copernicus sentinels 4 and 5 mission requirements ... · the copernicus atmosphere service (cas)...

Prepared by Mission Science Division Reference EOP-SM/2413/BV-bv Issue 2 Revision 0 Date of Issue 7 July 2017 Status issued Document Type Mission Requirements Traceability Document (MRTD) Distribution public

ESA UNCLASSIFIED – For Official Use

estec

European Space Research and Technology Centre

Keplerlaan 1 2201 AZ Noordwijk The Netherlands�

Tel. (31) 71 5656565 Fax (31) 71 5656040

www.esa.int�

COPERNICUS SENTINELS 4 AND 5 MISSION REQUIREMENTS TRACEABILITY DOCUMENT

/166

S-4/-5 MRTD

Date 7 July 2017 Issue 2 Rev 0


DISTRIBUTION

ESTEC/EOP ESRIN/EOP D. Aminou J. Aschbacher

G. Bagnasco M. Borgeaud

G. Bazalgette Courrèges- Y.-L. Desnos

J.-L. Bézy A. Dehn

P. Blythe G. Filippazzo

M. Davidson P. Goryl

M. Drinkwater N. Hanowski

M. Erdmann S. Jutz

T. Fehr H. Laur

P. Goudy E. Monjoux

R. Koopman A. Tassa

G. Levrini C. Zehner

D. Martin

G. Mason

K. McMullan EUMETSAT R. Meynart B. Bojkov

H. Nett M. Dobber

D. Schuettemeyer G. Fowler

H. Stark C. Kaiser

P. Silvestrin A. Lacan

B. Simpson R. Munro

H. Stark P. Schluessel

N. Wright R. Stuhlmann

ESTEC /TEC OTHER

B. Ahlers Sentinel-4/-5 Mission Advisory Group

H. Weber

Sentinel-5P Mission Advisory Group

/166

S-4/-5 MRTD



Reason for change Issue Revision Date

The first issue of the Mission Requirements Traceability Document is based on the document EOP-SM/1507/JL-dr, which was first issued on 02.04.2007, and which has evolved taking into account recommendations from GAS IG, internal reviews, Camelot and other scientific support studies, and from the from Sentinel-4/-5 MAG;

1 0 20.09.2012

Traceable to System Requirements Documents for Sentinel-4 (S4-ESA-UVN-RS-0581, issue 2, rev. 2) and Sentinel-5 (S5-RS-ESA-SY-0002, issue 3, rev. 0), see change record for details;

2 0 31 May 2017

Issue Revision

Reason for change Date Pages Paragraph(s)

Mission scenario formulations for Sentinel-4 and -5 harmonized;

31 May 2017 59 Section 5.2.1

The S5 related requirements listed below have been changed (for details see Annex 5): MR-LEO-SYS-35, 37, -39, -40, -60, -61, -62, -63, -64, -65, -130, MR-LEO-UVN-80, -87, -88, -90, -100, -120, -130, -150, -160, -180, -185, -190, -220, -257, -258, -260, MR-LEO-SWIR-50, -57, -110, 120, -130, -140, -150, -160, -170, -180, -185, -198, -287, -290, -300

31 May 2017 Section 6

The S4 related requirements listed below have been changed (for details see Annex 5): MR-GEO-SYS-5, -20, -30, -80, MR-GEO-UVN-30, -37, -110, -120, -130, -190

31 May 2017 Section 7

Editorials (reference to Copernicus instead for GMES, launch dates updated, etc.)

31 May 2017 Entire document

/166

S-4/-5 MRTD



Table of contents:

EXECUTIVE SUMMARY ........................................................................................................................... 7 1 INTRODUCTION ................................................................................................................................ 9 2 BACKGROUND AND JUSTIFICATION .............................................................................................. 11 2.1 Background ....................................................................................................................................................................... 11 2.2 Justification ...................................................................................................................................................................... 12 3 USER INFORMATION REQUIREMENTS INTRODUCTION ............................................................. 15 3.1 Atmospheric Services ....................................................................................................................................................... 15 3.2 Copernicus Service (MACC) Requirements .................................................................................................................... 17 3.2.1 GreenHouse Gases (GHG) ............................................................................................................................................. 17 3.2.2 Global Reactive Gases (GRG) ........................................................................................................................................ 17 3.2.3 Aerosols .......................................................................................................................................................................... 17 3.2.4 Regional Air Quality (RAQ) ........................................................................................................................................... 17 3.2.5 Space Observation Requirements ................................................................................................................................. 17 3.3 Downstream Service Requirements ................................................................................................................................ 18 3.4 Applications vs. Environmental Theme and User Information ..................................................................................... 19 4 LEVEL-2/3 DATA REQUIREMENTS ................................................................................................. 22 4.1 Derivation and General Characteristics of Requirements .............................................................................................. 22 4.1.1 Data Requirements Table Format and Definition of Height Ranges .......................................................................... 22 4.1.2 Coverage and Sampling Requirements ......................................................................................................................... 24 4.1.3 Uncertainty, Spatial Resolution and Revisit Time Requirements ............................................................................... 25 4.2 Geophysical Observation Requirements per Application ............................................................................................. 28 4.2.1 Theme A: Stratospheric Ozone and Surface UV .......................................................................................................... 28 4.2.2 Theme B: Air Quality ..................................................................................................................................................... 37 4.2.3 Theme C: Climate .......................................................................................................................................................... 45 4.3 Application Priorities and Contributions of Existing and Planned Missions ................................................................ 54 5 SYSTEM CONCEPTS ........................................................................................................................ 56 5.1 Candidate Remote Sensing Techniques and Mission Scenarios .................................................................................... 56 5.1.1 High temporal and spatial resolution space-based measurements of tropospheric composition, including the planetary boundary layer (PBL), for air quality applications .................................................................................................. 56 5.1.2 High spatial resolution and high precision monitoring of tropospheric climate gases (CH4, CO (precursor) and CO2) and aerosols with sensitivity to PBL concentrations for climate protocol monitoring ................................................. 57 5.1.3 High vertical resolution measurements in the upper troposphere/lower stratosphere region for stratospheric ozone/surface UV and climate near-real time and assessment applications .......................................................................... 57 5.2 Overall System.................................................................................................................................................................. 58 5.2.1 Mission Scenario ............................................................................................................................................................ 59 5.3 The Need for a Sentinel-5 Precursor (S5P) Mission ...................................................................................................... 60 5.4 Requirement Terms, Conventions and Definitions ........................................................................................................ 61 6 LEVEL-1B PRODUCT REQUIREMENTS FOR LEO MISSION (SENTINEL-5 AND SENTINEL-5 PRECURSOR) ......................................................................................................................................... 62 6.1 LEO-SYSTEM ................................................................................................................................................................... 62 6.2 UV-VIS-NIR (LEO-UVN) ................................................................................................................................................66 6.2.1 LEO-UVN, Geometrical Requirements ........................................................................................................................66 6.2.2 LEO-UVN, Spectral Requirements ...............................................................................................................................66 6.2.3 LEO-UVN, Radiometric Requirements ........................................................................................................................ 72 6.3 SWIR (LEO-SWIR) .......................................................................................................................................................... 78 6.3.1 LEO-SWIR, Geometrical Requirements ....................................................................................................................... 79 6.3.2 LEO-SWIR, Spectral Requirements .............................................................................................................................. 79 6.3.3 LEO-SWIR, Radiometric Requirements ....................................................................................................................... 81 6.4 Thermal Infrared (LEO-TIR) .......................................................................................................................................... 89 6.4.1 LEO-TIR, Geometrical Requirements ......................................................................................................................... 89

/166

S-4/-5 MRTD



6.4.2 LEO-TIR, Spectral Requirements ................................................................................................................................ 90 6.4.3 LEO-TIR, Radiometric Requirements .......................................................................................................................... 92 6.5 Cloud Imagery Data ......................................................................................................................................................... 95 6.6 Aerosol Data ..................................................................................................................................................................... 95 6.6.1 LEO-3MI, Geometric Requirements ............................................................................................................................. 95 6.6.2 LEO-3MI, Spectral Requirements ................................................................................................................................96 6.6.3 LEO-3MI, Radiometric Requirements .........................................................................................................................96 6.6.4 LEO-3MI, Auxiliary Data Requirements ...................................................................................................................... 97 6.7 Sentinel-5 Precursor (S5P) ............................................................................................................................................. 98 7 LEVEL-1B PRODUCT REQUIREMENTS FOR GEO MISSION (SENTINEL-4) ................................... 99 7.1 GEO-SYSTEM ..................................................................................................................................................................99 7.2 UV-VIS-NIR Requirements (GEO-UVN) ...................................................................................................................... 101 7.2.1 GEO-UVN, Geometrical and Temporal Requirements .............................................................................................. 101 7.2.2 GEO-UVN, Spectral Requirements ............................................................................................................................. 102 7.2.3 GEO-UVN, Radiometric Requirements ...................................................................................................................... 104 7.3 Thermal Infrared (GEO-TIR) ........................................................................................................................................ 110 8 DATA DELIVERY REQUIREMENTS ............................................................................................... 111 9 SYNERGIES AND INTERNATIONAL CONTEXT .............................................................................. 112 10 REFERENCE DOCUMENTS ............................................................................................................ 113 LIST OF ACRONYMS AND ABBREVIATIONS ..................................................................................................................... 116 APPENDIX 1: REQUIREMENT TERMS, CONVENTIONS AND DEFINITIONS ........................................ 119 APPENDIX 2: COPERNICUS ATMOSPHERIC SERVICE REQUIREMENTS TRACEABILITY ................. 127 APPENDIX 3: DATA REQUIREMENTS TABLES ................................................................................... 130 APPENDIX 4 GAIN VECTORS ............................................................................................................... 150 APPENDIX 5 DETAILED CHANGE RECORD ......................................................................................... 161

/166

S-4/-5 MRTD

Date 19.05.2014 Issue 1 Rev 1


EXECUTIVE SUMMARY

This Mission Requirements Traceability Document (MRTD) elaborates and further clarifies in detail the Copernicus (former GMES) Sentinel-4, -5, and -5Precursor (S5P) Mission Requirements specified in the Sentinel-4/-5 Mission Requirements Document (MRD) (/ESA2007/) through the definition of unambiguous and traceable numbered requirements used by the Sentinels-4, -5 and -5P implementation teams. The Sentinel-4/-5 MRTD and MRD are managed by the Sentinel-4, -5 and 5P mission scientists according to the procedure set out in /ESA2012/.

The heart of the document is formed by the Level-1b requirements for these missions (Chapters 6 and 7). The document was first issued in March 2007, and is considered a living document which will be updated according to new insights. The current version was released after including results from the CAMELOT study (\CAM2009\), from a scientific study for Sentinel-4 led by IUP-Bremen, from ad hoc discussions with scientists, and the ON-TRAQ study (/ON2009/).

Chapter 1 of this MRTD contains an introduction.

Chapter 2 contains the background and justification for the Copernicus Sentinel-4, -5, and -5P Missions. It describes the attribution of a number of atmospheric change processes to human factors, describes the protocols in place and assesses the influence of atmospheric observations on policy making.

In Chapter 3, the user information requirements are presented. This includes a description of the Copernicus framework and requirements stemming from it. It highlights the three environmental themes and the three types of user information domains.

Chapter 4 focuses on the Level-2/3 data requirements. In this Chapter, a detailed explanation is provided of which data are needed for the various applications, identified by Copernicus and outlined in Chapter 3.

The topic of Chapter 5 is system concepts, meaning an overview of different remote sensing techniques, wavelength regions to be applied, and mission scenarios. From the presented mission scenarios, the choice made is to fly one GEO and one LEO mission. However, flying this combination of missions only would lead to a gap in the data series, and hence the need for a precursor mission has been identified.

At the end of Chapter 5, a section on mission requirements, a number of definitions and default interpretations have been included, leading up to the two chapters that contain the Level-1B mission requirements, namely Chapter 6, focusing on the LEO missions Sentinel-5 and Sentinel-5P, and Chapter 7, focusing on the GEO mission, Sentinel-4. Chapters 6 and 7 are divided into sections as follows: first a system section, followed by sections ordered by wavelength range considered. For the GEO system, additional requirements on the Eumetsat-delivered observations are presented.

Chapter 8 contains the data delivery requirements. Chapter 9 is a brief description of synergies between satellite systems, and provides the international context. Chapter 10 is a list of references, followed by various lists (Tables, Figures).

/166

S-4/-5 MRTD



In Appendix A, definitions used throughout the document are given, which are important for a proper interpretation of the presented requirements. In Appendix B, the tables stemming from Copernicus with the Level-2 requirements are given. Those were the basis for derivation of the Level-1 requirements.

/166

S-4/-5 MRTD



1 INTRODUCTION

Atmospheric chemistry observations from space have been made for over 30 years. They have always been motivated by the concern about a number of environmental issues which are outlined below. Most of the space instruments have been designed for scientific research, improving the understanding of processes that govern stratospheric ozone depletion, climate change and the transport of pollutants. Long-term continuous time series of atmospheric trace gas data have been limited to stratospheric ozone and a few related species. According to current planning, meteorological satellites will maintain these observations over the next decade. They will also add some measurements of tropospheric climate gases, however their measurement being motivated by meteorology and vertical sensitivities / accuracies being marginal for atmospheric chemistry applications. With the exception of stratospheric ozone, reliable long-term space-based monitoring of atmospheric constituents with quality attributes sufficient to serve atmospheric chemistry applications still need to be established.

The need for a GMES (now Copernicus) atmospheric service (GAS), its scope and high level requirements were laid down in an orientation paper organised by the European Commission /GAS2006/, /GAS2006a/ and then updated by an Implementation Group (IG), backed by four working groups, advising the Commission on scope, architecture, in situ and space requirements. The resulting report of the IG /GAS2009/ includes a section resulting from the working group on space observation infrastructure.

The Copernicus Atmosphere Service (CAS) will provide coherent information on atmospheric variables in support of European policies and for the benefit of European citizens. Services cover: air quality, climate change/forcing, and stratospheric ozone and solar radiation.

The main functions of the Copernicus Atmosphere Service are the acquisition and processing of space and in-situ observations (Near-Real-Time, historic and ancillary), analysis and forecasting, product generation, dissemination and archiving. In particular, the Copernicus Atmosphere Service will provide:

Standard Global and European data on which downstream services will be based;

Information for process assessments;

Daily analysis of the atmosphere at various space/time scales;

Key information on long range transport of atmospheric pollutants;

European overviews and initial and boundary conditions for air quality models;

Sustained monitoring of green-house gases, aerosols and reactive gases such as tropospheric ozone.

The European Union (EU) Framework Programme (FP) 6 integrated project for Global and regional Earth-system (Atmosphere) Monitoring using Satellite and in-situ data (GEMS), see http://gems.ecmwf.int, is a large activity with the aim to generate a prototype service for assimilated chemical data, provided a basis of a future operational Copernicus service.

/166

S-4/-5 MRTD



Its data products and sources are listed in /Hol2008/. Development, demonstration and evaluation of 69 dedicated services based on the usage of existing space data in the ESA Copernicus Service Element for Atmosphere project PROMOTE (see www.gse-promote.org) provides a further basis, in particular with respect to end-user requirements /PRO2009/. In June 2009, GEMS and PROMOTE were followed by the EU FP7-funded Copernicus service MACC (Monitoring Atmospheric Composition and Climate, see http://www.gmes-atmosphere.eu).

The general framework for spaceborne atmospheric composition measurements in synergy with ground-based and airborne measurements and integration with atmospheric models and data assimilation schemes has been outlined in IGOS-IGACO Theme Report /Bar2004/. That document includes also quantitative observation requirements, summarised for scientific and operational applications. Several other efforts have been made to identify the needs of long-term atmospheric composition data, such as the GMES-GATO report /GAT2004/, the Eumetsat position paper on observation requirements for nowcasting and very short range forecasting in 2015-2025 /EUM2003a/, and a Eumetsat-commissioned study to identify requirements for geostationary platforms in the context of Meteosat Third Generation /EUM2003b/.

An ESA study on “Operational Atmospheric Chemistry Monitoring Missions” (“CAPACITY”) /Kel2005/ gathered all available inputs and generated comprehensive observational requirements by environmental theme, by user group, and by observational system (ground / satellite). The study also assessed the contributions of existing missions to the fulfilment of these requirements, and identified priorities of observational techniques for future Copernicus Sentinel-4 and -5 missions. Tentative requirements at radiance level and other instrument and system related requirements were also identified. These requirements rely partly on a rich experience with the usage of existing similar instrumentation, and partly on retrieval simulations.

Requirements at radiance level have been further investigated in the frame of ad-hoc expert meetings. Results of Eumetsat requirement processes in the frame of MTG /EUM2007/ and EPS-SG (Eumetsat Polar System – Second Generation) /EUM2006/, /EUM2013/ missions have been taken into account.

An ESA study, CAMELOT, has been performed in order to consolidate mission requirements for Sentinels-4, -5, and -5P /Lev2009/.

The MRTD for Sentinels-4 and -5 covers the end-to-end system including high-level requirements, mission operations, data product development and processing, data distribution and data archiving. The Sentinel-4/-5 MRTD and MRD are managed by the Sentinel-4/-5 and -5p mission scientists according to the procedure set out in /ESA2012/.

/166

S-4/-5 MRTD



2 BACKGROUND AND JUSTIFICATION

2.1 Background

The human activities on a planetary scale are appreciably altering the processes that control the Earth’s system. The system comprises a complex set of biological, physical and chemical processes taking place in and between the oceans, continents and the atmosphere. Our atmosphere has sufficient oxygen to maintain life, and quantities of trace gases, which serve to maintain the UV-protecting stratospheric ozone layer and a greenhouse effect that maintains habitable temperature conditions at the surface. The system of trace gases is maintained by a complex web of chemical processes that also serve to maintain the atmosphere in a sufficiently ‘clean’ condition for life to continue.

Historically, human activities have always interacted with the atmosphere, but the growth of population and industrialisation in the 19th and 20th century has led to dramatic changes in the Earth system. The developments of large mega-cities and the changes in land usage, occurring in the last 50 years, are two outstanding examples. Air pollution, the over-burdening of the atmosphere with emitted substances, has been known from the earliest times but it was always a local nuisance. In the 20th century pollution spread to regional and global scales and became a subject of concern both nationally and internationally. Table 2.1 lists some of the noticeable environmental effects attributable to the atmosphere (adapted from /Bar2004/).

Table 2.1 Some atmospheric problems attributed to human activities.

Climate Change and the Greenhouse Effect

The likely change in climate due to a continuing increase in the average global temperature, resulting from the increase in carbon dioxide (e.g. from industrial and vehicular combustion) and other trace gases such as methane, ozone and the HCFCs. Aerosols also play a part, by perturbing the Earth’s radiative balance, both directly through the scattering and absorption of sunlight, and indirectly through the influence on clouds and the hydrological cycle. Changes in both tropospheric and stratospheric temperatures are likely to have negative (or amplifying) effects on many issues.

London Smog

Visible soot particles and sulfur dioxide, trapped in a static high pressure weather system, resulting from domestic heating and industrial activities. Now largely eliminated in the developed countries but still a problem in the developing world.

Los Angeles Smog or "summer" smog

Products of photo-oxidation, notably ozone and PAN, and other lachrymatory compounds, and particles, resulting from the photo-oxidation of industrial and vehicular emissions in the presence of sunlight. Endemic under static weather conditions throughout the world, despite countermeasures taken in the developed and developing world.

Acid Rain

Acid precipitation, which threatened natural ecosystems and forests over Europe and North America, resulting from industrial emissions of sulfur dioxide, which is photo-oxidised to form sulfate. Although sulfur emissions have been reduced, acidity is still a problem with the growing contribution of nitrogen compounds from industry and vehicles to acidity, and the SO2 and NOx from ships.

/166

S-4/-5 MRTD



Ozone Depletion in the Stratosphere

Loss of ozone from the stratosphere with a consequent increase in UV radiation reaching the surface, resulting from the release of CFCs from industrial processes, which are inert in the troposphere but photolyse in the stratosphere. The ozone hole is a dramatic seasonal demonstration of such loss. Although the production of CFCs has been curbed, ozone depletion will remain for a long time because of the slow rate at which these compounds, still mostly present in the troposphere, diffuse to the stratosphere.

Nutrification of coastal waters and freshwater lakes; eutrophication

Change in coastal and freshwater ecosystems with the increased deposition of nitrogen compounds from the air. Resulting from industrial and vehicular emissions. Eutrophication and "tides" of poisonous algae are more obvious manifestations.

The increase in background tropospheric ozone.

Observed in the northern hemisphere, the increase is attributed to the intercontinental transport of photo-oxidants. Current background levels are comparable to those known to affect plant growth.

Enhanced aerosol and photo-oxidant levels due to biomass burning

Observed annually over the Atlantic and Indian Oceans, as well as South-East Asia as a result of biomass burning and agricultural activities in Africa, South America and Asia.

Aerosols in and downwind of regions of high population

Recently observed "brown cloud" over the western Indian Ocean, far from land; originates from industrial, agricultural and vehicular activities in the Indian subcontinent and South-East Asia and possibly China.

The intercontinental transport of pollutants and aerosols

These have been observed across the North Atlantic, Pacific and southern Indian Ocean

Stability of the Atmospheric Oxidation Efficiency

The efficiency of the atmosphere for removing greenhouse gases and pollutants has been called into question.

2.2 Justification

Issues related to changing atmospheric composition can be grouped into the following themes:

Stratospheric Ozone and its impact on surface UV radiation,

Air Pollution,

Climate Change.

These affect human, animal and ecosystem health and have considerable economic consequences. Such effects have been recognised for some time and the public concern about the atmosphere is reflected in a number of international agreements.

The Convention on Long-Range Transboundary Air Pollution (CLRTAP) (1979) /CLRTAP/ with protocols on Sulfur (1985 & 1994), Nitrogen Oxides (1988), Volatile Organic Compounds (VOC) (1991), Heavy Metals (1998), Persistent Organic Compounds (POP) (1998), and Acidification, Nutrification and Ground Level Ozone (1999).

/166

S-4/-5 MRTD



The Vienna Convention on the Protection of the Ozone Layer (1985) /Vie2006/ and the Montreal Protocol on Substances That Deplete the Ozone Layer (1987), and the subsequent adjustments and amendments /Mon1987/.

The Framework Convention on Climate Change (1992) /UN1992/ and the Kyoto Protocol (1997) /UN1998/.

The global nature of the problem requires a worldwide coordinated approach. The aim of these Conventions and Protocols is to stem or reverse adverse environmental change. To be effective, these Protocols require timely, reliable and long-term information for assessment, monitoring and verification purposes. In addition to the need to ascertain the effectiveness of Protocols, there is a need to predict future change. Daily forecast systems are presently emerging in various stages of development. A number of local and national authorities are already providing air quality and UV forecasts to serve public awareness and provide advance warning systems similar to the weather forecast service. On a different level, an intense research effort is directed at climate predictions and understanding the consequences of global change. The quality of predictions very much depends on valid theoretical models and accurate measurements of the state and evolution of the atmosphere.

Observational data and theoretical models together result in increased understanding of atmospheric change. This synthesis is needed for policy assessment and, in general, to advance our knowledge.

The need for information on atmospheric composition is driven by the potentially huge impact that global atmospheric change has on human health and safety, ecosystem balance and socio-economic development.

Direct needs for atmospheric composition information derive from monitoring and verification requirements of Protocols designed to regulate and mitigate the effects of human induced atmospheric change. This information is often needed on a country (signatory) by country basis. There is a need for independent global information for Protocol verification, separate from reporting obligations by individual signatories. This need calls for the ability to probe the atmospheric planetary boundary layer (PBL) on a global scale at high spatial and temporal resolution.

The process of policy formulation that leads to the implementation of Protocols is a multi-stage process, which starts with the scientific discovery of change, the assessment and understanding of the issues involved, checked by the usual process of scientific scrutiny and independent verification. Good quality observations and reliable theoretical models are essential at this explorative stage.

This stage is followed by the political process of policy formulation and appraisal of policies. Autonomy and self-reliance of the European Union and Member States requires the ability to carry out independent investigations and assessment of environmental and climate issues. There is a strategic need for reliable environmental and climate information to be available at the negotiating table when politicians and policy-makers need to decide on new policies. Access to high quality environment and climate data at all levels is required in order to be effective in policy negotiations.

/166

S-4/-5 MRTD



Forecasts are necessary in order to anticipate episodes of risk to health and safety and to provide advance warnings to the public and the responsible authorities. Predictions of long-term environmental change are necessary in order to abate and mitigate the socioeconomic consequences and to formulate policy and research agendas for sustainable development. Here, information based on a combination of measurements and models turns out to be necessary, and in the case of forecast, the delivery of this information needs to take place in near-real time.

The need for stratospheric ozone information derives from the harmful effects of excess UV B dose on health and biosphere. The Montreal Protocol calls for quadrennial ozone assessments and monitoring of stratospheric ozone concentrations and emissions of ozone destructing substances. Forecast of stratospheric ozone and surface UV prediction are possible and necessary in order to issue warnings and raise public awareness.

Understanding of the ozone layer behaviour includes that of chemistry–climate interactions, which is a subject of scientific research. Continued assessment and improvement of regulatory action is needed until the recovery of the ozone layer is a fact, currently not expected to happen before 2050.

The need for air pollution monitoring and forecast is driven by health and safety directives and conventions. The Convention on Long-Range Trans-boundary Air Pollution (CLRTAP) and several EC directives regulate the emission of air pollutants. The Clean Air For Europe (CAFE) initiative has led to a thematic strategy setting out the objectives and measures for the next phase of European air quality policy. Air Quality (AQ) forecasts are important to serve as health warning in polluted areas. Environmental agencies need the AQ information in order to support implementation of regulatory actions on emissions from sources such as vehicular traffic and electric power generation. Reliability, timeliness, continuity and quality of this information is important. The temporal and spatial scale of requirements poses a challenge to both observational and modelling capability, ranging from street level to continental transport and from diurnal variability to decades of chemical lifetimes.

The need for climate information and prediction stems from the impact of climate change on society, which can be enormous. Policy on greenhouse gas regulation will deeply affect the energy resource management, the transport sector and the economy as a whole. The UN Framework Convention on Climate Change (UNFCCC) adopted at the Earth Summit of Rio de Janeiro in 1992 and the resulting Kyoto Protocol (1997) commits the signatories to cut emissions of greenhouse gases by 8% in the period 2008-2012 compared with 1990 levels. The EU and some hundred other nations have ratified the Protocol, but major players like the USA have not, whereas developing nations like China and India are not committed. Climate predictions are limited by a range of uncertainties depending on economic development scenarios assumed and on the validity of models employed to describe the Earth System. Understanding climate change includes understanding the chemistry-climate interactions at all levels in the atmosphere, indeed in the entire system Earth. This subject is one of the great challenges for future global observation and modelling development.

/166

S-4/-5 MRTD



3 USER INFORMATION REQUIREMENTS INTRODUCTION

3.1 Atmospheric Services

In the frame of Copernicus, the EC distinguishes two types of services,

Copernicus services

o these are global and pan-European, multi-purpose information service capacities,

o linked to EU information needs (EU policies and international commitments) or to decisions to share capacities at EU level

o require sustained public funding (EU & Member States);

Downstream services

o tailored for specific applications at local, regional, national, European levels (public good or private use)

o use Copernicus services as one of the inputs

o EU not directly driving the service and not responsible for service requirements.

GAS Services component should include the following (as described in /GAS2009/)

Air quality

integrated global and European air quality analysis;

integrated global and European air quality forecast;

historic records of Global and European atmospheric composition.

Climate forcing

improved and sustained monitoring of the state of the climate system (surface and upper air meteorology and composition) and its variability;

integrated global, European and regional concentration fields of key greenhouse gases (CO2, CH4 and related tracers) enabling determination of sources and sinks.

Stratospheric ozone and solar radiation

improved and sustained monitoring of the current status and trends in stratospheric ozone depletion and ozone depleting gases;

routine provision of updated ozone, UV and solar radiation maps and forecasts;

historic European UV and solar radiation records and mapping.

Generic services

validation, bench-marking, and documentation service;

data dissemination and archiving infrastructure assurance service;

/166

S-4/-5 MRTD



near-real time satellite data provision service.

An indicative list of possible downstream services includes the following (as described in /GAS2009/)

Air quality

Local air quality forecasts (urban scale)

Improved air-quality-related alerts and forecasts by health services supporting vulnerable communities (COPD, respiratory diseases, asthma (including pollen-induced allergies))

Supporting integrated air quality indices.

Enhanced assessment of air quality within a specific region, supporting development of effective air pollution abatement measures through proper apportionment of sources and assessment of impacts (human exposure) etc.

Improved air-quality-related alerts and forecasts for extreme events involving the combined effects of heat stress, high UV-B exposure and poor air quality;

Analysis of national, regional and local air pollution abatement policies and measures through inverse modelling, validation/improvement of emission inventories and reconciling bottom-up and top-down emission inventories;

Support to implementation of indicators related to following different aspects of policy effectiveness, as for example public exposure assessment, transboundary contributions (at a particular site/regions, rather than at large scale), footprint of cities, contribution from a transport mode etc.

Climate forcing

Identification, assessment and monitoring of regional/local sources and sinks of greenhouse gases and pollutants and related tracers in support of emission and sink verification and mitigation policy.


Solar-radiation potential analysis, policy scenario analysis, energy yield mapping, support to electricity transmission network management together with site audits and plant management for solar power plants.

A GMES (now Copernicus) Service project was launched in 2009 by the EC in the frame of the first FP7 call for proposals for space-based applications at the service of the European society /EU2007/. The project, Monitoring Atmospheric Composition and Climate (MACC), see http://www.gmes-atmosphere.eu, is based on the developments made in the FP6 integrated project GEMS (http://gems.ecmwf.int), with some components of the GMES Service Element (GSE) PROMOTE /PRO2009/.

An invitation to tender for Downstream Services was released in 2008 in the Second FP7 Space Call. A proposal for Downstream Services on air quality, drawing extensively from developments in the GSE project PROMOTE, was written. The project, called PASODOBLE, has started in early 2010.

/166

S-4/-5 MRTD



In the following sections the currently available services are briefly described. The traceability between these services and the data products coming from the Sentinel-4 and -5 satellites is provided in Appendix 2.

3.2 Copernicus Service (MACC) Requirements

This will be based on the MACC project . The key functions of the service will be

acquisition and processing of observations (both space and in-situ);

analysis and forecasting;

product generation, dissemination and archiving.

The above mentioned core service components will be implemented in four sub-projects as given below.

3.2.1 GreenHouse Gases (GHG)

The mission of this sub-project is to develop the first operational system to monitor the concentrations of greenhouse gases, and their associated surface sources and sinks, and to attribute these sources and sinks to controlling processes.

3.2.2 Global Reactive Gases (GRG)

The overall purpose of this sub-project is to set-up an operational data assimilation system for chemically reactive gases on a global scale providing products to end-users on a day-by-day basis. GEMS has developed a stratosphere-troposphere chemical analysis system with an initial focus on the troposphere. In addition to ozone, four other reactive gases (NOx, CO, SO2 and HCHO) are incorporated in ECMWF's Integrated Forecasting System and the 4D-Var assimilation system.

3.2.3 Aerosols

The mission of aerosol sub-project is to improve weather analysis and reanalysis products (and possibly ultimately weather forecasts), and to provide operational aerosol products to end-users.

3.2.4 Regional Air Quality (RAQ)

The Regional Air Quality sub-project has as its focus the production of regional forecasts of chemical species and air quality indices based on an ensemble of air-quality models on the European scale.

3.2.5 Space Observation Requirements

Requirements for space measurements of chemical species have been elaborated in /GAS2009/, in summary they include

for air quality: vertical profiles of particulate matter (PM), ozone (O3), nitrogen dioxide (NO2), carbon monoxide (CO) and sulphur dioxide (SO2), with hourly

/166

S-4/-5 MRTD



temporal sampling during daytime with specific focus over the European area and its neighbourhood;

for climate forcing: vertical distribution of tropospheric Essential Climate Variables as identified by WMO and GCOS, including water vapour, ozone, aerosols (optical and chemical properties), and greenhouse gases with focus on carbon dioxide (CO2) and methane (CH4), with global coverage;

for ozone, UV and solar energy: total content and profiles (especially in the upper troposphere and stratosphere) of ozone, plus information on water vapour, active nitrogen components, nitrogen reservoir and source species (NOx, HNO3 and N2O), active halogens and halogen reservoirs, aerosol optical properties and methane with global daily coverage and vertical resolutions ranging from 0.5 to a couple of km.

In a recent update of GCOS /GCO2010/ the precursors for ozone and aerosols, namely NO2, SO2, CO and HCHO, have also been included in the Essential Climate Variables (ECV), see table at http://gosic.org/ios/MATRICES/ECV/ecv-matrix.htm.

3.3 Downstream Service Requirements

Complementary to the main Copernicus services currently dedicated downstream services are developed within EU FP 7 projects. These services link space, in-situ and modelling capacities down to specific applications according to user needs. For example, PASODOBLE aims at developing local and regional air quality services to improve information for the public and people at risk, to support the health community and to deliver policy relevant data and advise to local authorities and regional agencies (www.myair-eu.org ). Exploiting the Copernicus Services the project ENDORSE seeks to develop regional services promoting the energy use from sun, wind, and biomass, electricity grid management and building engineering through daylighting in buildings. The European Volcano Observatory Space Services project (EVOSS) utilises satellites for monitoring of major volcanic Hazards. CARBONES focuses on a 30-year re-analysis of global and European carbon fluxes and pools (www.carbones.eu) while EURO4M sets out to monitor climate extremes in Europe using seamlessly integrated datasets of in situ, satellites and reanalyses sources (www.euro4m.eu).

An indicative list of possible downstream services includes:

Air quality

Local air quality forecasts (urban scale).

Improved air-quality-related alerts and forecasts by health services supporting vulnerable communities (with Chronic Obstructive Pulmonary Disease [COPD], respiratory diseases, asthma (including pollen-induced allergies)).

Supporting integrated air quality indices.

Enhanced assessment of air quality within a specific region, supporting development of effective air pollution abatement measures through proper apportionment of sources and assessment of impacts (human exposure) etc.

/166

S-4/-5 MRTD



Improved air-quality-related alerts and forecasts for extreme events involving the combined effects of heat stress, high UV-B exposure and poor air quality.

Analysis of national, regional and local air pollution abatement policies and measures through inverse modelling, validation/improvement of emission inventories and reconciling bottom-up and top-down emission inventories.

Support to implementation of indicators related to following different aspects of policy effectiveness, as for example public exposure assessment, transboundary contributions (at a particular site/regions, rather than at large scale), footprint of cities, contribution from a transport mode etc.

Climate forcing

Identification, assessment and monitoring of regional/local sources and sinks of greenhouse gases and pollutants and related tracers in support of emission and sink verification and mitigation policy.


Solar-radiation potential analysis, policy scenario analysis, energy yield mapping, support to electricity transmission network management together with site audits and plant management for solar power plants.

3.4 Applications vs. Environmental Theme and User Information

Consistent with the GAS needs /GAS2006/ and as elaborated in /Kel2005/, user information requirements need to address three major environmental themes:

(A) Stratospheric Ozone and Surface UV radiation

(B) Air Quality

(C) Climate

(In theme (A), “surface UV radiation” has been generalised to “solar radiation”, ie. support to exploitation of solar energy /GAS2006/, however no space observation requirements beyond those already satisfied by the European meteorological infrastructure have been identified /GAS2009/.)

Further, three main drivers were identified for operational spaceborne observations of atmospheric composition. These drivers are

(1) The provision of information on treaty verification and protocol monitoring

(2) The facilitation and improvement of operational applications and services, including forecasts, using near-real time monitoring information on the atmospheric composition

(3) The contribution to scientific understanding and knowledge acquisition for environmental assessments to support policy.

/166

S-4/-5 MRTD



Each of the three overall drivers contributes to policy support. The first item with the direct delivery of the required monitoring information, the second one with applications and services using actual information and forecasts on the atmospheric state for warning systems and to support real-time decision making, and the third one via environmental assessments and their summaries for policy makers (WMO ozone assessments, European and global-scale environmental assessments on Air Quality and IPCC climate assessments).

Furthermore, in addition to the three overall drivers, spaceborne operational monitoring of atmospheric composition will be valuable:

To promote scientific research with unique long-term consistent data products

To contribute to numerical weather prediction (NWP), climate monitoring, and, in a broader perspective, Earth system monitoring

To improve atmospheric correction for surface remote sensing

To strengthen public awareness on environmental themes

Different levels of information will be needed, which can be associated with different user categories. On a first level of information there are the users that are involved in the monitoring of protocols and directives (Compliance User), e.g. governmental institutes on different administrative levels and international organisations associated with international treaties and protocols. The data requirements of these users are typically Level-4 data requirements, such as long-term 3-dimensional global distributions of trace gases, aimed at complete monitoring of the atmospheric state and its evolution in time.

On a second level of information there are users that would like to apply the available data products to operational applications and services, e.g. meteorological institutes, to improve early-warning systems and to increase public awareness. These users typically need the data in near-real time, i.e., within a few hours after observation. NWP centres may wish to receive Level-1 data (typically radiances and irradiance) in order to process them to Level-2 in near-real time and within the running applications.

The services may involve different user categories with specific data requirements, e.g., they may be directed to support policy-makers for control strategies and security, health and environmental law enforcement, e.g. on measures to be taken in air pollution episodes. The services can also be directed to the general public for health warnings (concentrations exceeding standards, UV radiation levels) and planning of outdoor activities (e.g. a Marathon in Athens) as well as for general awareness. Scientists could use actual information on the atmospheric composition for campaign planning and climate monitoring. Other specific organisations could use the data, e.g. to improve safety of air and road transport by provision of warnings on environmental hazards (forecast of plumes related to volcanic eruptions, extreme forest fires, etc.).

On a third level are scientists assessing the technical basis for abatement strategies, typically summarised in environmental assessment reports (Technical User) and the scientists using the information for (fundamental) scientific research (Research User). Key to these users is the understanding of the atmospheric state and its evolution. The data requirements are typically more stringent in comparison to the monitoring requirements and these users will require Level-1 and/or Level-2 data products in addition to Level-4.

/166

S-4/-5 MRTD



The most important aspect of operational missions for these users is the perspective of unique long-term and homogeneous data sets with global coverage.

An overall summary is provided in Table 3.1.

Table 3.1 Applications per environmental theme and user information (UTLS = Upper Troposphere-Lower Stratosphere).

Environmental Theme

Information

Ozone Layer &

Surface UV radiation

A

Air Quality

B

Climate

C

Protocols

1

UNEP Vienna Convention; Montreal and subs. Protocols

CFC emission verification

Stratospheric ozone, halogen and surface UV distribution and trend monitoring

UN/ECE CLRTAP; EMEP / Göteborg Protocol; EC directives EAP / CAFE

AQ emission verification

AQ distribution and trend monitoring

UNFCCC Rio Convention; Kyoto Protocol; Climate policy EU

GHG and aerosol emission verification

GHG/aerosol distribution and trend monitoring

Services 2

Stratospheric composition and surface UV forecast

NWP assimilation and (re-) analysis

Local Air Quality (PBL); Health warnings (PBL)

Chemical Weather (PBL/FT)

Aviation routing (UT)

NWP assimilation and (re-) analysis

Climate monitoring

Climate model validation

Assessments

3

Long-term global data records

WMO Ozone assessments

Stratospheric chemistry and transport processes;

UV radiative transport processes

Halogen source attribution

UV health & biological effects

Long-term global, regional, and local data records

UNEP, EEA assessments

Regional & local PBL AQ processes; Tropospheric chemistry and long-range transport processes

AQ source attribution

AQ Health and safety effects

Long-term global data records

IPCC assessments

Earth System, climate, rad. forcing processes; UTLS transport-chemistry processes

Forcing agents source attribution

Socio-economic climate effects

/166

S-4/-5 MRTD



4 LEVEL-2/3 DATA REQUIREMENTS

4.1 Derivation and General Characteristics of Requirements

This section will provide the geophysical observation requirements (Level-2/3 data requirements) separately for each environmental theme and user category. The requirements have been generated for an integrated system, encompassing ground-based, airborne and satellite observations, and data integration tools such as data assimilation and inverse modelling tools including atmospheric models. In this document, only the space data requirements are reproduced.

4.1.1 Data Requirements Table Format and Definition of Height Ranges

The data requirements are tabulated per theme (A, B, C) and per user category (1,2,3) following the structure defined in the previous section, i.e., for monitoring / compliance users (A1, B1, C1), for forecast / near-real time applications and services (A2, B2, C2) and for environmental assessments / technical and research users (A3, B3, C3). The requirements are further split into Level-2 satellite data requirements and auxiliary requirements. Thus, for example, table A1 summarises the data requirements from spaceborne platforms for Theme A (ozone layer), user category 1 (monitoring, compliance user). Table 4.1 summarises the list of available data requirement tables, which are provided as Appendix 3 to this MRD. The auxiliary requirements are also described.

Table 4.1 List of the data requirements tables.

Table code Environmental Theme

Application User category

A1 Ozone Layer Monitoring Compliance

A2 Ozone Layer Forecast Near-real time

A3 Ozone Layer Assessment Technical/research

B1 Air Quality Monitoring Compliance

B2 Air Quality Forecast Near-real time

B3 Air Quality Assessment Technical/research

C1 Climate Monitoring Compliance

C2 Climate Forecast Near-real time

C3 Climate Assessment Technical/research

/166

S-4/-5 MRTD



Table 4.2 Format of the data requirements tables.

Reference code Environmental Theme

Requirement

Data

Product

Driver

Height Range

Horizontal resolution

Vertical resolution

Revisit Time

Uncertainty

The data requirements tables have the general format presented in Table 4.2. We distinguish per data product the relevant height range (for a profile) or a total column, or a partial column (e.g. tropospheric column). In general, the height-range requirements should be interpreted understanding that even when vertical profile information is required, information from column observations could still contribute to the application, although not fulfilling the vertical resolution requirement. Further, the required horizontal and vertical resolution and revisit time are given, for which the first value is a goal (or target) requirement, and the second value, separated by a slash (/), is the threshold requirement. In the last column the threshold uncertainties that can be allowed for the given (threshold) resolution requirements are presented.

For the height ranges, reference is made to the compartments of the atmosphere that are commonly distinguished in atmospheric research (see Figure 4.1). All boundaries should be interpreted as approximate values. In the troposphere distinction is made between the PBL, the Free Troposphere (FT), the Upper Troposphere (UT) and the Tropical Tropopause Layer (TTL). In the stratosphere a distinction is made between the lowermost stratosphere (LS), the middle stratosphere (MS), and the upper stratosphere (US). The mesosphere is denoted with (M).

/166

S-4/-5 MRTD



Figure 4.1 The atmospheric compartments that are distinguished for the height-range specifications in the data requirement tables. The boundaries have been set at fixed altitudes and latitudes for simplicity and only represent an approximation to the mean state neglecting atmospheric variability. Tropics [0° – 30°], Mid-latitudes [30° – 60°], Polar region [60° – 90°], in both hemispheres.

The PBL typically extends up to less than 2 km above the Earth’s surface. The PBL is usually thicker above continents than above oceans and typically up to 1 km altitude at polar regions. The FT is defined as the region between the top of the PBL and the tropopause. The tropopause in polar regions is typically at an altitude of ~8 km, at mid-latitudes at ~12 km, and at tropical latitudes near ~16 km. The TTL is located in the FT between about 12 and 16 km at tropical latitudes. The UT refers to tropospheric air above about ~6 km altitude. The LS refers to stratospheric air below ~20 km altitude. The MS represents the middle stratosphere between ~20 km (i.e. excluding the lowermost stratosphere) and ~35 km. The upper stratosphere plus mesosphere are defined to extend from ~35 km up to ~80 km altitude globally. No requirements for atmospheric composition above ~80 km have been specified. The given domains and their boundaries are all to be considered as a simplified version of the real, variable atmosphere. Thus, none of the defined boundaries should be interpreted as hard numbers.

4.1.2 Coverage and Sampling Requirements

In general, for each of the listed satellite products the goal coverage is global. This requirement directly reflects the global nature of the three driving environmental themes. Only for the air quality theme, with its additional focus on local, regional and continental scale environmental air quality issues, the required coverage for European-scale operational applications is the European continent, including Turkey, and Europe’s

/166

S-4/-5 MRTD



surrounding coastal waters as well as the closest parts of the North-Atlantic, which typically impact on the PBL in Europe by long-range transport.

The general goal requirement on sampling is (near-)contiguous sampling. It is clear that no measurement (sub-)system can be envisioned, nor it is desirable or necessary, with continuous and global-scale sampling on the defined spatial resolution and with the defined revisit times. The integration of a single measurement (sub-)system in an integrated system may allow for ‘data gaps’ in time and space to a certain extent.

On the other hand, in order to have an efficient overall measurement system, the aim of the measurement (sub-)system should be to maximise the number of independent observations to be made by that measurement system, the sampling being mostly limited by the other data requirements on uncertainty, spatial resolution and revisit time. Subsystems with (severe) limitations in coverage and sampling will contribute less to the integrated system and therefore typically should have less priority for operational applications.

4.1.3 Uncertainty, Spatial Resolution and Revisit Time Requirements

The following strategy to the derivation of quantitative data requirements has been followed. First, for each application a list of observables has been compiled for which the data requirements on spatial resolution and revisit time have been specified. In a next step, and on the basis of the given spatial resolution and revisit times, the requirements for the uncertainty have been specified.

This logic has been followed because the data requirements on spatial resolution and revisit time reflect the atmospheric variability of the observable, which is primarily a function of the time- and spatial scales of the atmospheric and surface processes that are relevant for the observable. Given the relevant temporal and spatial scales the amount of variability of the observable on these scales can be investigated. The different requirements for an observable (uncertainty, spatial resolution, revisit time) cannot be assessed independent from each other.

4.1.3.1 Uncertainty Requirements

The uncertainties that are given for each of the observables should be read as the maximum (threshold) uncertainty that is allowed in order to obtain information on the observable at the specified spatial resolution and revisit time. Whether the uncertainty is reached with a single retrieval or with a combination of retrievals will depend on the sampling and measurement techniques used. Requirements for these have not been specified.

In data assimilation systems the (assumed) uncertainty of the measurement drives the potential impact of the observation on the system. Therefore, the requirements on uncertainty are the most quantitative and, in fact, leading requirements, at least in comparison to the related requirements on spatial resolution and revisit time. The uncertainty for which the requirement is set will typically contain both a random component (‘root mean square error’) and a systematic (‘bias error’) component. The latter component should be established by a long-term validation with independent measurements. Constant biases are often less important. Requirements on regional biases and random errors are more difficult to define separately, and their relative importance

/166

S-4/-5 MRTD



will be dependent on the application (e.g. trends). The relative contributions of random errors and biases will also be very much dependent on the observational technique. The uncertainty requirements in the tables are to be understood as the root sum square of random and systematic errors.

A representation error may contribute to the uncertainty, which should be taken into account in the assessment of the uncertainty requirements. In general, the requirement is that the measurements be sufficiently representative for the given spatial resolution and revisit time. For satellite measurements the representation errors will typically contribute less to the uncertainty than for ground-based data, at least as long as the pixel sizes of the space data and the model grid sizes are of the same order of magnitude or the pixel sizes are larger.

General requirements on sampling and coverage have been specified above. Sampling is also constrained by the given spatial resolution and revisit time requirements. In some cases enhanced temporal or spatial sampling could somewhat relax the uncertainty requirement on an individual retrieval. However, the extent to which relaxation is possible typically depends on the forecast correlation lengths of the assimilation system. These are dependent on atmospheric conditions (see also below). The main limitation on sampling is that the additionally sampled observations need to be independent. A clear advantage of extensive, independent sampling is that a large number of available observations from prolonged data sets with stable retrievals and limited instrumental drifts during the mission lifetime typically will help the data assimilation system to better characterise the random and systematic components of the uncertainty. In this way sampling is related to the uncertainty.

The impact of observations with a certain uncertainty on a data assimilation system will also depend on the (assumed) model forecast uncertainties. These will typically vary in time and place. This is a complicating factor that has not been taken into account in the derivation of the measurement uncertainty requirements. It can be anticipated that at locations and times with small model uncertainty (e.g. because in-situ observations are available) the uncertainty requirements on the observations can be relaxed to a certain extent. This effect will become more important as models will improve in describing atmospheric transport and chemistry in the future. On the other hand, atmospheric composition is also to a large extent determined by intermittent processes and ‘unpredictable events’. Because of the unpredictable nature of atmospheric composition (in time and space) it is not desirable to relax a data requirement based on limited model uncertainties in transport or chemistry.

4.1.3.2 Horizontal Resolution Requirements

The horizontal resolution requirements are somewhat less quantitative than the uncertainty requirements. As a rule of thumb, the horizontal resolution should be at least a factor 2-3 smaller than the error correlation length in the model that is used in the assimilation of the observable. In fact, the assimilation typically combines the available observations within an area defined by the model forecast error correlation length. These are typically a function of altitude in the atmosphere and are mainly determined by the spatial scales of the relevant atmospheric processes and by the resulting spatial variabilities

/166

S-4/-5 MRTD



in the observables. Typically, the correlation length decreases from several hundreds of kilometres in the (lower) stratosphere to several tens of kilometres in the lower troposphere and even smaller in the PBL. Correlation lengths in the upper stratosphere and mesosphere are typically smaller than in the lower stratosphere. In some special cases the observation of scales smaller than those defined by the model forecast error correlation length might be very useful as well, e.g., to validate the model on the cascade of processes as a function of spatial scale and parameterisations of sub-grid scale processes.

4.1.3.3 Vertical Resolution Requirements

The vertical resolution requirements are in the first place related to the gradients of the observable in the vertical direction. Present-day estimates of vertical correlations show very short correlation lengths in the lower stratosphere and Upper Troposphere-Lower Stratosphere (UTLS) region due to their stratified nature, and much longer correlation lengths in the well-mixed troposphere. In the middle and upper stratosphere the distributions of the observables vary more smoothly in space and the requirements can be limited to a few kilometres in vertical resolution. In contrast, in the UTLS the vertical gradients (and thus the model error correlation lengths) can be very steep and highly variable in time. This results in rather stringent requirements. The vertical gradients in the troposphere typically depend on the synoptic situation and are mainly controlled by convective events and large-scale subsidence. Note that in contrast to turbulent mixing, convection can either steepen or smooth gradients. The faster overturning in the troposphere transports the information coming from observations more efficiently throughout the model vertical domain than in the UTLS. Therefore the vertical resolution requirements can typically be somewhat more relaxed in the free troposphere than in the UTLS region. Especially in the UTLS region and lower stratosphere the vertical fine-structure of models (dynamics) is not well tested due to a lack of high-resolution vertical information, e.g., with respect to atmospheric waves, and relevant for the general (Brewer-Dobson) circulation.

4.1.3.4 Revisit Time Requirements

Requirements on the revisit time can, in principle, be determined from examination of the anomaly correlations in an assimilation system. One could argue that if the anomaly correlation drops below a certain predefined threshold, the time evolution as described by the model is not sufficiently adequate and a new analysis based on observations, is needed. The lifetime of the analysis increments depends on the growth of the model forecast error in time. Following this argument the required update frequency would determine the required temporal resolution for an observable. However, it is difficult to estimate the extent to which future (and likely improved) models are able to describe the time evolution of the atmosphere. Current assimilation models have already proven skill for the prediction of stratospheric transport up to more than a week ahead (and possibly longer, depending on the required accuracy). Model skill to describe the evolution of tropospheric transport is much more limited because of the intermittent and unpredictable nature of several processes and events. The model skill on predictability is often limited by the predictability of the meteorological variables (wind, temperature) on which atmospheric composition typically has little influence, at least in the troposphere.

/166

S-4/-5 MRTD



Here, instead of using extensive studies on the anomaly correlation or the model error growth per time step, the requirements on the revisit time for the observables are derived from the typical model forecast error correlation lengths and the atmospheric variability in time of the observable. For example, at the higher altitudes the observables with a diurnal cycle should be observed at least twice daily (e.g. day/night, etc.), while for the other observables daily to weekly observations would probably suffice. The required revisit time typically increases in the lower troposphere and PBL, as does the complexity of models to describe the time evolution of the atmosphere. Depending on the relevant atmospheric processes and the geographic location, the required revisit time in the PBL can typically vary from several times daily to less than one hour.

Finally, it is noted that the spatial and temporal resolutions that are or will be used in present-day and future atmospheric models play only a (minor) role for the resolution requirements, because the requirements are determined by the scales of atmospheric processes, which may be either resolved or sub-grid processes in a model. It should also be noted that revisit time requirements are typically not related to (either near-real time or offline) data delivery time requirements.

4.2 Geophysical Observation Requirements per Application

In this section the geophysical observation requirements are outlined and the resulting data requirements table corresponding to each application can be found in Appendix 2. In sub-sections 4.2.x.y and 4.2.x.y.z, labels A1 … C3 refer to Table 4.1.

4.2.1 Theme A: Stratospheric Ozone and Surface UV

4.2.1.1 A1 - Protocol Monitoring and Treaty Verification

4.2.1.1.1 A1 - Relevant Species and Processes The Montreal Protocol and its subsequent Amendments and Adjustments form the main driver to the monitoring of stratospheric ozone and surface UV radiation. Long-term monitoring of the expected decrease in polar and global ozone loss in response to the measures taken based on the Montreal Protocol and its amendments is required. The ultimate goal is to obtain accurate information on the evolution of the ozone layer (total column) and its effect on surface UV, together with the monitoring of columns of ozone depleting substances (ODS); CFC’s and their replacement HCFCs, and halons. Specifically information on the changes (trends) in chlorine loading is needed, both in the troposphere and in the stratosphere.

More detailed policy-relevant information includes the monitoring of the height distribution of ozone and ODS compounds, in addition to total column information. Ozone profile information also allows separation of long-term changes in tropospheric component, mainly relevant to the Air Quality and Climate themes, from changes in the stratospheric component relevant to the Montreal Protocol. These aspects are all considered under ‘Assessment’ in Section 4.2.1.3.

Another user requirement is that the sources of ODS need to be identified and quantified. Currently this is done bottom-up from country-wise official figures. However, independent

/166

S-4/-5 MRTD



verification by inverse modelling of the concentration distributions would be highly desirable. Limiting factor for inverse modelling of ODS is however their fairly homogeneous distribution.

The user requirements for operational surface UV radiation monitoring relevant to the Montreal and subsequent protocols need some consideration. In fact the protocols are directed to reduce UV increases that are related to (anthropogenic-induced) changes in total ozone column. On the other hand, the importance of these ozone-related long-term UV changes also need to be viewed in relation to (possibly larger) surface UV changes induced by long-term variations in other processes, including the locally and in time varying effects of clouds, aerosols and surface albedo.

For the long-term monitoring of the surface UV radiation it suffices to monitor on a global scale the clear-sky UV Index and the daily UV dose. The clear-sky UV Index is an adequate measure that is directly related to (variations or trends in) the total ozone column amount. Next to the total ozone column the main other modulators of the UV Index are the solar spectral irradiance, solar zenith angle and Sun-Earth distance, surface elevation, surface albedo, stratospheric temperatures (via ozone absorption) and aerosol optical parameters. A global daily monitoring of the noontime clear-sky UV Index will also give information on the occurrence of extreme values, which are typically related to ozone depletion events.

The daily UV dose is defined as the 280-400 nm spectrally-integrated erythemally-weighted surface irradiance integrated over daytime. In the interpretation of UV dose variations and trends due to ozone depletion other processes that may result in long-term changes in surface UV radiation levels should be taken into account. Most important for long-term UV dose monitoring, i.e., over decades, are possible systematic changes in the effects of clouds, aerosols, UV surface albedo, and the solar spectral irradiance.

4.2.1.1.2 A1 - Measurement Strategy and Data Requirements Given the user requirements on long-term homogeneity and global coverage of the data sets and the trend requirements, the most advantageous approach for protocol monitoring is the integration of spaceborne and ground-based data in an assimilation system. Specific requirements call for spaceborne total ozone columns (5% standard deviation; 5% bias). The ozone profile should distinguish different atmospheric domains, at least including the lower troposphere, the upper troposphere, the lower stratosphere, and the upper stratosphere and mesosphere. The threshold ozone monitoring requirements can be summarised as follows: Horizontal resolution: 100 km; Vertical resolution: column (mandatory), 4 independent pieces of information (desirable); Temporal resolution: 24 hrs; Uncertainty: standard deviation 5%, bias 5%). Note that some of these requirements are covered under Assessment in Section 4.2.1.3.

Based on present-day experience with the assimilation of total ozone column information in chemistry-transport models, the required information can be obtained by global spaceborne observations with about 3-days revisit time such as typically provided by ERS-2 GOME. The user requirement on trend detection is rather stringent (~0.1% per year). Although this number applies to the zonal monthly means, the trend requirement is driving the uncertainty requirement of 3% on an individual total ozone column measurement. Neglecting biases, typically ~900 independent measurements per zonal band (of 100 km

/166

S-4/-5 MRTD



width) and per month would then suffice to reduce uncertainty by a factor 30 as required (3% => 0.1%).

The monitoring of (the trend in) the ODS in troposphere and stratosphere can be best performed using a representative surface network, measuring weekly background surface concentrations and total column amounts of the various regulated ozone depleting substances as listed by, e.g., WMO in the ozone assessment reports. In the data requirement table only the most abundant ODS are listed. Furthermore, especially ODS for which surface-based historical records are available at present are the most relevant for future protocol monitoring. A representative surface network, with at least one background station in each ~10 degrees latitude band, will typically suffice for the determination of total equivalent chlorine in the atmosphere as well as for the derivation of trends in CFC concentrations and trends in their emissions. For the annual trends, typically zonally averaged, weekly representative values with uncertainties of ~2% are needed for the CFCs and other long-lived ODS, and ~5% for the HCFCs.

Independent verification of ODS emissions by inverse modelling of the concentration distributions would be desirable. However, owing to the long chemical lifetime of the ODS, and hence their fairly uniform global distribution, this would be a challenging task. On the other hand, it has been shown already that trajectory analyses of surface-based time series of long-lived compounds sufficiently close to emission regions can be used to trace back the emissions to a certain region. Currently it is not foreseen that such detailed studies can be performed on an operational basis. Spaceborne observations of ODS columns are not likely to contain sufficient information to contribute significantly to inverse modelling of ODS emissions. A rather dense surface network would be required to derive country-based (monthly) ODS emission numbers, typically one station per country and further every 10000-100000 km2.

Operational surface-based observations from a global representative surface network are needed for continuous validation of the ozone column spaceborne observations. Ozone sonde observations, especially in the polar regions, are needed to provide additional information on ozone that could be difficult to obtain by spaceborne observations, including the altitude(s) of extreme ozone loss.

The surface UV radiation requirements include a requirement on long-term time series and regional maps of the daily noontime clear-sky UV Index, typically with at most 1 index point accuracy. The uncertainty of a Level-2 UV index product based on satellite observations should be better than ~10% for UV Index higher than 5 index points, and 0.5 index point for smaller UV Index values.

Given the known sensitivity of the UV Index for different parameters, the UV Index requirements can be translated into requirements for Level-2 products, e.g., maximal a few percent of change in UV Index per change of 0.1 in aerosol optical depth. The relevant products include, besides the total ozone column, the solar spectral UV irradiance and its modulations over time, the aerosol optical depth and absorption optical depth and the UV surface albedo. Trace gases such as NO2 and SO2 absorbing in the UV spectral range have a very minor effect on UV radiation levels.

/166

S-4/-5 MRTD



For the surface UV daily dose the estimated uncertainty requirement is 0.5 kJ m-2 (for reference: a maximum daily dose at tropical latitudes is ~ 8 kJ m-2, typical values range from 1 to 5 kJ m-2). Apart from the effect of clouds considered below (Section 4.2.1.1.3), the same Level-2 products as for the clear-sky UV Index are needed to derive the daily UV dose.

4.2.1.1.3 A1 - Auxiliary Data Requirements The ozone layer monitoring requires assimilation of the observations in an atmospheric model. Therefore, additional information is needed on the meteorological state of the stratosphere. At the time this information is assumed to be available from the analyses of NWP models.

For the attribution of UV changes to ozone changes, auxiliary information is needed on the global distribution and possible changes over time in:

3-D cloud optical and geometric parameters (mainly cloud optical depth and cloud cover)

Stratospheric temperatures (determining the UV absorption for a given ozone amount)

The UV extraterrestrial solar spectrum, covering the 200-400 nm spectral range

3-D aerosol optical parameters in the UV (mainly aerosol optical depth and single scattering albedo)

2-D UV surface albedo global distribution

The latter three bullets are covered in the data requirement table of A1-S. Stratospheric temperatures are assumed to be available with adequate accuracy from the analyses of NWP models.

Quantitative requirements on cloud properties are not yet available. However, given the large, often dominating effect of clouds on the daily UV dose and its changes over time and place, the required cloud information needs to be quite detailed in time and space in order to be able to derive information on surface UV variations and trends that can be related to ozone changes as required here for protocol monitoring. Typically, for the interpretation of the UV dose accurate cloud information is needed on cloud cover and cloud optical depth as a function of time over a day with time steps of about an hour or less. Here, it is assumed that the required information on cloud parameters will be available from existing or planned meteorological platforms (e.g. MSG, GOES). A good cloud mask (on/off) is the most crucial requirement.

In the mapping of the UV daily dose the various Level-2 data products that are needed are typically gridded (Level-3/-4) before these are combined. Requirement on co-location of the various products are therefore not considered very stringent. The different products may be derived from different platforms, including for example a platform in low orbit for total ozone, the solar spectrum, aerosols, and surface albedo, and a geostationary platform for variables that typically change significantly over the day (cloud parameters and, possibly, aerosol parameters).

/166

S-4/-5 MRTD



4.2.1.2 A2 - Near-Real Time Data Requirements

4.2.1.2.1 A2 - Relevant Species and Processes Forecasts of ozone fields and surface UV radiation are required for different user groups. Near-real time ozone data are required for improved radiances in NWP models and as input data for surface UV forecast. For forecasts a data assimilation system is needed to integrate the near-real time observations and to combine these with transport information from the model forecast. It has been shown that with present-day NWP models reliable total ozone and clear-sky UV Index forecasts are possible up to ~1 week ahead.

Near-real time information on the ozone layer is also required during periods of severe (polar) ozone loss to inform policy-makers, the media and the general public. Currently, near-real time data relevant to Arctic ozone loss has been intended for scientific use only, e.g., related to Arctic measurement campaigns. Especially extensive ozone loss that takes place in the Arctic during cold winters is a cause of great concern due to its proximity to inhabited areas. Forecast of, e.g., the Antarctic vortex break-up would contain important information especially for some countries in the Southern Hemisphere including Argentina, Chile, New Zealand and some small islands.

For surface UV radiation forecasts, such as provided in most countries by the meteorological institutes, total ozone column forecast information is needed, typically for a few days ahead. Additional forecast information is required on clouds, aerosols and surface albedo. However, given the present-day uncertainties associated with the forecast of these additional parameters, current forecasts of surface UV radiation are often limited to so-called clear-sky values (at most including a fixed aerosol correction and in some countries taking into account known surface albedo variations). The reported clear-sky value therefore typically represents the most extreme case. Near-real time observations of aerosols and surface albedo are needed to reduce the uncertainty in their effect on the clear-sky UV predictions.

In some countries, an uncertainty range is presented on the UV Index forecast where the given range mainly reflects the prediction of the possible reduction of UV radiation by clouds. Improved cloud forecasts (mainly on cloud cover and cloud optical depth) would help to reduce the uncertainties that are associated with cloud predictions.

Note that UV Index forecasts need to report the highest expected value for the day, which is typically around noontime.

4.2.1.2.2 A2 - Measurement Strategy and Data Requirements Modelling of the evolution of the ozone layer over a couple of days (e.g., up to ~10 days) requires information on the full three-dimensional ozone layer distribution. Based on the initial field, the meteorological forecasts will be used to transport ozone in all dimensions and this will result in a new ozone field from which the required forecast of the spatial distribution of the ozone columns can be derived.

In order to accurately forecast ozone columns, near-real time spaceborne ozone profile measurements are needed in the UTLS region and above. In the troposphere a measurement of the tropospheric column suffices. Total ozone column observations can also be used, although at the expense of accuracy. Typically for the ozone profile the

/166

S-4/-5 MRTD



required vertical resolution decreases from about 2 km (threshold) in the UTLS region to ~5 km in the upper stratosphere and mesosphere. If the complete ozone profile cannot be covered by the measurements, additional information will be needed from measurements of the total ozone column.

Near-real time availability of surface-based observations of total ozone columns is needed to complement and validate the spaceborne observations. Furthermore, a representative ozone sonde network is needed for validation of the assimilated ozone distribution. In-situ ozone profiles are also needed to enhance the vertical profile information in the troposphere and lower stratosphere.

Both ground-based measurements and spaceborne estimates of the UV dose and UV Index are needed for the validation of the UV forecasts. Although the quality of (derived) surface UV radiation measurements is highly correlated with the quality of the total ozone observations, some differences between both data sets will occur because clear-sky surface UV radiation products are additionally weighted with solar zenith angle, aerosol load, and surface albedo. Some information on possible long-term changes in the incoming UV solar irradiance at the top of the atmosphere would also contain valuable information for UV forecasts.

A complicating factor for validation of the spaceborne surface UV estimates is the variability in aerosols and clouds. Near-real time observations of the UV spectral aerosol (absorption) optical depth and UV spectral surface albedo will help to reduce uncertainties in UV forecasts. The required spectral range for these products is the 280 – 400 nm spectral region. The required spectral resolution is typically 5 to 10 nm in the UV-B range (280 – 320 nm) and 10 to 20 nm in the UV-A range (320 – 400 nm).

A requirement for the forecast model is that the dynamics of the stratosphere are well-predicted and also that changes in ozone due to dynamics can be distinguished from changes in ozone that are related to chemical and/or radiative processes. Good vertical resolution is crucial to better represent stratospheric waves. It has been shown that inclusion of a parameterisation of heterogeneous ozone loss processes can improve the forecasted ozone distribution. For the stratospheric radiation budget the most important gases to assimilate together with ozone are H2O, CO2, CH4, and N2O.

Assimilation of tracer observations of SF6 or CO2 could be used to better separate between ozone transport and ozone chemical processing. Currently, parameterisations on ozone loss are based on the prediction of temperature. Ozone loss processing can be better constrained by observations of PSCs, enhanced ClO, and aerosol extinction.

Operational in-situ aircraft measurements in the UTLS region, co-located with ozone observations, of H2O, CO, HNO3 and HCl would be desirable to better constrain the stratosphere-troposphere exchange processes. Operational spaceborne observations of these gases in the UTLS region could possibly contribute as well.

Finally, near real time data delivery for this application implies that the data needs to be available to an operational modelling environment within a couple of hours after observation. In that case a significant part of today’s observations can still be used for the analysis on which the required forecast for tomorrow will be based.

/166

S-4/-5 MRTD



4.2.1.2.3 A2 - Auxiliary Data Requirements Ozone forecasts rely on an operational assimilation system including the meteorological analysis and forecast of stratospheric transport. The required meteorological fields, up to at least one week ahead, can only be delivered by NWP centres. Therefore, it is foreseen that forecast services will be run by these meteorological centres. The operational atmospheric composition products will contribute to the overall assimilation system. Significant experience has been obtained in the GEMS and MACC projects.

UV radiation forecasts are typically most relevant for clear-sky conditions as these typically represent the maximum level that can possibly be obtained. However, forecasts including the effect of clouds would be more realistic. Therefore, improved all-sky UV radiation forecasts would profit from improved forecasts of cloud parameters. Most important parameters for all-sky UV Index forecasts are, next to the information on ozone, aerosols and surface albedo, cloud cover, especially around noontime, and cloud optical depth.

For forecasts of the UV dose, a forecast of (the distribution of) the sunshine duration over the coming days would be the most crucial parameter, together with the above-mentioned cloud parameters relevant for the UV Index.

Improving cloud forecasts, especially with the aim to improve surface radiation forecasts, is extremely challenging. Even with near-real time availability of cloud observations current scientific knowledge of cloud processing likely does not allow for accurate forecasts of cloud distribution for the purpose of improving UV forecasts for typically 24 hours ahead. No requirements on cloud parameters are available.

Global radiation (pyranometer) measurements from the surface radiation networks could be another independent set of observations that can account for the cloud and aerosol effects on UV. Also, forecasts of global radiation are becoming available from NWP centres and these could give additional information that is useful to improve upon the UV forecasts.

4.2.1.3 A3 - Assessment

4.2.1.3.1 A3 - Relevant Species and Processes More detailed policy information than required for direct protocol monitoring (total ozone column, surface UV; Section 4.2.1.1) will be based on the monitoring of the height distribution of ozone and ODS compounds, related compounds and parameters other than ozone that affect the surface UV radiation. For example, ozone profile information is necessary in order to separate long-term changes in the troposphere ozone component, mainly relevant to the Air Quality and Climate themes, from changes in the stratospheric component relevant to the Montreal Protocol.

For the ODS altitude information would also give indication on the effectiveness of treaty implementation. Desirable is information on the stratospheric halogen loading, which includes also reservoir species such as HCl, ClONO2, HBr and BrONO2. In addition, monitoring of these reservoir species might be relevant for another reason: it is anticipated that changes in reservoir species typically would precede changes in total chlorine content and therefore would give an early indication of changes in equivalent chlorine. Certain active chlorine and bromine components (ClO and BrO) and PSCs are indicators of the

/166

S-4/-5 MRTD



amount, severity and extent of ozone depletion events, which is additional relevant information for treaty verification.

The main drivers for a better understanding of the ozone layer evolution and long-term changes in surface UV radiation are the following long-term science questions:

Understanding of the trends in total ozone, largely by examination of the evolution of the ozone layer and the changes in the ozone distribution over time

Understanding of the effects on the ozone layer of the policy measures taken in response to the Montreal Protocol and its amendments

Understanding of the global ozone chemical budget, including the relative roles of denitrification, heterogeneous chemistry and other ozone loss processes

Understanding of the processes resulting in interactions between ozone recovery and climate change, related to radiation, dynamics and/or chemistry

Understanding of the long-term changes in surface UV radiation levels, their attribution to either total ozone changes or other processes, and their effects on health and the environment

Understanding of the distribution of the ozone depleting substances and the trends in their concentrations

To answer these questions the scientific users require long-term global monitoring of the three-dimensional distribution of ozone, ozone depleting source gases, and some other long-lived key gases in the stratosphere, as well as stratospheric aerosols and PSCs. To understand changes in surface UV radiation additional information is needed on the various processes that affect surface UV radiation, most importantly besides ozone: clouds, aerosols, surface albedo and the solar spectrum.

Long-term operational data sets will be most essential to validate ‘slow’ processes in atmospheric chemistry models. With ‘slow’ processes reference is made to processes that are predicted to have significant effect on, e.g., the ozone layer on the long term, although the direct effect can be difficult to obtain from dedicated measurements that are typically limited to short time periods. One such ‘slow’ process is, e.g., the continuous increase of CO2 and other greenhouse gases concentrations in the atmosphere, which is predicted to affect the ozone layer by inducing changes in, e.g., the temperature distribution in the stratosphere. Another example is the observed slow increase in stratospheric water vapour, partly caused by CH4 increases, but largely not well understood. Further, the increase in stratospheric N2O concentrations is expected to enhance the relative role of the nitrogen cycle in stratospheric chemistry.

Operational measurements of atmospheric composition can mostly be limited to the longer-lived compounds. The measurement of short-lived compounds on an operational basis is considered of less relevance because a lack of scientific understanding of a certain chemical or physical process is likely to benefit more from dedicated (campaign) measurements than from operational data. Operational measurements, however, can help to quantify the relative importance of different (fast) processes on the long term, e.g. in

/166

S-4/-5 MRTD



relation to the contribution of the hydrogen, nitrogen and halogen cycles to the chemical ozone budget.

4.2.1.3.2 A3 - Measurement Strategy and Data Requirements The monitoring of changes in the vertically-resolved concentration distributions in the global stratosphere is crucial to understand the long-term evolution of the ozone layer. Long-term ozone changes occur at different altitudes and at each altitude different chemical, dynamical and radiative processes play a role. Vertical resolution is most critical in the UTLS region where stratosphere-troposphere exchange processes result in large gradients in the ozone distribution. A goal vertical resolution of 1 km is given for spaceborne ozone observations, with a threshold of 3 km resolution. Especially in the latter case the spaceborne observations in the UTLS would benefit if complemented by more detailed ground-based and airborne observations. In the middle and upper stratosphere the ozone distribution is less variable and the required vertical resolution is typically relaxed to 3-5 km. Total column information is needed in cases when the vertical profile is not covered in all atmospheric domains. Spaceborne observations of the tropospheric column (in combination with an averaging kernel) would help to distinguish from total ozone observations between changes in tropospheric ozone and changes in stratospheric ozone. Ground-based networks and airborne UTLS observations are needed to enhance the profile information on tropospheric ozone and to better quantify changes in the net ozone flux from the stratosphere into the troposphere.

Monitoring of the total stratospheric halogen loading requires spaceborne stratospheric profile observations of the main reservoir gases: HCl, ClONO2, HBr and BrONO2. The reservoir gases are spatially and temporally much more variable than the ODS. Vertical profiles with about 3 km resolution covering the lower and middle stratosphere suffice. Additional HNO3 stratospheric profile information is desirable to observe possible long-term changes in denitrification. Typically a zonal mean uncertainty of ~20% could be allowed for data that is representative for a few days to one week. For some gases an uncertainty for a 1000 km-average has been specified to account for anticipated longitudinal variations in these compounds.

The uncertainty requirements for ClO (for enhanced levels) and BrO of ~50% are set, e.g. as occurring in spring in the polar stratosphere. For protocol monitoring these short-lived gases, responsible for at least 50% of springtime stratospheric ozone loss, are mainly desirable to detect the number, location and extent of events with excessive ozone loss, i.e. statistics. For the same reason the data requirement on PSCs is also limited to detection only (instead of full characterisation, see the section on ‘ozone layer: understanding’).

Several long-lived gases are important to be monitored for a better understanding of the evolution of the ozone layer. These include at least H2O, CH4, N2O and HNO3. The gases play multiple roles in the stratospheric physical system. Most important is the long-term trend of these gases as well as information on possible changes in their vertical and zonal distributions, e.g., changes in the HNO3 distribution can be related to long-term changes in denitrification. Also NO2 observations are considered very useful in this respect.

Ground-based networks of surface concentrations and total columns are most suited for the determination of trends in ODS, of which the most important are CFC-11, CFC-12 and

/166

S-4/-5 MRTD



HCFC-22. Desirable information for understanding the ozone layer evolution in response to policy measures taken in response to the Montreal Protocol and its amendments would be further the measurement of the gases CFC-113, HCFC-123, HCFC-141b, HCFC-142b, CCl4, Halon 121, Halon 1301 and Halon 2402. In this list CH3CCl3 is neglected because it is assumed to be of minor relevance for ozone depletion after 2010. Spaceborne observations are useful to complement the ground-based measurements and to verify the representativeness of the ground-based networks for global trend determination of the ODS concentrations. Spaceborne profile observations of other source gases such as CH3Cl and CH3Br as well as reservoir gases such as HCl, ClONO2 would further aid understanding the diminishing role of the anthropogenic ODS to the ozone layer evolution.

Spaceborne measurements of SO2 and volcanic aerosol would be needed for understanding the ozone layer evolution in case of severe volcanic eruptions polluting the stratosphere for a couple of years, e.g., comparable to the effect of the Pinatubo eruption in 1991.

Understanding of surface UV radiation changes and their possible effects on health and the environment requires long-term spaceborne monitoring of the 3-D ozone distribution (i.e., preferably ozone profiles), the UV aerosol optical depth, the UV aerosol absorption optical depth or single scattering albedo, the UV surface albedo, and the extraterrestrial solar spectrum in the UV range.

Finally, operational ground-based measurements from a representative global network are needed for continuous validation of the mentioned spaceborne measurements and derived surface UV products.

4.2.1.3.3 A3 - Auxiliary Data Requirements For the ozone assessment the interpretation of the combination of ground-based observations and spaceborne observations would be most beneficial if the observations are assimilated in chemistry-transport models. The main auxiliary requirement is therefore on the availability of state-of-the-art chemistry-transport models, preferably covering the atmosphere from the surface to the mesosphere and making use of analysis fields of NWP models, detailed emission databases (both natural and anthropogenic), and adequate chemical schemes.

In addition, the interpretation of long-term variations and trends in stratospheric composition requires information on climate and climate evolution. Especially relevant is climate monitoring of the variations and trends in the main meteorological parameters in the stratosphere and mesosphere (temperature, air density, winds, Brewer-Dobson circulation, etc).

4.2.2 Theme B: Air Quality

4.2.2.1 B1 - Protocol Monitoring and Treaty Verification

4.2.2.1.1 B1 - Relevant Species and Processes Within the Air Quality theme the main drivers for protocol monitoring are the EMEP and Gothenburg Protocols of the UN/ECE CLRTAP convention, the National Emission Ceilings, as well as complementary regulations related to EU Air Quality policy, e.g. in

/166

S-4/-5 MRTD



relation to the CAFE (Clean Air for Europe) program (see Table 3.1). The user requirements include the monitoring of the total abundances (incl. threshold value exceedances) and concentration distribution of the regulated gases and aerosols as well as the detection and source attribution of the related emissions for verification. In order to observe peak concentration levels, e.g. as related to rush hours or to accidental chemical releases, monitoring of hourly surface concentrations are typically needed. In order to monitor the effect of policy measures it is necessary to be able to derive information on trends in concentrations and emissions within a time frame of maximum a few years.

Air Quality data requirements primarily should respond to the need for information on pollution levels at ground level and in the planetary boundary layer (PBL, typically between surface and ~1-2 km altitude) where they impact on the health and safety of people and of the biosphere. However, additional information on the composition of the adjacent free troposphere is also important as boundary condition to the PBL. The long-range transport and free-tropospheric photochemistry determine the background concentrations of the longer-lived pollutants on which locally pollution builds up.

The compounds for which the surface concentrations are regulated include O3, SO2, NOx, Particulate Matter (PM10, PM2.5, PM1 in (g .m-3), denoting particles with diameters smaller than, respectively, 10, 2.5 and 1 microns), CO, benzene (C6H6), Poly Aromatic Hydrocarbons (PAHs), and some heavy metals (Pb, Ni, As, Cd, and Hg). Regulations on PM1 are anticipated.

Driving the requirements on emissions are the National Emission Ceiling Directives for SO2, NOx, Volatile Organic Compounds (VOCs), NH3 and fine particulate matter. Also the CLRTAP convention, which includes Europe, Russia, US and Canada, sets emission ceilings on SO2, NOx, VOCs and NH3, by the EMEP and Gothenburg protocols. The Gothenburg protocol also regulates surface ozone levels.

With reference to the GMES-GATO report /GAT2004/ it is recommended to anticipate possible future regulation of ship emissions. The most important ship emissions include CO, NO2, SO2 and particles. Concentrations of these compounds need to be monitored for operational shipping in harbours, main waterways, and over coastal waters.

4.2.2.1.2 B1 - Measurement Strategy and Data Requirements Traditionally, the requirements for monitoring and verification of air quality have been formulated based on the means already available for verification and enforcement, which consist of ground-based networks at the local and regional authority level. Even though the data quality issue is addressed in the EC framework directive, at present these data are often of limited use in a global observation network because of lack of standardisation of the instruments employed and the data generated. Furthermore, continental and hemispherical or global coverage cannot be obtained by ground-based networks in practice. An optimal strategy for air quality protocol monitoring and verification would be based on a synthesis of satellite observations, ground-based networks and air quality model information through data assimilation on different spatial scales.

It has been shown that local to regional air quality models are very useful to complement the ground-based networks, e.g. to interpolate in time and space. However the models are also essential because these include meteorological information on the PBL, e.g. based on

/166

S-4/-5 MRTD



NWP model output. For example, the PBL height is crucial for the surface concentration levels that are attained as it determines the extent of the planetary boundary layer and as such the atmospheric volume in which surface emissions are injected. Other meteorological variables that can be delivered by the air quality model and are essential for the surface pollution levels include the wind speed and direction, turbulent mixing, temperature, water vapour, UV radiation, clouds and convection. In addition the model can include detailed information on natural emissions, also based on surface characteristics such as vegetation and snow cover. For example, ozone levels in rural, moderately polluted regions are known to be very sensitive to meteorology-dependent isoprene and monoterpene emissions.

Spaceborne observations can help to fill in gaps in the surface networks, although global-scale satellite measurements cannot be expected to be of sufficient resolution and accuracy to deliver accurate information on local surface concentration levels. Spaceborne observations are crucial, however, for the boundary conditions of the air quality models. These models are typically limited to a certain region and therefore highly dependent on appropriate boundary conditions, especially for the meteorology and the longer-lived compounds. These boundary conditions, e.g., for chemical compounds over the oceans, can typically be delivered by global model output in which satellite observations of tropospheric composition have been assimilated.

Inverse modelling will be needed to derive emissions based on concentration distributions. Currently, the intrinsic limitations of ground-based observations also hamper the emission verification using inverse modelling. Independent observations from satellites will help to better constrain the inverse modelling. Note that especially the performance of the air quality model will be crucial for the quality of the emissions that can be inferred using inverse modelling techniques. It is anticipated that with the increasing level of detail incorporated in the air quality models the uncertainties related to inverse modelling of emissions will become smaller in the coming years. In order to derive emissions on a country-by-country basis or better the density of the surface network should be typically 10000-100000 km2, with at least one measurement station per country.

The surface network for protocol monitoring should be representative for the polluted regions in Europe and include at least surface concentration measurements of O3, SO2, NOx, PM10, PM2.5, PM1, CO, benzene (C6H6), Poly Aromatic Hydrocarbons (PAHs), ammonia (NH3) and heavy metals (Pb, Ni, As, Cd, and Hg). Note that requirements for ground-based measurements are limited to compounds for which spaceborne observations play a role, i.e., requirements for, e.g., PAHs, ammonia and heavy metals have not been derived. Long-term homogeneous measurement series are needed in order to derive trends in the surface pollution levels. About 10% uncertainty on individual measurements should be sufficient both for the hourly peak levels and for the detection of small long-term trends in monthly mean peak values.

The satellite measurements of trace gases should include preferably tropospheric profiles of O3, SO2, NO2, CO and formaldehyde (CH2O), at least separating the PBL from the free troposphere. The threshold vertical resolution requirement is for a tropospheric column measurement, in combination with an averaging kernel in order to have information on the sensitivity of the satellite measurement as a function of altitude. Note that formaldehyde is required because it will contain important information to constrain the VOC emissions.

/166

S-4/-5 MRTD



The required revisit times are typically between half-hour (goal) to several hours and are directly related to the protocol requirements to observe hourly peak pollution values, in combination with the fast chemistry and mixing time scales of the PBL. The revisit time requirements are typically for daytime only (this is the threshold requirement) as photochemistry is a major driver for the pollution levels. The extension to full 24 hour coverage, i.e., including the night-time evolution, is a goal requirement and can be a useful additional constraint to air quality models, especially for ozone and nitrogen compounds (NOX, N2O5, HNO3, PAN).

The uncertainty requirements typically do not pertain to very clean or background levels. However, it is still needed to measure in the background atmosphere and to assign these pixels as being background or below the detection limit. This is especially true for SO2, NO2 and CH2O satellite observations for which the threshold uncertainty is expressed in absolute terms. Column amounts of <1.3·1015 molecules cm-2 correspond to background conditions, with column average concentrations below 1 ppbv. The uncertainty is given in absolute terms (1.3·1015 molecules cm-2) and corresponds, e.g., for NO2 with 100% relative uncertainty for a column of 1.3·1015 molecules cm-2 to <10% uncertainty for columns larger than 1.3·1016 molecules cm-2. Note that satellite NO2 measurements are assumed to suffice for constraining NOx emissions and ambient levels. This assumption sets some basic requirements on the chemical scheme to be used in the Air Quality model for the NO/NO2 conversions.

Maximum uncertainties for PM10 and PM2.5 surface concentrations have been fixed in absolute terms at two times the measured background concentration in Europe /Din2004/. For PM1 requirements could not be specified as information on the background concentrations is lacking.

The vertical resolution requirements on the satellite observations of aerosol optical depth are similar to the satellite requirements on trace gases, with a goal to distinguish between aerosols in the PBL and free troposphere and a threshold for the tropospheric aerosol optical depth. The required uncertainty (0.05) is again expressed in absolute terms and based on different earlier assessments. The aerosol optical depth observations can be used to constrain the surface concentrations of PM. Information from satellite on aerosol type would be desirable.

The requirement on ship emissions extends the need for surface measurements to coastal waters. These ground-based measurements should include at least CO, NO2, SO2 and particles. The same compounds over coastal waters measured from satellite would add significantly to the ship data.

In addition to the monitoring network for surface concentrations, ground-based observations are also needed for the validation of the models and satellite observations in the troposphere. The observations should include ozone profiles from the sonde network as well as tropospheric column data at representative sites for the validation of the modelled and spaceborne observations of tropospheric ozone. Lidar observations at specific sites are very useful to validate the vertical tropospheric profiles of O3, NO2, SO2 and CH2O. PBL concentration profiles from towers at a few locations would also help to validate satellite data and models.

/166

S-4/-5 MRTD



4.2.2.1.3 B1 - Auxiliary Data Requirements Air quality protocol monitoring relies heavily on the combination of ground-based observations, air quality models and satellite data. The main auxiliary requirement is therefore on the availability of state-of-the-art air quality models making use of analysis fields of NWP models, detailed emission databases (both natural and anthropogenic), adequate chemical schemes and detailed descriptions of surface-atmosphere exchange processes.

4.2.2.2 B2 - Near-Real Time Data Requirements

4.2.2.2.1 B2 - Relevant Species and Processes The main societal drivers for air quality forecasting are health and safety warnings. Surface concentration predictions are needed from local street-level to regional and national scales. Typically the maximum delay time allowed for data delivery is very short, about 30 minutes. The so-called Air Quality index, according to EC directives, is based on a mixture of O3, NO2, PM10, SO2, and CO. These compounds are affecting respiratory health. Because particle size is important, distinction is made between PM10, PM2.5 and PM1. Particles are possibly also related to cardiovascular health. Metals in particles could also be an issue.

With respect to safety natural hazards such as volcanic eruption, forest fires and man-made hazards such as biomass burning and chemical and nuclear releases require plume transport and dispersion model forecast fed by observations. An additional driver here is air traffic management, including both air routing and early warnings for the mentioned unpredictable events.

An important requirement for health and safety is further near-real time source detection and attribution of the emissions of aerosols and aerosol and ozone precursors (NO2, SO2, and CO).

Additional information on methane (CH4), water vapour (H2O), formaldehyde (CH2O) as well as the UV-VIS photolysis rates is important for the forecasting of the photochemical activity. These observations are needed to constrain the chemical conversion rates and help to determine the atmospheric residence time of pollutants.

4.2.2.2.2 B2 - Measurement Strategy and Data Requirements The optimal strategy for air quality forecasting is similar to the strategy for air quality monitoring described in Section 4.2.2.1 and based on a synthesis of satellite observations, ground-based networks and air quality model information through data assimilation on different spatial scales. The main difference is the requirement on the timely availability of the forecast information.

Typically, environmental agencies require air quality forecasts for the day to be available in the early morning. The time delivery requirement on the observations for air quality forecasts is therefore mainly determined by the need for the data to be available for the integrated forecast system at the time that the analysis run is performed on which the forecast run will be based. In practice, the analysis run will have to be performed in the late evening or early night in order to do a forecast run that finishes in early morning. The delivery time requirement is therefore about several hours. Given that the daytime

/166

S-4/-5 MRTD



observations are most relevant, the most stringent delivery requirements are for the last daytime measurements of the day. The user requirements on the timely availability of the forecasts prevents the need for observations of the same day as for which the forecast is being made. This is true for satellite observations as well as for observations of the ground-based networks.

The revisit time satellite data requirements are for daytime only (threshold), except for N2O5, HNO3 and PAN for which especially night-time observations would be desirable, given their role in the night-time NOy budget, which is an important constraint on the amount of NOx released from reservoir species after sunrise. The threshold revisit time requirements of 2 hours are mainly related to the diurnal cycle of air pollution levels as well as the short timescales of the mixing and chemical processes in the PBL.

The satellite measurements of trace gases should include preferably tropospheric profiles of O3, H2O, SO2, NO2, CO and formaldehyde (CH2O), at least separating the PBL from the free troposphere. The threshold vertical resolution requirement is a tropospheric column, in combination with an averaging kernel in order to have information on the sensitivity of the satellite measurement as a function of altitude. Note that formaldehyde is required because it contains information on the amount of photochemical activity caused by hydrocarbons. Water vapour profile information is important for the effect of relative humidity on aerosols as well as for the primary OH production, which controls the photochemical activity together with the ozone concentration and UV-VIS actinic flux.

The aerosol requirements on the satellite observations are on the aerosol optical depth and the aerosol type, with a goal to distinguish between aerosols in the PBL and free troposphere and a threshold for the total tropospheric aerosol optical depth. The required uncertainty (0.05) is expressed in absolute terms and is based on different earlier assessments. The aerosol types to be distinguished include at least standard categories such as sulphate, dust, sea salt, organic carbon, black carbon, and mixed aerosol. The requirement on aerosol type is that mis-assignments should be limited to less than about 10% of the cases.

For air traffic management the threshold coverage requirements on aerosol optical depth and SO2 are global scale, while all other air quality forecast applications have a threshold coverage requirement which is limited to Europe and its coastal waters.

Ground-based networks can significantly add to the air quality forecasts, especially by adding information on the local scale. The measurements should preferably include O3, and H2O profiles from sonde measurements, as well as surface concentrations of O3, SO2, NO2, CO, CH4 and CH2O from a representative network. Information on surface CH4 concentrations is relevant, because CH4, although being relatively well-mixed, is an important competitor for the OH radical, and therefore variations in its abundance affects the lifetime of other compounds, especially CO.

Finally, near-real-time data delivery for this application implies that the data needs to be available to an operational modelling environment within a couple of hours after observation. In this case a significant part of today’s observations can still be used for the analysis on which the required forecast for tomorrow (etc.) will be based. It should be noted that current practice of data time handling at ECMWF is not favourable for Air

/166

S-4/-5 MRTD



Quality forecasts. Data are collected twice a day (till 3 am and 3 pm) to provide forecasts in the morning and evening. For Air Quality forecasts it would likely make sense to include also the late afternoon observations of today in the Air Quality forecast for tomorrow, which should be available to operational agencies in the very early morning of the day to come.

4.2.2.2.3 B2 - Auxiliary Data Requirements Air quality forecasting relies heavily on the combination of ground-based observations, air quality models and satellite data. The main auxiliary requirement is therefore on the availability of state-of-the-art air quality forecast model making use of analysis fields of NWP models, detailed emission databases (both natural and anthropogenic), adequate chemical schemes and detailed descriptions of surface-atmosphere exchange processes.

4.2.2.3 B3 - Assessment

4.2.2.3.1 B3 - Relevant Species and Processes In order to feed into environmental assessments and within the Air Quality theme the main drivers for understanding are the following long-term science questions:

What is the impact on air quality of the spatial and temporal variations and possible trends in the oxidising capacity?

What is the impact on air quality of spatial and temporal variations and possible trends in the long-range transport of longer-lived compounds and aerosols?

What is the impact on air quality of long-term changes in the distribution and total burden of the tropospheric ozone, carbon monoxide and methane background levels?

Can we relate the observed changes in atmospheric pollution levels to changes in certain emissions (source attribution)?

To answer these questions scientific users require long-term data sets of the total abundances and global concentration distribution of the pollutants as well as the detection and source attribution of the related emissions. For trend detection typically, monthly mean to annual values are needed in order to be able to relate changes in concentration levels to changes in emissions, possibly in response to policy measures.

The oxidising capacity of the atmosphere is largely governed by the OH and tropospheric ozone budget. Analysis of the causes for changes in the OH production and loss rates can be derived from simultaneous measurements of the global distribution (and spatial and temporal changes therein) of the longer-lived compounds in the OH budget, including H2O, O3, NOx, CO, CH4, CH2O and higher hydrocarbons, in combination with numerical modelling of chemistry, transport and mixing, emission and deposition, and UV-VIS radiative transfer for the photolysis rates.

Important for the tropospheric ozone budget are the mixing and transport processes, including stratosphere-troposphere exchange, ozone deposition, the ozone precursor gases (mainly NOx, CO and CH2O) and their chemistry, photolysis rates (mainly of NO2 and O3),

/166

S-4/-5 MRTD



water vapour and temperature. The trend of tropospheric ozone requires accurate monitoring of the tropospheric ozone profile.

Information on long-range transport is most important for CO, NOx, NOy, O3, and aerosols.

The trend of tropospheric ozone, carbon monoxide and methane requires accurate monitoring of the tropospheric ozone profiles and of CO and methane surface concentrations at background stations.

Inverse modelling will be used to derive emissions. Required emissions include aerosol emissions and aerosol and ozone precursor emissions including SO2, NO2 and CO.

4.2.2.3.2 B3 - Measurement Strategy and Data Requirements The goal coverage for understanding air quality issues should be global. The threshold coverage for the PBL can be Europe, incl. coastal waters, and for the free troposphere the threshold coverage includes at least parts of the North-Atlantic that impact on the surface air quality levels in Europe. The European scale mainly refers to use in scientific assessments by, e.g., the European Environmental Agency and understanding of European scale air quality issues.

To understand the oxidising capacity of the atmosphere related to air quality issues, measurements are needed of the global distribution (and spatial and temporal changes therein) of the longer-lived compounds in the OH and tropospheric ozone budgets, including H2O, O3, NOx, CO, CH4, CH2O and higher hydrocarbons, most notably isoprene and monoterpenes. Additional information would come from observations of the UV-VIS actinic flux, and N-reservoir species, especially at night, including HNO3, PAN, organic nitrates and N2O5.

The revisit time satellite data requirements are typically for daytime only. However, for example for O3, CO, and especially N2O5, HNO3 and PAN night-time observations would certainly be worthwhile. The N-compounds would give information on the night-time NOy budget, which is an important constraint on the amount of NOx released from reservoir species after sunrise. The threshold revisit time requirements of 2 hours are mainly related to the diurnal cycle of air pollution levels as well as the short timescales of the mixing and chemical processes in the PBL.

The understanding of long-range transport of pollutants requires global observations on: CO, NOx, NOy, O3, aerosol optical depth, aerosol type, POPs, and Hg. Distinction between PBL and free troposphere would be desirable, although the threshold requirement for the satellite observations related to long-range transport are on the tropospheric column (in combination with an averaging kernel). The assumption is that the height at which transport takes place can be traced from the model’s meteorological information.

The trend of tropospheric ozone and methane requires accurate monitoring of the tropospheric ozone profile and methane surface concentrations.

For source attribution the requirements are on aerosol observations and aerosol and ozone precursor observations, including SO2, NO2 and CO. Formaldehyde is required because it will contain important information to constrain the VOC emissions.

/166

S-4/-5 MRTD



The aerosol requirements on the satellite observations are on the aerosol optical depth and the aerosol type, with a goal to distinguish between aerosols in the PBL and free troposphere and a threshold for the total tropospheric aerosol optical depth. The required uncertainty (0.05) is expressed in absolute terms and based on different earlier assessments. The aerosol types to be distinguished include at least standard categories such as sulphate, dust, sea salt, organic carbon, black carbon, and mixed aerosol. The requirement on aerosol type is that mis-assignments should be limited to less than about 10% of the cases.

Note that satellite NO2 measurements are assumed to suffice to constrain NOx emissions and NOx ambient levels. This assumption sets some basic requirements on the chemical scheme that is to be used in the Air Quality model for the NO/NO2 conversions.

Separate measurements of the isotopes (12C, 13C, 14C) of C for CO (and possibly CH4) could be useful, both spaceborne and ground-based to distinguish between, e.g., fossil fuel and biomass burning emissions.

A representative ground network is needed for the validation of the Air Quality models and the spaceborne observations. Surface concentrations typically suffice. Additional measurements of PBL profiles (Lidars, Towers) at specific sites would be useful to validate PBL mixing processes in the models.

Spaceborne estimates of the spectral actinic flux profile, necessary to determine photodissociation rates, would be desirable, especially in combination with validation of the surface level actinic fluxes using a representative surface network of UV radiation measurements. Methods exist to translate spectral UV irradiance measurements into spectral actinic fluxes. The most relevant spectral range is the 280-420 nm spectral region as the most important photodissociation reactions are limited to this range. The required spectral resolution is typically ~5 nm.

4.2.2.3.3 B3 - Auxiliary Data Requirements Air quality forecasting heavily relies on the combination of ground-based observations, air quality models and satellite data. The main auxiliary requirement is therefore on the availability of state-of-the-art air quality forecast model making use of analysis fields of NWP models, detailed emission databases (both natural and anthropogenic), adequate chemical schemes and detailed descriptions of surface-atmosphere exchange processes.

4.2.3 Theme C: Climate

4.2.3.1 C1 - Protocol Monitoring and Treaty Verification

4.2.3.1.1 C1 - Relevant species and processes Within the climate theme the main drivers for protocol monitoring are the UNFCCC and the resulting Kyoto Protocol (for CO2, CH4, N2O, HFCs, PFCs and SF6), as well as complementary regulations related to EU climate policy (Climate Change Committee), see /UN1992/ and /UN1998/. The user requirements include the monitoring of the total abundances and global concentration distribution of the radiatively active gases and aerosols as well as the detection and source attribution of the related emissions. Typically,

/166

S-4/-5 MRTD



monthly mean values are needed. It would be highly desirable to be able to derive yearly trends in concentrations and emissions within a timeframe of a decade or less.

Several of the regulated greenhouse gases are ‘well-mixed’, i.e., their abundance in the troposphere and lower stratosphere is almost uniform over the globe. Good examples include the gases SF6, CF4, HFCs (HFC-134a is the most abundant), and CFCs (CFC-11 and CFC-12 are the most abundant). The atmospheric residence time of these gases is very long compared to the mixing time scales of the troposphere (typically in the order of months to one year). However, continuing but unevenly distributed emissions will maintain a latitudinal gradient and a global trend. Possible future changes in the zonal distribution of emissions, e.g. from mid-latitudes to (sub-)tropical latitude bands, may affect the latitudinal gradient. Inverse modelling can be used to trace the latitudinal concentration distribution of well-mixed gases back to latitudinal emission distributions. The applicability of inverse modelling for verification purposes was analysed recently in quite some detail in an inverse modelling workshop at Ispra /Ber2003/. HCFCs (of which HCFC-22 is the most abundant) are not inert in the troposphere. Therefore, the column data of these compounds will contain variability due to atmospheric transport, in addition to latitudinal gradients. Also the columns of N2O and CH4, and to a lesser extent, CFCs and CO2 will contain variability introduced by transport, mainly in the stratosphere. Clearly, dynamically-induced variabilities need to be corrected for before the column data of these gases can be used in addition to the surface measurements for the inverse modelling of emissions.

CO2 and CH4 are also often referred to as ‘well-mixed’, however these gases are not completely inert in the PBL and have large and variable natural sources and sinks, besides their anthropogenic emissions. For this reason the concentration distribution of these gases show more spatial and temporal variability in the troposphere. Especially for CO2 there is a strong diurnal cycle in the PBL, mainly due to the respiration and photosynthesis of the vegetation. Natural CH4 emissions (mainly from wetlands) are very uncertain, but the available observations also suggest large variability. Also anthropogenic CH4 emissions are assumed to be more variable than anthropogenic CO2 emissions because of their origins in, e.g., agriculture (rice paddies, ruminants), landfills, coal mining and related to fossil-fuel production.

Ozone and aerosols are relatively short-lived and show large variability in time and space throughout the atmosphere. For the ozone radiative forcing we should further make a distinction between tropospheric and stratospheric ozone as changes in their distribution and their trends have very different origins. Stratospheric ozone is expected to recover in the coming decades (see the Ozone Layer theme), in response to the measures taken on the emissions of halogenated compounds. Although a potent greenhouse gas, tropospheric ozone is nowadays mainly subject to air quality regulations (see the Air Quality theme). Ozone is not emitted but photochemically produced in the atmosphere. The two major precursor gases for tropospheric ozone are NO2 and CO (besides CH4 and non-methane hydrocarbons). It is anticipated that especially the NOx and CO emissions may become subject to regulation in the future if climate policy measures are to be taken to reduce the radiative forcing by tropospheric ozone. NO2 and CO are both short-lived and therefore show large variability throughout the troposphere.

/166

S-4/-5 MRTD



For the direct effect of tropospheric aerosols on climate the aerosol radiative properties are crucial, especially the aerosol extinction (‘cooling’) and aerosol absorption (‘warming’) optical depth. Large volcanic eruptions can inject large amounts of aerosol into the stratosphere, which can also have considerable climate effects over prolonged periods of time.

4.2.3.1.2 C1 - Measurement Strategy and Data Requirements A representative surface network with stations in different latitude bands separated by ~10 degrees latitude will be well suited to monitor (changes in) the latitudinal gradient and trend of well-mixed gases. The monitoring at a certain station should include a surface concentration, representative of the tropospheric background abundance in the latitude band, and a total column, representative of the total atmospheric abundance in the latitude band. The surface concentration observations will allow to derive information on (changing) zonal monthly emission distributions and yearly emission trends using inverse modelling. The total column measurements will confirm the representativeness of the surface observations. Weekly-representative observations will typically suffice to arrive at the required monthly means for concentrations and emissions. In order to be able to derive trends over a decadal time-frame the uncertainty on the individual observations should be very small. It is estimated that about 2% uncertainty for weekly-representative surface-based observations would typically suffice in this respect. Enhanced sampling, e.g. hourly or daily observations, can also help to reduce the uncertainties. The network should measure the regulated gases, including CO2, CH4, N2O, SF6, CF4, HFCs, and (H)CFCs. A high-density surface network is needed to derive emissions on a country-by-country basis, typically one station every 10000-100000 km2 and with at least one station per country. This would be very valuable. Sites should be close to emission regions for this purpose.

For CO2 and CH4, the global yearly trend in concentrations and emissions and the zonal distribution of the abundance and (monthly) emissions can be obtained from a representative surface network, as explained above. However, zonal distributions are of limited use for protocol verification. In order to better separate the variable natural emissions from the (more constant, although likely increasing) anthropogenic emissions, additional information on the spatial concentration and emission distribution may be derived from spaceborne observations. The same is true for the CO and NO2 concentrations and CO and NOx emissions. Although tropospheric profile information with global coverage will likely be optimal to constrain emissions, tropospheric columns or total column, in combination with an averaging kernel, with horizontal resolutions of 10x10 km2 (goal) to 50-50 km2 (threshold) are estimated to contain sufficient information to improve upon emission estimates from surface networks alone and especially help to improve emission estimates on country-by-country basis, as typically required for the protocols.

From available results on inverse modelling, the required uncertainty for spaceborne CO2, CH4, CO and NO2 column observations in order to be useful for improved emission estimates have been derived. The uncertainty of an individual CO2 column retrieval on the given horizontal resolution and with 6 to 12 hours revisit time (to capture the diurnal cycle) typically needs to be better than ~0.5% with sensitivity to the PBL. For the CH4 columns, on the same horizontal resolution but with only 1-day to 3-days revisit time (to capture the synoptic variability), the uncertainty of an individual retrieval needs to be better than ~2%

/166

S-4/-5 MRTD



with sensitivity to the PBL. For the much more variable CO columns, ~25% uncertainty would suffice, while for NO2 columns a maximum absolute uncertainty of ~1.31015 molecules cm-3 has been derived. The latter requirement in absolute terms implies that spaceborne observations of the variations in the background NO2 concentrations are not considered relevant.

By assimilation of sufficiently long and homogeneous time series, possible biases in the satellite columns can likely largely be accounted for by analysis of the observation minus forecast fields, especially in combination with the assimilation of the observations from surface networks. Further, if needed to reduce uncertainties, combinations of independent observations over a certain region and/or time period can be made to retrieve emissions over longer periods (e.g. months to years) and/or larger regional domains (e.g. continents). Crucial for the CO2, CH4, CO and NO2 column observations is the requirement for sensitivity to the PBL in order to be able to relate column variability with emissions. If the columns would reflect mainly the variability in the free troposphere, the inverse modelling is very much less constrained and emission estimates are likely limited to values representative of (very) large regions or hemispheres.

Tropospheric ozone, CO, NO2 and aerosols are short-lived and show variability in time and space to an extent that cannot be captured by surface-based networks or in-situ observations and thus their global distribution is best monitored by spaceborne observations. However, a distributed surface network is needed for the validation of the spaceborne measurements, either columns or profiles. Spaceborne tropospheric profiles should have at least ~5 km vertical resolution in order to contain at least two points outside the tropics and three points within in the tropical troposphere.

Monitoring of the height distribution of tropospheric aerosols from satellite is considered of minor relevance for climate monitoring, except to distinguish between tropospheric and stratospheric (volcanic) aerosols. For the inverse modelling of aerosol emissions the data requirements are comparable to those for NOx and CO emissions, i.e., total aerosol optical depth on similar horizontal resolutions and with a revisit time between 6 hours (goal) to 3-days (threshold). The shortest revisit time would be needed to include monitoring of dust storms with very-short lived large aerosol particles.

For the selection of ozone depleting halogen compounds the requirements in the tables are limited to the three Montreal gases that are responsible for the majority of climate forcing by halogenated compounds (CFC-11, CFC-12 and HCFC-22).

4.2.3.1.3 C1 - Auxiliary Requirements For long-term monitoring of the three-dimensional state of the atmospheric composition it is considered essential to assimilate the available observations in an atmospheric-chemistry numerical transport model in order to make optimal use of the available meteorological information. Furthermore, the (institutional) users will prefer complete, gridded and validated data sets with well-established uncertainties in terms of accuracy and possible biases. These requirements can be best fulfilled by an assimilation system, e.g. by systematic analysis of observation minus forecast error fields. Cross validation between different data sets will be facilitated by an assimilation system.

/166

S-4/-5 MRTD



In the case of using column observations to retrieve emissions the aid of a numerical transport model is also needed in order to be able to correct for dynamically-induced column variabilities that should not be related to emissions.

Another important requirement for inverse modelling of emissions is the availability of a priori emission distributions, both for the anthropogenic and the natural emissions. These inventories do exist and are widely available. Nevertheless, the spatio-temporal patterns of these inventories may still be very uncertain in many cases.

4.2.3.2 C2 - Near-Real Time Data Requirements

4.2.3.2.1 C2 - Relevant Species and Processes Climate monitoring relies to a large extent on the NWP centres, and especially on the re-analysis projects that these centres perform. For various reasons it would be impracticable if the assimilation of atmospheric composition data by NWP centres would be limited to reanalyses projects and would be excluded from the near-real time processing. NWP centres also do not have the resources to maintain different systems. Moreover, it also could lead to inconsistencies between different model versions. Therefore, in order to improve climate monitoring it is most advantageous to include atmospheric composition observations in near-real time in the operational assimilation system of the NWP centres.

Driving the near-real time data requirements for the climate theme is therefore the assimilation of atmospheric composition observations in NWP models in order to improve the analysis of the physically coupled climate system. Depending on the improvement of the analyses also improvement of the weather forecasts can be envisioned, although atmospheric composition typically impacts the atmosphere most on the longer, climatic time scales.

In addition to climate monitoring, a service to make near-real-time data sets quickly available to NWP and climate research centres will allow for a continuous process of validation of the latest NWP and climate models for present-day atmospheric conditions. Near-real-time validation of adjustments in NWP models is crucial to the NWP centres. Also the capability of a climate model to simulate the latest changes in the atmospheric state is generally considered as an important model requirement to gain confidence in its ability to simulate future climate change.

In order to justify the efforts, it is required that atmospheric composition data that are intended for assimilation in NWP models should have a non-negligible impact on the model simulations. Here, two types of contributions can be distinguished: a direct impact of the atmospheric composition observations on the physical climate system, e.g., stratospheric ozone largely determines stratospheric heating rates; and an indirect impact by improving the application of other available observations. One example of the latter effect is the impact of atmospheric composition on model-simulated radiances, e.g. to constrain the temperature profile retrieval or the outgoing long-wave radiation at the top of the atmosphere.

The most important chemical species for NWP is water vapour. Water vapour plays a central role in the atmosphere, e.g. in the atmospheric radiation and energy budgets, in the hydrological land-ocean-atmosphere system and in several parameterisations such as for

/166

S-4/-5 MRTD



convection and precipitation formation. Accurate profiles of water vapour are needed in NWP models, throughout the troposphere and also in the lower to middle stratosphere.

In the PBL atmospheric composition impacts on the atmospheric absorption with the largest contributions coming from aerosol absorption, water vapour, CO2, and ozone. Also the scattering of solar radiation by aerosol particles has significant effect on how the physical climate manifests, e.g. on surface temperature and incident solar radiation. Aerosols also impact on several other remote sensing observations and improved characterisation in NWP models will reduce uncertainties related to aerosol correction.

In the free troposphere the same components as in the PBL are relevant for NWP, although the effect of the spatial and temporal variability in CO2 is probably negligible in the free troposphere and only long-term trend monitoring is required.

In the upper troposphere and lower stratosphere water vapour, (ice) particles including cirrus and PSCs, and ozone impact on the physical climate. Observations of ozone, water vapour, CO2, CH4, and N2O throughout the stratosphere are important for the radiation budget. The assimilation of radiatively active gases will improve the simulation of the local heating rates and outgoing long-wave radiation at the top of the atmosphere.

Observations of inert stratospheric tracers, e.g. SF6 or HF, but also other tracers including CO2, CH4, N2O, HDO, will help to better constrain the large-scale transport in the stratosphere. These observations will be complementary to direct observations of the wind vector, planned by, e.g., the ESA Atmospheric Dynamics Mission (ADM) Aeolus. Direct observations of the wind vector observations will constrain in the first place the dominant large-scale motions that are most relevant on the short-term to NWP. In addition, tracer observations will help to better constrain the residual Brewer-Dobson circulation and associated vertical and lateral motions. Tracers represent air masses and have a memory of the flow over the preceding time.

Although tracer information would be most profitable on longer, seasonal and climatic time scales, it is hypothesised that sufficiently accurate inert tracer profile measurements with the given goal revisit time may also positively impact on the stratospheric dynamics on short time scales. However, because absolute tracer concentrations are being measured and mixing ratios are conserved during transport this would possibly also require accurate information on the atmospheric density profile in the stratosphere as well as information on gravity waves. At this stage it is rather uncertain what could be the impact of tracer observations for NWP on short time scales relevant to weather prediction.

It is noted that stratospheric observations of the tracers CH4 and HDO, in addition to H2O, can help to better constrain the stratospheric water vapour budget.

4.2.3.2.2 C2 - Measurement Strategy and Data Requirements For NWP and climate monitoring applications the three-dimensional water vapour distribution in the PBL, free troposphere and stratosphere is required with global coverage. Therefore, an integrated approach of spaceborne observations, a representative global in-situ network of radiosonde and surface-based remote-sensing techniques is needed, coupled with model information. Two-to-three kilometre vertical resolution for H2O would be very advantageous, threshold for the satellite contribution is the distinction of PBL, free

/166

S-4/-5 MRTD



tropospheric and stratospheric water vapour sub-columns. For climate purposes the goal horizontal resolution in the troposphere is about 10x10 km2, although water vapour spatial variability is large and structures with less than one kilometre are associated with e.g. fronts. Uncertainty of column data typically needs to be better than ~5% to improve upon current modelling capabilities of weather centres. Tropospheric water vapour has a strong diurnal cycle and the required revisit time for spaceborne observations is typically ~6 hours. The revisit time can be limited to one day to one week (threshold) in the stratosphere. Given the spatio-temporal variability in water vapour the optimal strategy to water vapour is likely combined use of ground-based systems (e.g. GPS), radiosondes, polar orbiting and geostationary platforms.

Aerosol absorption and aerosol scattering are important for the radiation budget and atmospheric corrections. Threshold requirements for operational use include the separation of the total extinction optical depth in an absorption and scattering contribution. Distinction between boundary-layer, free-tropospheric, and stratospheric aerosol would be advantageous, as well as further aerosol characterisation, in particular the aerosol phase function given the important radiative effects of aerosols. The same set of requirements applies to cirrus and PSC ice particles (optical depth, phase function) albeit limited to the higher altitudes. Spatial scales for aerosol are typically comparable to water vapour. Revisit times for tropospheric aerosols can typically be limited to about one (goal) to a few days (threshold) and to a couple of days to a week in the stratosphere.

Ozone profile information is most relevant in the stratosphere and upper troposphere where the (variation in) ozone radiative forcing is most effective. Tropospheric ozone threshold requirements are limited to column observations (in combination with an averaging kernel), while distinction between the PBL and free troposphere would be advantageous. In the UTLS region, co-located profile observations of O3 with HNO3, HCl and/or CO are desirable to help to constrain stratosphere-troposphere exchange processes. Hereto, the observations need to be both rather accurate and have high vertical resolution (1 km goal, 3 km threshold).

In-situ CO2 observations in the PBL and total column CO2 observations can be obtained from a surface network. Spaceborne CO2 column observations (in combination with an averaging kernel) can help to provide global coverage provided they are sufficiently accurate to include the natural variations in CO2. The column data need to be sensitive to the PBL. A representative surface network would be needed for validation and corrections of possible biases.

As explained in the former section, tracers can constrain the stratospheric circulation. Suitable candidates are inert gases as SF6 and HF, but other long-lived compounds such as CO2, N2O, CH4 and HDO can be used as well. The tracer that can be observed most accurately needs to be measured. The required uncertainty is directly related to the gradient over the specified spatial resolution (100–200 km horizontally, 1-3 km vertically). Goal revisit times are about 12 hours. With the threshold revisit time of one week, only information on the circulation on seasonal to multi-annual time scales will be obtainable.

For the radiation budget, stratospheric profiles are required for the radiatively active gases H2O, O3, CO2, CH4 and N2O. The stratospheric water vapour budget can be constrained by

/166

S-4/-5 MRTD



measurements of H2O, HDO and CH4. Profiles are needed with three-to-five kilometre vertical resolution throughout the stratosphere.

Finally, near real time data delivery for this application implies that the data needs to be available to an operational modelling environment within a couple of hours after observation. In that case a significant part of today’s observations can still be used for the analysis of the day.

4.2.3.2.3 C2 - Auxiliary Data Requirements The main users for near-real time data within the climate theme are the NWP centres. These centres need near-real-time information on numerous aspects of the land-atmosphere-ocean-cryosphere system that all contribute to the analysis of the atmosphere and therefore to the initial state on which the weather prediction is based, and on which climate monitoring relies. Atmospheric composition is one of the key elements for the monitoring of the climate system.

4.2.3.3 C3 - Assessment

4.2.3.3.1 C3 - Relevant Species and Processes Within the Climate theme the operational data requirements for understanding need to be based on long-term science questions relevant to understand the interactions between atmospheric composition and the physical climate. The relevant issues are typically addressed in the regular IPCC scientific assessments.

Important science questions that require long-term operational monitoring are related to:

Understanding of the radiative forcing of climate and the changes in forcing over time, including possible volcanic eruptions, and also including the forcing of climate on local to regional scales

Understanding of the abundance, evolution, and, if relevant, spatial distribution of the forcing agents

Understanding of the stratospheric water vapour budget and the monitoring of the water vapour trend in the UTLS and above.

Understanding of the role of the ozone layer evolution on climate change

Understanding of the role of possible changes in the Brewer-Dobson circulation on climate change, including possible changes in the position and strength of the polar and sub-tropical jets, changes in the position and strength of the inter-tropical conversion zone (ITCZ) as well as changes in the mesosphere (air density)

Understanding of the role of long-term changes in the oxidising capacity of the troposphere for its effect on the atmospheric residence time of the climate gases

Concentration monitoring for the detection and attribution of long-term changes in the natural as well as anthropogenic emissions of the forcing agents and their precursors.

/166

S-4/-5 MRTD



Data requirements, related to the understanding of the role of atmospheric composition changes for climate, have been laid down in the Atmospheric Composition Explorer for CHEMistry and climate interaction (ACECHEM) Report for Assessment /ESA2001/ and the report of the preceding Atmospheric Chemistry Explorer (ACE) requirements study /Ker2001/]. The reader is referred to these documents for additional scientific background.

4.2.3.3.2 C3 - Measurement Strategy and Data Requirements The processes underlying the interactions between climate change and atmospheric composition change are typically rather slow (months, years, decades) and therefore can only be better understood by increasing the amount of available long-term and homogeneous data sets on atmospheric composition. Although global coverage is required for most observations, the UTLS layer it probably the most important atmospheric domain because it is both chemically and radiatively very active. However, other atmospheric layers are relevant as well, e.g. the long-term trend in stratospheric water vapour is badly understood and this needs to be monitored by long-term accurate global-scale profile measurements including the stratosphere above the UTLS layer. Profiles of H2O, HDO and CH4 are needed with three-to-five kilometre vertical resolution. Column data can be useful and should be given in combination with an averaging kernel. For the radiation budget vertical profiles are required for the radiatively active gases H2O, O3, CO2, CH4 and N2O in both the UTLS and the overlying stratosphere. Tracer measurements to constrain the Brewer-Dobson circulation also need to extend over the full stratosphere, and possibly should even include parts of the mesosphere. Changes in the mesosphere, e.g., in air density, could give also indication of temperature changes in the middle atmosphere. Suitable tracer candidates for diagnosing the Brewer-Dobson circulation are typically inert gases such as SF6 and HF, but other long-lived compounds such as CO2, N2O, CH4 and HDO can be useful as well. Likely the tracers that can be observed most accurately need to prevail.

Atmospheric composition related climate processes in the troposphere include, e.g., gaseous and aerosol absorption and aerosol scattering in the PBL, secondary aerosol formation relevant for cloud formation, aerosol deposition on ice surfaces affecting the ice surface albedo and oceanic dimethylsulfide also affecting cloud condensation nucleii.

In-situ observations in the UTLS by operational aircraft measurements will be useful in addition to satellite measurements and should include preferably O3, CO, NOy (or HNO3), NOx, HCl and H2O. The airborne measurements can especially help to better constrain stratosphere-troposphere exchange as well as chemical processes

Surface-based atmospheric composition measurements contributing to understanding of climate are most relevant to monitor the long-term evolution in the long-lived gases. In addition the networks are crucial for the validation of the global-scale satellite measurements. The need for a detailed knowledge on the 3-D water vapour distribution and its changes over time would be improved by surface-based networks such as the radiosonde network and GPS-based configurations. The monitoring of the 3-D distribution of ozone and aerosols would be improved by surface based monitoring of surface concentrations and total columns as well as a network of profile measurements of sondes and LIDARs.

/166

S-4/-5 MRTD



4.2.3.3.3 C3 - Auxiliary Data Requirements Climate research centres need long-term information on numerous aspects of the land-atmosphere-ocean-cryosphere system that all contribute to the analysis of the climate system. Atmospheric composition is only one component of the Earth System. The usefulness of atmospheric composition data for the study on climate change will partly depend on the information that will be available for the other components of the Earth System.

4.3 Application Priorities and Contributions of Existing and Planned Missions

Of the three types of applications considered in the previous section, the protocol monitoring and near-real-time services account fully for the operational character of the Sentinel-4/-5 mission and therefore have the highest priority. Climate monitoring is included in the “near-real-time” service category; there is general agreement that climate monitoring requirements are not more stringent than requirements for short-term monitoring and forecast. The third category, environmental assessment, covers important aspects such as support to IPCC assessment; however in this case there is no clear user relationship. Requirements related to assessment applications are therefore of lower priority in the context of a Copernicus mission and are not supposed to drive the mission concept.

An extensive survey of the potential contributions of other existing and planned space missions to the fulfilment of the mission requirements has been carried out /Kel2005/. As a result, in the post-Envisat timeframe two families of satellites are expected to be operational. The European MetOp / EPS mission carries the UV-VIS-NIR sounder “GOME-2” and the FTIR “IASI” (both in nadir-viewing geometry), NASA’s S-NPP and NOAA’s JPSS mission have the UV-VIS sounder “OMPS” (limb and nadir) and the IR sounder “CrIS” (nadir) on board. The preliminary assumption is that European users will have access to JPSS data, although this will need to be verified.

UV-VIS observations will be available once per day from each of the instruments (MetOp/GOME-2, S-NPP&JPSS/OMPS) and are expected to cover requirements of application A1 (protocol monitoring and treaty verification applications for the stratospheric ozone and surface UV theme). They will provide a small contribution to other stratospheric applications, however limited by their poor vertical resolution and only partial coverage of required species. Their contribution to tropospheric applications is considered limited since, due to their large ground pixels, most of the tropospheric measurements are contaminated by clouds and the horizontal resolutions are too coarse; and in the case of OMPS the specifications on spectral coverage and resolution and SNR will limit its use to stratospheric applications.

IR observations will be made twice per day by each of the instruments (IASI and CrIS). These will make a contribution to tropospheric applications, in particular as they have the potential to provide some altitude resolution for a subset of the needed species. However, these measurements are not suited to obtain the crucial information on PBL composition, and their spectral resolution limits the profiling capabilities (in particular CrIS).

/166

S-4/-5 MRTD



There are four overall requirements that cannot be met by the existing or planned systems and therefore need to be addressed by future missions:

High temporal and spatial resolution space-based measurements of tropospheric composition including the PBL for Air Quality applications (B1, B2, B3);

Vertical distribution of aerosol needs to be characterized in order to estimate near surface concentrations of particulate matter (B1). Note that requirements regarding column integrated aerosol properties (B1) are covered, namely by Copernicus Sentinel-3 with high spatial resolution (0.25-1.0 km);

High spatial resolution and high precision monitoring of tropospheric climate gases (CH4, CO (precursor) and CO2) and aerosols with sensitivity to PBL concentrations (C1);

High vertical resolution measurements in the upper troposphere/lower stratosphere region for Stratospheric Ozone/Surface UV and Climate near-real time and assessment applications (A2, A3, C2, C3).

/166

S-4/-5 MRTD



5 SYSTEM CONCEPTS

In this chapter, possible measurement techniques and mission scenarios to satisfy the geophysical observation requirements are identified.

5.1 Candidate Remote Sensing Techniques and Mission Scenarios

Possible remote sensing principles and specifications have been investigated in /Kel2005/ which includes an extensive assessment of the capabilities of spaceborne atmospheric chemistry instrumentation (existing and proposed for Sentinels 4 and 5) with respect to the observational requirements in Chapter 6 and 7 of this document, a prioritisation of observational capabilities per application as well as further references justifying individual specifications for suggested space instruments. Following the results of this study, potential implementation of the required observations can be identified as described below.

5.1.1 High temporal and spatial resolution space-based measurements of tropospheric composition, including the planetary boundary layer (PBL), for air quality applications

Suitable instrument types with sensitivity to required chemical constituents are

a UV-VIS-NIR (UVN) spectrometer to cover the most important species listed in Section 4.2.2 and to provide PBL sensitivity (daytime);

a SWIR spectrometer for the CO and CH4 total columns (bands near 2.3 m, 1.6 m) with PBL sensitivity (daytime); and possibly another SWIR channel for aerosol (daytime);

a thermal IR (TIR) spectrometer for CO and O3 vertical profiles, possibly complementary species and day and night coverage of species.

The mission scenario may depend on the key observational requirements for this mission which are the temporal sampling of 0.5 – 2 h and horizontal sampling of 5 – 20 km. Given

the uncertainty of the temporal sampling requirement in the absence of any spaceborne atmospheric composition data coming even near to it and of suitable observation system simulation experiments,

the fact that in another assessment of geophysical observation requirements for the same application, a team tasked by Eumetsat /EUM2006/ arrived at a threshold temporal sampling requirement of 4 hours,

the fact that the trade-off between temporal sampling and geographical coverage depends on the setting of priorities among various air quality applications,

/166

S-4/-5 MRTD



a range of combined temporal sampling and geographical coverage requirements can be identified as specified below. The choices made from these options are described further in this section.

temporal sampling of 1 hour covering Europe and surrounding areas (30°W - 45°E [@40°N] and 25/30ºN (goal / threshold) - 65°N, nominal geographical coverage area) and daily global coverage at fixed local time;

temporal sampling of 2 hours covering the most polluted latitude ranges of 25/30ºN (goal / threshold) to 65/60ºN (goal / threshold) at all longitudes and temporal resolution of 6 hours at lower latitudes; no sampling of high latitudes required;

temporal sampling of 4 hours globally.

The general goal requirement on sampling is (near-)contiguous sampling. The integration of the space-borne data within a data assimilation framework may allow for data gaps in time and space to a certain extent.

5.1.2 High spatial resolution and high precision monitoring of tropospheric climate gases (CH4, CO (precursor) and CO2) and aerosols with sensitivity to PBL concentrations for climate protocol monitoring

monitoring CO2 with the accuracy necessary to serve operational applications such as treaty verification is very challenging due to severe random and systematic error requirements, uncertainties in inverse modelling (e.g. transport between PBL and free troposphere) and uncertainties in the modelling of natural surface fluxes (an order of magnitude larger than the anthropogenic fluxes). This application is therefore considered immature at this time /Bre2003/, /Kel2005/, but may be re-considered once experience with upcoming research missions is available. In addition, /Hun2010/ concludes that separation of natural and anthropogenic components from space might be very challenging.

CO and aerosols can be observed in the same way as for the air quality mission; CH4

with PBL sensitivity can be measured in the 1.6 m band and also in the 2.3 m band as CO.

This mission requires global coverage, with emphasis on inhabited areas. Due to the large overlap of requirements, it can best be combined with the LEO (part of the) air quality mission, be it in sun-synchronous or with lower orbital inclination.

5.1.3 High vertical resolution measurements in the upper troposphere/lower stratosphere region for stratospheric ozone/surface UV and climate near-real time and assessment applications

These data will best be obtained by a limb-sounding technique in either the thermal IR or in the mm-wave region. However, scientific prototypes of both sensing techniques on Envisat and EOS-Aura are just being assessed for their operational capabilities; the most suitable technique cannot be identified yet.

/166

S-4/-5 MRTD



5.2 Overall System

In conclusion, the following implementation priorities are recommended:

1. a satellite with UV-VIS-NIR (UVN), SWIR and TIR observational capabilities, also using auxiliary cloud and aerosol data, which serves air quality and climate protocol monitoring in LEO. This satellite complies with all temporal sampling / geographical coverage requirement scenarios and will provide continuity and improvement with respect to the OMI and Sciamachy missions. From the considered orbit scenario’s, namely sun-synchronous or lower inclination, the sun-synchronous option was chosen. The specified mission scenario is described in Section 5.2.1;

2. an extension of this mission to obtain regularly ≤ 1 hour revisit time as required for air quality applications. This extension could consist of a GEO platform carrying instrumentation with similar Level-1b performance specifications. It was decided to implement this mission, in addition to one LEO platform as specified under item 1, as described in Section 5.2.1;

3. a limb-sounding mission observing the UTLS either in the mm-wave or infrared region to serve stratospheric ozone / surface UV and climate near real time and assessment applications. This mission will not be part of the Sentinel-4/-5 system and will not be further considered in this document;

4. requirements regarding aerosol requirements shall be addressed via a multi-viewing, multi-channel, multi-polarisation capability. Column integrated aerosol properties will be covered by Copernicus Sentinel-3.

MR-SYSTEM-10. Observations in the solar spectral range shall be made whenever the solar zenith angle (SZA) at the observed ground pixels is less than 92°; observations in the thermal spectral range shall be made continuously.

MR-SYSTEM-20. The geographical coverage and revisit time of the Sentinel-4/-5 system shall satisfy the requirements listed in Table 5.1. For observations in the solar spectral range, these requirements shall be met for SZA < 80°. For LEO, global coverage within one day is required. For GEO, an area over Europe and a revisit time are specified in Table 5.1 and MR-GEO-SYS-10, MR-GEO-SYS-20, MR-GEO-SYS-30.

/166

S-4/-5 MRTD



Table 5.1 Coverage and revisit time requirements.

Latitude range Longitude range Revisit time [hours]

30oN – 65oN (3)

30oW – 45oE (@ 40oN) (3)

0.5 – 1 (for solar bands) (2)

0.5 – 2 (for TIR bands) (2)

90oS – 90oN All (1) 24 (@ fixed local time)

(1) Minor (drifting) gaps in the geographical and temporal coverage are tolerated, especially at latitudes below ±20o between consecutive orbit tracks.

(2) Highest temporal resolution needed for solar bands due to their sensitivity to the PBL. TIR bands are sensitive to the free troposphere with less temporal variability, allowing a relaxed revisit time requirement.

(3) Indication only; for details see MR-GEO-SYS-10, MR-GEO-SYS-20, MR-GEO-SYS-30. MR-SYSTEM-30. The nominal operational lifetime of the Sentinel-4/-5 system shall be

at least 15 years.

MR-SYSTEM-40. Level-1b data contain radiometrically and spectrally calibrated and geo-located radiance and irradiance data.

MR-SYSTEM-50. The ground segment of each mission shall allow reprocessing of the respective Level-1b and Level-2 data on a yearly basis.

The following chapters provide the Level-1b requirements for Sentinel-4, Sentinel-5, and Sentinel-5 Precursor.

5.2.1 Mission Scenario

Following recommendations expressed in /GAS2009/ and based on high level agreements, Sentinels 4 and 5 will get implemented as additional payloads on Eumetsat platforms, as follows.

Sentinel-4 will be a realised as

o addition of a UVN spectrometer on the MTG-S platforms;

o utilisation of TIR data from the MTG-S InfraRed Sounder (IRS); and

o utilisation of imager data from the MTG-I platforms.

Sentinel-5 will be a realised as

o a UVNS spectrometer embarked on the EPS-SG platforms;

o utilisation of TIR data from the EPS-SG Infrared Atmospheric Sounder (IAS);

o utilisation of imager data from the EPS-SG VIS/IR Imager (VII); and

o utilisation of EPS-SG Multi-Viewing Multi-Channel Multi-Polarisation Imager (3MI).

The expected launch dates for MTG-S (≥2021) and EPS-SG (≥2021) imply a data gap following the end of life of the Envisat mission (end of operations in 2012, including SCIAMACHY) and the EOS-Aura mission (launched in 2004, including OMI and TES), affecting in particular short-wave measurements with sufficient quality for tropospheric

/166

S-4/-5 MRTD



applications. Hence, the need for an additional mission was identified, as described in Section 5.3

5.3 The Need for a Sentinel-5 Precursor (S5P) Mission

A precursor for Sentinel-5 is needed to bridge between the current research missions and Sentinel-5, for several reasons:

Continuity of data:

o CO and CH4 data with the essential PBL sensitivity are needed for air quality and climate applications. Without a Sentinel-5 Precursor, provision of such data has stopped with the end of life of Sciamachy in 2012. GOME-2 on MetOp does not have the required short-wave infrared band; IASI measurements of CO and CH4 have little sensitivity to the PBL;

o High spatial resolution is mandatory for tropospheric applications to resolve emission sources and to obtain an acceptable fraction of spectra without cloud contamination. After the end of life of OMI on EOS-Aura, there will be no other UV-VIS spectrometer with comparable resolution;

o The pixel size of GOME-2 in the standard observation scenario implies a tropospheric sampling density of about one strictly cloud-free sample per geolocation per week. This is not expected to be sufficient for operational tropospheric applications, since sampling frequency and spatial resolution are still unacceptable;

o Daily global coverage will end after the end of life of OMI, due to the narrower swath of GOME-2.

The Sentinel-5 Precursor (S5P) will fly in an early afternoon orbit (as recommended in /Lev2009/) in which it will pick up more pollution signal. This is essential for the predictive value of the data. GOME-2 data at a local time of 9:30 do not contain the information needed for air quality forecast of the next day.

The improved radiometric sensitivity of S5P will allow to make measurements at low albedo, thereby helping to identify smaller pollution events and to improve the accuracy of air quality assessments and forecast and of CH4 surface fluxes.

S5P data can be combined with IASI data and with GOME-2 (to the extent that GOME-2 can be exploited for tropospheric applications). This will help building up synergetic data use and for the first time provide better than daily coverage. (Some trials are done now with OMI and Sciamachy, however limited by Sciamachy revisit time of 6 days)

MR-SYSTEM-60. A Sentinel-5 Precursor mission is required to bridge the gap between contemporary missions and EPS-SG, and to provide a smooth transition to operational applications. The operational lifetime of the precursor mission shall extend from 2017 to 2024.

/166

S-4/-5 MRTD



5.4 Requirement Terms, Conventions and Definitions

Requirements are labelled according to the structure

MR-x-y-n where

x is either the overall Sentinel-4, -5, and -5P SYSTEM, or the LEO, GEO, or S5P component,

y is omitted if x is the SYSTEM on platform level, otherwise it denotes an observation concept on the platform such as UVN, SWIR, TIR or 3MI, and

n is a number.

In Appendix 1 the terms, conventions and definitions used for the requirements are described. For clarity, a number of conventions and definitions is summarized here:

Requirements apply to Level-1B data, or if explicitly stated, to derived quantities like the reflectance (which is defined in Appendix 1).

Error distributions are to be understood as Gaussian and the threshold values given refer to one standard deviation (1), unless otherwise stated.

A recipe for the combination of error contributions to overall errors is given in the LEO and GEO chapters.

Level-1B requirements are generally derived from a combination of the relevant Level-2 requirements; these are intended to cover a typical range of geophysical scenarios, but usually not all extremes.

A number of requirements have separate numbers for “goal”, “breakthrough”, and “threshold”, which are defined in Appendix 1. The implementation or not of the goal requirements will be decided by the Agency after proper analysis of the implications.

/166

S-4/-5 MRTD



6 LEVEL-1B PRODUCT REQUIREMENTS FOR LEO MISSION (SENTINEL-5 AND SENTINEL-5 PRECURSOR)

The LEO part of the Sentinel-4/-5 system, which is implemented as the Sentinel-5 and Sentinel-5 Precursor (S5P) missions, shall have observational capabilities in the UVN, SWIR and TIR. There is also the need for auxiliary cloud imaging data as input to the different retrieval algorithms. In addition, there is a need for aerosol data for air quality purposes, and as input to retrievals from the other instruments. For the latter, in the case of Sentinel-5 the data from 3MI are assumed to be available.

In Sections 6.1 – 6.6, the Level-1b requirements for each of the spectral ranges are outlined (for Sentinel-5 and S5P together).

6.1 LEO-SYSTEM

MR-LEO-SYS-5. During the nominal operational phase (i.e. after commissioning), the system shall provide an availability of the Level-1b data of at least 95%.

For a proper allocation of sources and sinks, accurate geolocation knowledge of the observations is required.

MR-LEO-SYS-10. The geo-location knowledge shall be better than 0.1 spatial sampling distance (SSD) for all spatial samples.

Synergetic retrievals are envisaged from various bands:

ozone from combination of UV and TIR bands,

CO from combination of the SWIR and the TIR bands,

CH4 from combination of the SWIR and the TIR bands,

cloud information from NIR (i.e., O2-A) band used in UV-VIS and SWIR retrievals,

cloud information from VII used in UV-VIS, SWIR and TIR retrievals,

refined cloud characterisation from the combination of all instruments,

aerosol information from UVN and 3MI, also used in trace gas retrievals in UV-VIS

Geometrical Requirements

Requirements MR-LEO-SYS-20 through MR-LEO-SYS-110 shall hold for all spatial samples.

MR-LEO-SYS-15. The SSD at SSP in the UV-VIS-NIR and SWIR shall be smaller than or equal to 15 km (T) / 7 km (B) / 5 km (G) for λ>300 nm, except for band UV-1 (see Table 6.2) where it can be smaller than or equal to 50 km (T) / 15 km (G).

/166

S-4/-5 MRTD



MR-LEO-SYS-20. The SSD shall be identical within 5% in all spectral channels of the UV-2 (for definition of UV-1 and UV-2 see Table 6.2), VIS, NIR, and SWIR. The SSD in channels in the UV-1 shall be identical to an integer multiple of the SSD in channels of the other bands, within 5%.

The sampling shall be such that the centroids of the UV-1 grid (Fig 6.1, red) match the centroids of a grid obtained by combining an inter number of pixels from eg the UV-2 (Fig. 6.1, green).

Figure 6.1. Alignment of regular sampling grids with SSD ratio (SSD red/SSD green) = 2.

With the breakthrough spatial sampling, observations with an SSD equal or better than the nominal SSD at SSP are achieved with daily global coverage up to the mid-latitudes.

MR-LEO-SYS-35. The SSD in across-track direction shall be equal to or better than the nominal SSD at SSP for all viewing angles (goal). As breakthrough, the off-nadir SSD in across-track direction between spatial samples up to 1000 km away from the SSP shall be equal or better than the nominal SSD at SSP. As threshold, the off-nadir SSD in across-track direction shall be obtained by maintaining a regular angular sampling of SSA.

It is desired that data with a spatial sampling better than what is defined by MR-LEO-SYS-35 are down-linked and that spatial binning is performed on-ground. Moreover, it is preferred that on-ground spatial binning is performed in a flexible programmable way thus allowing observations with enhanced spatial sampling but degraded SNR in the case of emission events. All requirements that are influenced by spatial sampling (e.g. MR-LEO-SYS-10, 40, -60, -61, -62, -63, -64, -65, MR-LEO-UVN-150, -160, -180, -185, MR-LEO-SWIR-70, -150, -160, -185) apply on the achieved spatial sampling.

MR-LEO-SYS-38. The SSD in along-track direction at off-nadir spatial samples shall be identical to the SSD at SSP in along-track direction.

MR-LEO-SYS-39. Radiance data with an along-track SSD smaller than 0.2 (G) / 0.33 (T) times the nominal along-track SSD shall be available on-ground for the spectral samples located in the (TBS) spectral ranges (e.g. 310-315 nm, 338-343 nm, 415-420 nm, 460-465 nm, 750-755 nm, 760-763 nm, 770-775 nm, and 2310-2315 nm). Note that all other geometric, spectral, and radiometric requirements apply to the nominal along-track SSD defined by MR-LEO-SYS-015 and MR-LEO-SYS-36.

/166

S-4/-5 MRTD



Intra- and inter-band co-registration

The synergetic retrievals require observations of a consistent scene resulting in requirements for inter-band spatial co-registration and requirements on sampling with the same spatial resolution. The retrieval of data within a spectral band also requires a good spatial co-registration, which in this case results in an intra-band spatial co-registration requirement. Spatial co-registration requirements are specified relative to the achieved SSD.

MR-LEO-SYS-40. Intra-band spatial co-registration shall be better than the values specified in Table 6.1.

MR-LEO-SYS-60. Inter-band spatial co-registration between bands shall be better than the values specified in Table 6.1. In case of different SSDs, the larger one applies. More specifically, the co-registration with UV-1 shall be done with the center of the composite grid (red in Fig 6.1) of smaller SSDs fitting in the UV-1 SSD (green in Fig 6.1).

Table 6.1 Intra-band spatial co-registration (values on the diagonal) and inter-band spatial co-registration (off-diagonal values) expressed in SSD.

Band UV-1 UV-2 Vis NIR-1 NIR-2 SWIR-1 SWIR-3 UV-1 0.1 0.2 - - 0.2 - -

UV-2 0.15 0.2 - 0.3 - -

Vis 0.15 - 0.3 - -

NIR-1 0.1 0.2 - -

NIR-2 0.1 0.3 0.3

SWIR-1 0.15 0.3

SWIR-3 0.15

MR-LEO-SYS-61. The nadir System Integrated Energy (SIE) shall be between 66% and 83% over an area of dimension 1.0 SSD along

track by 1.0 SSD across track larger than 90% over an area of dimension 1.5 SSD along track by

1.5 SSD across track.

MR-LEO-SYS-62. At nadir, the SIE over an area of dimension 1.0 SSD along track by 1.0 SSD across track shall vary by less than 10 percent points over the spectral range containing the UV-2, Vis, NIR-1, and the NIR-2.

MR-LEO-SYS-63. At nadir, the SIE over an area of dimension 1.0 SSD along track by 1.0 SSD across track shall vary by less than 10 percent points over the spectral range containing the NIR-2, SWIR-1, and SWIR-3.

MR-LEO-SYS-64. For any spatial sample, the Instrument Integrated Energy (IIE) over an area of dimension 1.0 SSD along track by 1.0 SSD across track shall vary by less than 10 percent points over the spectral range containing the UV, Vis, NIR-1, and the NIR-2.

/166

S-4/-5 MRTD



MR-LEO-SYS-65. For any spatial sample, the IIE over an area of dimension 1.0 SSD along track by 1.0 SSD across track shall vary by less than 10 percent points over the spectral range containing the NIR-2, SWIR-1, and SWIR-3.

For SWIR retrievals, a selection of the best pixels (e.g., very low cloud fraction) will be made, which will be below a TBD observation viewing angle.

MR-LEO-SYS-67. The temporal co-registration of measurements of the same spatial sample between any pair of spectral channels within the UVNS shall be better than 5 seconds.

Inter-instrument co-registration

For consistent sampling of the atmosphere, measurements of a scene by different bands / instruments need to be made within a time shorter than atmospheric variability, in particular cloud cover (i.e., between TIR and VII) and with sufficient co-registration knowledge. Hence it is recommended that

temporal co-registration of UVN or SWIR measurements of a given spatial sample and TIR or VII measurements of the same spatial sample is better than 10/60 seconds (G/T);

inter-band spatial co-registration between the VII and any of the other instrument data is known to better than 1.0 (T) / 0.5 (G) of the VII SSD.

inter-band spatial co-registration between any band in the TIR and any band in the UV-1, UV-2, VIS, NIR, or SWIR is known to better than 0.2 (T) / 0.1 (G) of the SSD of the UV-1, UV-2, VIS, NIR, or SWIR respectively.

Data delivery time:

Fast delivery services are required for air quality forecast applications (see also Sections 3.2.4 and 3.3.3).

MR-LEO-SYS-120. Level-1b data with full spectral coverage shall be available 2h 15’ after sensing and Level-2 data shall be available 2h 45’ after sensing, with full spatial coverage.

MR-LEO-SYS-130. Level-0 data shall be available in real time to local users during overpass.

/166

S-4/-5 MRTD



6.2 UV-VIS-NIR (LEO-UVN)

The UVN instrument shall make observations at the sunlit part of the Earth’s atmosphere according to requirement MR-SYSTEM-10. The physical quantity used by the retrieval algorithms is the ratio of the radiance and the solar irradiance, called reflectance. For this reason it is important to perform regular measurements of the solar irradiance. The frequency of the solar irradiance measurements needs to be consistent with the spectral and radiometric stability and is recommended to be performed once per day /Lev2009/.

MR-LEO-UVN-40. Sun irradiance shall be measured in the same spectral channels as specified for Earth observation

6.2.1 LEO-UVN, Geometrical Requirements

The UVN instrument shall make observations at the sunlit part of the Earth’s atmosphere according to requirements MR-SYSTEM-10 and MR-SYSTEM-20.

MR-LEO-UVN-10. The instrument shall be capable of continuously measuring the radiance from the illuminated side of the Earth’s atmosphere.

MR-LEO-UVN -20. The instrument’s performance requirements shall be met for SZAs smaller than 80o.

MR-LEO-UVN -30. The instrument shall make observations not exceeding an OZA of 66o.

To characterise local sources and sinks, and to minimise cloud contamination of observations, spatial sampling and horizontal resolution need to be finer than the scale of the variations induced by the sources and sinks. A relaxed spatial resolution and SSD are required for UV-1 because radiance from this wavelength range is mostly sensitive to stratospheric ozone, which has less horizontal structure than the signal arising from the troposphere.

MR-LEO-UVN -40. Moved to Section 0

Requirement MR-LEO-UVN-85 poses a lower limit to the width of the IPSF in order to exclude spatial undersampling.

MR-LEO-UVN-85. The IPSF in all channels in the UVN shall be sufficiently broad to allow spatially contiguous observations.

MR-LEO-UVN-88. The spatial variation of the IIE over all channels in the UVN shall be below 25% (TBC).

6.2.2 LEO-UVN, Spectral Requirements

The UVN bands are used to measure several trace gas species, and gain information on aerosols and clouds. In Table 6.2, these data products are listed together with the required spectral ranges and associated priorities in the context of Copernicus service requirements (see Appendix 2 for traceability).

/166

S-4/-5 MRTD



Table 6.2 LEO-UVN data products, spectral ranges and priorities.

Level-2 data product Wavelength range [nm]

Priority

Ozone vertical profile 270 – 330 1

Sulphur dioxide 308 – 325 1 (1)

Albedo 310 – 775 2

Total ozone 325 – 337 1

Aerosol (2) 336 – 340 2 (6)

Formaldehyde 337 – 360 1

Bromine monoxide 345 – 360 2

Rayleigh scattering (cloud), aerosol absorption

360 – 400 2, gaps may be acceptable between 375 nm and 380 nm and between 384 nm and 400 nm.

Aerosol (2) 400 – 430 2 (6)

Nitrogen dioxide 405 – 500 1

Glyoxal 430 – 460 2 (3)

Aerosol (2) 440 – 460 2 (6)

Cloud (O2-O2) 460 – 490 1 (4)

Water vapour and cloud (effective scattering height)

685 - 710 (G) /

710 – 750 (T)

2

Cloud (O2-A band) 750 – 775 1 (4)

Aerosol profile (O2-A band) 750 – 775 1 (4, 5)

(1) Only observed in strongly polluted areas (volcanic eruptions, major power plants)

(2) Column averaged aerosol optical thickness, and aerosol type.

(3) Considered as a by-product rather than a species driving the requirements. It has a weak spectral signature, only temporal averages will be retrieved.

(4) ‘Priority 1’ not as geophysical product itself, but for use in correcting retrievals of other products.

(5) Used for estimating the aerosol profile for air quality and climate applications (see Section 4.2.2 and 4.2.3).

(6) priority ‘2’ in the context of instrument development; priority ‘1’ in the context of Level-2 product development.

/166

S-4/-5 MRTD



Table 6.3. LEO-UVN spectral bands and spectral requirements.

Band ID (1) Spectral range [nm]

Spectral resolution

[nm]

Spectral sampling

ratio

Priority

LEO-UV-1 270–300 1.0 3 1

LEO-UV-2 300–370(2) 0.5 3 1

LEO-VIS 370–500(2) 0.5 3 1

LEO-NIR-1 (3,4)

685–710 0.4 (T) / 0.2 (G) 3 2

710–745 0.4 (T) / 0.2 (G) 3

LEO-NIR-2 750–775 (G) / 755-773 (T)

0.4 (T) / 0.12 (B) (4) /0.06 (G) (4,5) 3 1

(1) It is not mandatory that the wavelength region 270 – 775 nm is split into physically separate bands.

(2) An overlap region of the LEO-UV-2 and the LEO-VIS bands is recommended (360 nm to 380 nm).

(3) The spectral range of the LEO-NIR-1 band is 685-710 nm (goal) or 710-750 nm (threshold). This band is primarily used for water vapour retrieval but is also beneficial to aerosol retrieval.

(4) At least one of the following two capabilities must be achieved: a) a wide spectral range in the NIR (including the NIR-1 and NIR-2 bands) or b) a high (breakthrough or goal) spectral resolution in the NIR-2 band.

(5) The spectral resolution of 0.06 nm is needed for aerosol profile retrievals.

/166

S-4/-5 MRTD



Table 6.4. LEO-UVN spectral bands and spectral requirements and SNR requirements.

Band ID Spectral range [nm]

SNR(1,2)

@ Lref_TR_uvn

SNR(3)

@ Lsun

LEO-UV-1 270–300 100(3) @ 270 nm 8000

LEO-UV-2 300–370(4)

1000 @ 320 nm (T)

1000 @ 310 nm (G)

8000, for 300 nm < < 325 nm; 10000,

for 325 nm < < 370 nm

LEO-VIS 370–500(5)

1500 @ 420 nm, for < 430 nm

15000 1500 (T) / 2130 (G) @ 460 nm

for 430 nm < < 460 nm (6)

1500 @ 460 nm, for > 460 nm

LEO-NIR-1

685–710 (G) 500 @ 710 nm

5000 (T) / 10000 (G) (6)

710–745 (T)

500 @ 745 nm 5000 (T) / 10000 (G)

(6)

LEO-NIR-2 750–775

(G) / 755-773 (T)

500 @ 755 nm 5000 (T) / 10000 (G)

(6)

(1) The SNR at other wavelengths in a band is obtained by assuming that the SNR varies with the square-root of the radiance level. The SNR values apply at reference Spectral Sampling Interval (SSI) defined by the spectral sampling ratio and threshold spectral resolution. The SNR at other SSI is obtained by assuming that the SNR varies with the square-root of the SSI.

(2) The SNR requirements do not apply in the spectral ranges near strong Fraunhofer lines specified in Table 6.5.

(3) For this band on-ground binning up to threshold SSD is allowed in order to meet the SNR.

(4) Degraded performance may be acceptable in the region 300-312 nm.

(5) Degraded performance may be acceptable in the region 375–380 nm and 384–400 nm.

(6) The goal SNR shall not drive the mission.

/166

S-4/-5 MRTD



Table 6.5. Spectral ranges near strong Fraunhofer lines.

Lower limit [nm] Upper limit [nm]

271.7 272.7

273.5 275.5

278.2 281.5

284.2 286.2

287.5 288.7

296.7 297.2

298.1 298.9

300.0 300.7

301.7 302.5

309.0 310.5

315.6 316.5

318.0 318.6

319.3 319.7

321.0 321.8

322.5 324.0

335.7 337.7

343.8 345.0

356.7 357.4

357.9 359.0

373.2 375.4

381.7 384.4

385.7 386.4

387.0 389.0

392.5 394.5

396.0 398.0

429.7 431.5

433.8 434.5

438.2 438.9

485.5 487.5

/166

S-4/-5 MRTD



MR-LEO-UVN-90. The UVN spectral bands shall include the ranges specified in Table 6.3 taking into account the priorities and optional bands.

MR-LEO-UVN-100. The spectral resolution shall be smaller than or equal to the values specified in Table 6.3 For the NIR-2 band, optimization of spectral resolution takes priority over optimization of SNR and over implementation of the NIR-1 band.

The spectral variation of the spectral sampling should be smooth.

MR-LEO-UVN-110. The spectral sampling ratio shall be larger than or equal to the values specified in Table 6.3.

Accurate spectral calibration is required which can be fulfilled by analysis of in-flight data of solar irradiance using known Fraunhofer lines, and Earth radiance data using Fraunhofer lines and atmospheric absorption features. Since the spectral resolution will not be sufficient to fully resolve the narrow Fraunhofer lines, this analysis depends on high spectral stability between the solar and atmospheric measurements limiting errors due to resampling. Therefore, the spectral stability between two solar measurements is required. The fitting process also depends on high radiometric accuracy and SNRs of the radiance and solar irradiance spectra.

MR-LEO-UVN-120. The position of the spectral channel shall be known with an accuracy equal to or better than

- 0.02 SSI for spectral channels located below 300 nm, - 0.01 SSI for spectral channels located above 300 nm in the UV-vis. - 0.05 SSI (TBC) for spectral channels located in the NIR

for spatially uniform and non-uniform scenes. This knowledge requirement can be fulfilled by analysis of in-flight data relying on the specified spectral stability.

MR-LEO-UVN-130. For spatially uniform scenes, the position of the spectral channels in the radiance measurements shall not vary by more than 0.1 SSI for Level-0 samples acquired between two consecutive solar measurements.

For a spatially homogeneous scene the ISRF shall be known sufficiently accurately such that the following requirements are met for all applicable wavelengths and viewing angles.

MR-LEO-UVN-142. The ISRF shall be known to an accuracy better than 1% of the peak value of the ISRF in the spectral range where the ISRF is at least 1% of the peak value. The FWHM of the ISRF shall be known to an accuracy of 1%.

MR-LEO-UVN-146. The shape of the ISRF shall be sufficiently peaked such that the integrated areas satisfy

dd )(ISRF7.0)(ISRFFWHM

The wavelength range where MR-LEO-UVN-142 applies is chosen such that localized spectral stray light (ghosts) and spatial stray light do not dominate compliance. Outside

/166

S-4/-5 MRTD



this wavelength range the ISRF is constrained by stray light requirements. MR-LEO-UVN-146 excludes slit functions with strongly pronounced wings.

6.2.3 LEO-UVN, Radiometric Requirements

6.2.3.1 Radiometric Requirements for Earth Reflectance and Radiance

Radiometric requirements are specified using a number of reference radiance scenarios for tropical, high latitude and ozone hole conditions (Table 6.6) calculated for a standard atmosphere (US Standard 1976) assuming background aerosol, no clouds and no precipitation.

Table 6.6. LEO-UVN reference scenarios.

Scene SZA Surface albedo

ID

Tropical bright 0° 1.00 Lmax_TR_uvn

Tropical dark 0° 0.02 Lref_TR_uvn

High–latitude dark 75° 0.02 Lref_HL_uvn

Ozone hole bright 75° 1.00 Lmax_OH_uvn

/166

S-4/-5 MRTD



Figure 6.2. Earth radiance levels of the LEO UVN Reference Scenes specified in Table 6.6. The spectra in the UV-Vis region (upper panel) are shown with a spectral resolution of 1 nm (below 300 nm) or 0.5 nm (above 300 nm) assuming a Gaussian ISRF. The spectra in the NIR region (lower panel) are shown with a spectral resolution of 0.4 nm and 0.06 nm assuming a Gaussian ISRF.

/166

S-4/-5 MRTD



MR-LEO-UVN-150. The SNR of each spectral channel for (Earth) radiance measurements shall be larger than or equal to the values specified in Table 6.4. These SNR values are specified for the radiance levels of the reference scenario Lref_TR_uvn.

MR-LEO-UVN-160. The absolute radiometric accuracy (ARA) of the Earth radiance and reflectance shall be better than specified in Table 6.7. This requirement applies in the reference dynamic range specified in Table 6.7. This error budget includes all instrumental errors, but excludes radiometric noise and errors in radiometric standards. This requirement is applicable to polarised scenes with a degree of polarisation of 100% with any polarisation direction. This requirement is also applicable to spatial samples located in the dark zone outside of the transition zone in the contrast scene with the radiance levels of the scenarios Lref_TR_uvn and Lmax_TR_uvn (Llow and Lhigh in Figure 6.3). The width of the transition zone is 2 SSD (G)/10 SSD(T).

Table 6.7. LEO-UVN absolute radiometric accuracy (ARA) of the signal S and reference dynamic range. The signal Smin is defined by the High-latitude Dark reference scenario (Lref_HL_uvn). The signal Smax is defined by the maximum of the reference scenarios Tropical bright (Lmax_TR_uvn) and Ozone hole bright (Lmax_OH_uvn).

Band Reference dynamic range

ARA Min Max

LEO-UV-1 Smin(290 nm) for λ ≤290 nm

Smin(λ) for λ ≥ 290 nm Smax(λ) 3% S(λ)

LEO-UV-2 Smin(λ) Smax(λ) 3% S(λ)

LEO-Vis Smin(λ) Smax(λ) 3% S(λ)

LEO-NIR-1 Smin(710 nm) Smax(λ) 3% S(λ)

LEO-NIR-2 Smin(755 nm) Smax(λ) 3% S(λ)

MR-LEO-UVN-165. Radiometric errors in radiance and irradiance due to errors in radiometric standards shall be smaller than 1% (TBC).

The relative spatial radiometric accuracy is a viewing angle dependency with respect to the nadir viewing direction.

MR-LEO-UVN-170. The relative spatial radiometric accuracy of the (Earth) radiance and reflectance measurements shall be smaller than 0.25%.

/166

S-4/-5 MRTD



Requirements are posed on relative spectral radiometric accuracy in order to constrain instrumental spectral features that correlate with the absorption cross section of target trace gases.

MR-LEO-UVN-180. The peak-to-peak relative spectral radiometric accuracy (RSRA) of the (Earth) reflectance measurements shall be better than specified hereafter. This requirement is applicable to polarised scenes with a degree of polarisation of 100% with any polarisation direction. This requirement is also applicable to spatial samples located in the dark zone outside of the transition zone in the contrast scene with the radiance levels of the scenarios Lref_TR_uvn and Lmax_TR_uvn (Llow and Lhigh in Figure 6.3). The width of the transition zone is 2 SSD (G) / 10 SSD (T). When evaluating this requirement on reflectance it shall be assumed that radiance and irradiance are sampled on same spectral grid and that there is no spectral shift between radiance and irradiance.

0.25% in the UV and Vis, for any spectral window up to 3 nm wide 0.25% in the NIR, for any spectral window up to 7.5 nm wide.

MR-LEO-UVN-185. The Effective Relative Spectral Radiometric Accuracy (ESRA) shall be better than the values specified in the Table 6.8. The gain vectors to be used for evaluating ESRA are specified in the Appendix 4. This requirement is applicable to polarised scenes with a Degree Of Polarisation specified in Table 6.8 with any polarisation direction. This requirement is also applicable to spatial samples located in the dark zone outside of the transition zone in the contrast scene with the radiance levels of the scenarios Lref_TR_uvn and Lmax_TR_uvn (Llow and Lhigh in Figure 6.3). The width of the transition zone is 2 SSD (G) / 10 SSD (T). When evaluating this requirement on reflectance it shall be assumed that radiance and irradiance are sampled on same spectral grid and that there is no spectral shift between radiance and irradiance.

/166

S-4/-5 MRTD



Table 6.8. ESRA requirement and Degree Of Polarisation (DOP).

Spectral window [nm]

ESRA [1/sr] DOP Gain

310-325 4.20e-6 1 Figure A4.1

325-337 2.35e-4 1 Figure A4.2

337-360 3.35e-6 1 Figure A4.3

410-460 6.20e-5 0.6 Figure A4.4

685-710, 755-773 (1) 3.50e-5 1 Figures A4.5 and A4.6

755-773 (2) 1.40e-5 Spectrally varying (see Figure A4.8)

Figure A4.7

(1) for threshold spectral resolution

(2) for breakthrough spectral resolution

Figure 6.3. Illustration of a contrast scene considered for the radiometric requirements. The radiance levels Llow and Lhigh are defined by reference spectra specified in the radiometric requirements for each band. Radiometric requirements shall be met in the dark zone outside of the transition zone. Note that the scene can be oriented in any direction.

/166

S-4/-5 MRTD



6.2.3.2 Radiometric Requirements for Solar Irradiance Measurements

Regular solar measurements are needed for accurate determination of the Earth reflectance. For an impression of the solar irradiance levels in the UVN see the solar reference spectrum Es_uvn (/Dob2008/ for UV-VIS, /Cha2010/ for R) depicted in Fig. 6.4.

Figure 6.4. Extra-terrestrial UVN solar irradiance levels.

MR-LEO-UVN-220. The SNR of each spectral channel for solar irradiance measurements shall be larger than or equal to the values specified in Table 6.4.

MR-LEO-UVN-230. The absolute radiometric accuracy of the solar irradiance measurements shall be better than 1% (G) / 2% (T) excluding the uncertainty of the reference solar spectrum, errors on radiometric standards and radiometric noise. This requirement applies at irradiance levels defined by the reference solar spectrum Es_uvn (see Fig. 6.4)

MR-LEO-UVN-235. The absolute radiometric accuracy of the solar irradiance measurements shall be better than 5% (G) / 6% (T) of the reference solar reference spectrum in the continuum, including the uncertainty of the reference solar spectrum but excluding errors on radiometric standards and radiometric noise. This requirement does not apply at stronger solar Fraunhofer lines.

/166

S-4/-5 MRTD



MR-LEO-UVN-240. The relative radiometric accuracy of the irradiance within the full range of elevation and azimuth incidence angles of the sun (irradiance goniometry) shall be better than 0.5%.

6.2.3.3 Radiometric Stability

Assuming a stable and spatially uniform scene, two requirements on the radiometric stability are made. They apply to Level-1b, including corrections for degradation.

MR-LEO-UVN-257. For any stable and spatially uniform scene, the Earth spectral radiance and spectral reflectance shall be constant for any given spectral channel to within 0.5% over the day time part of the orbital period. This requirement applies over the reference dynamic range specified in Table 6.7.

MR-LEO-UVN-258. For any stable and spatially uniform scene, the long-term average Earth spectral radiance and spectral reflectance and Sun extraterrestrial spectral irradiance shall be constant for any given spectral channel to within 0.5% over the satellite lifetime. This requirement applies over the reference dynamic range specified in Table 6.7.

6.2.3.4 Polarisation

The light reflected by the Earth and the atmosphere can be strongly linearly polarised. The relative spectral variation of the polarization sensitivity should be characterized on-ground to an accuracy of (TBD).

MR-LEO-UVN-260. The polarisation sensitivity of each spectral channel shall be less than 0.005 in the UV, Vis and NIR-1, 0.005 in the NIR-2 in the central part of the swath (nadir ±27.1˚), 0.007 in the NIR-2 outside the central part of the swath.

MR-LEO-UVN-270. Removed.

6.3 SWIR (LEO-SWIR)

This section provides Level-1B requirements for the SWIR bands of the Sentinel-5 and -5P missions. From the 2.3 µm and 1.6 µm bands, the column densities of CO and CH4 including the contribution from the planetary boundary layer can be retrieved.

The CH4 accuracy requirement of 2% is very strict which makes the corresponding retrieval challenging. At present two approaches are proposed for the SWIR CH4 retrieval: The first retrieval approach aims to retrieve methane column density simultaneously with aerosol and cirrus properties. This retrieval approach makes use of a combination of both SWIR bands and the O2-A band. Alternatively the so-called proxy approach is suggested. Here the CO2 column retrieved from the 1.6 µm band is used as proxy for the effect of aerosols and cirrus on the light path. This approach requires prior knowledge on CO2. The Sentinel-5P mission will not have a 1.6 µm channel and will thus not have the option to use the proxy method.

/166

S-4/-5 MRTD



For CO retrieval from Sentinel-5 a two-band retrieval approach is used which combines the 2.3 µm band and the O2-A band.

6.3.1 LEO-SWIR, Geometrical Requirements

Observations in the SWIR will be made at the sunlit part of the Earth’s atmosphere according to requirement MR-SYSTEM-10.

MR-LEO-SWIR-10. Measurements in the SWIR shall be made of the reflected solar radiance on the illuminated side of the Earth’s atmosphere.

MR-LEO-SWIR-20. The performance requirements shall be met for SZAs smaller than 80o.

MR-LEO-SWIR-30. SWIR observations shall be made not exceeding an OZA of 66o.

Requirement MR-LEO-SWIR-55 poses a lower limit to the width of the IPSF in order to exclude spatial undersampling.

MR-LEO-SWIR-55. The IPSF in all channels in the SWIR shall be sufficiently broad to allow spatially contiguous observations.

Requirement MR-LEO-SWIR-70 shall hold for all pixels.

MR-LEO-SWIR-70. The SSD in all spectral channels shall be identical within 5%.

6.3.2 LEO-SWIR, Spectral Requirements

As mentioned before, the SWIR sounder has two main priority bands with one possible additional band, here to be considered as option for better data retrieval.

MR-LEO-SWIR-90. The SWIR sounder shall cover the spectral bands according to the ranges as specified in Table 6.9 taking into account the priorities.

MR-LEO-SWIR-100. The spectral resolution shall be smaller than or equal to the values specified in Table 6.10.


MR-LEO-SWIR-110. The spectral sampling ratio shall be larger than or equal to the values specified in Table 6.10.

/166

S-4/-5 MRTD



Table 6.9. LEO-SWIR data products, spectral ranges and priorities.

Level-2 data product Wavelength range [nm]

Priority

CH4 (& CO2) 1590–1675 1

TBD TBD 3

CO (& CH4) 2305–2385 1

Table 6.10. LEO-SWIR spectral bands and spectral requirements.

Band ID Spectral range [nm]

Spectral resolution [nm] Spectral sampling ratio

LEO-SWIR-1 1590–1675 0.25 3 (G) / 2.5 (T)

LEO-SWIR-2 TBD 0.25(1) 3 (G) / 2.5 (T)

LEO-SWIR-3 2305–2385 0.25 3 (G) / 2.5 (T)

(1) The LEO-SWIR-2 band may not drive the mission. In this band, achieving the specified spectral resolution has higher priority than achieving the specified SNR and spectral range.

MR-LEO-SWIR-120. The position of the spectral channels shall be known with an accuracy better than 0.006 nm (SWIR-1, SWIR-2 (TBC)) and 0.005 nm (SWIR-3) for spatially uniform and non-uniform scenes. This requirement will be fulfilled by analysis of in-flight data.

MR-LEO-SWIR-130. For spatially uniform scenes, the position of the spectral channels shall not vary by more than 0.012 nm (SWIR-1, SWIR-2 (TBC)) and 0.01 nm (SWIR-3) for Level-0 samples acquired between two consecutive solar measurements.

MR-LEO-SWIR-140. The ISRF shall be known to an accuracy better than 1% of the peak value in the spectral range where the ISRF is at least 1% of the peak value. Additionally, in the wings (in two spectral ranges within , each 0.125 wide, located at the upper and lower limit of respectively) the ISRF shall be known to an accuracy better than 20%. The FWHM of the ISRF shall be known to an accuracy of 1%. This requirement applies to spatially homogeneous scenes.

For a spatially homogeneous scene the ISRF shall be known sufficiently accurately such that the following requirements are met for all applicable wavelengths and viewing angles.

MR-LEO-SWIR-146. The shape of the ISRF shall be sufficiently peaked such that the integrated areas satisfy

/166

S-4/-5 MRTD




.

The wavelength range where MR-LEO-SWIR-140 applies is chosen such that localized spectral stray light (ghosts) and spatial stray light do not dominate compliance. Outside this wavelength range the ISRF is constrained by stray light requirements. MR-LEO-SWIR-146 excludes ISRFs with a stronlgy skewed shape or with strongly pronounced wings.

6.3.3 LEO-SWIR, Radiometric Requirements


The radiometric requirements of the radiance measurements applies to three reference scenes (see Table 6.11), including a minimum and maximum radiance case for both a tropical and a high-latitude situation. For the SNR requirements the low albedo (“minimum”) case Lref_HL_swir is applicable. For an impression of the radiance levels see Figure 6.5. Note that the radiometric accuracy requirements have been derived making the following simplifying assumption: In case of requirements for Earth radiance no radiometric bias on the solar irradiance has been assumed, while in case of the requirements for solar irradiance no bias on the Earth radiance has been assumed. For radiometric requirements on reflectance it is assumed that that Earth radiance and solar irradiance are sampled on the same spectral grid and that there is no spectral shift between Earth radiance and solar irradiance spectra.

Table 6.11. LEO-SWIR reference scenarios.

Scene SZA Surface albedo ID

SWIR-1 SWIR-2 SWIR-3

Tropical bright 10° 0.75 0.7 (TBC) 0.65 Lmax_TR_swir

Tropical dark 10° 0.05 0.05 (TBC) 0.05 Lref_TR_swir

High–latitude dark 70° 0.05 0.05 (TBC) 0.05 Lref_HL_swir

/166

S-4/-5 MRTD



Figure 6.5. Earth radiance levels of the LEO-SWIR Reference Scenes specified in Table 6.11 for the bands SWIR-1 and SWIR-3. The spectra are shown with a spectral resolution of 0.25 nm assuming a Gaussian ISRF.

/166

S-4/-5 MRTD



MR-LEO-SWIR-150. The SNR of (Earth) radiance shall be larger than the values specified in Table 6.12. These SNR values are specified for the maximum radiance of the reference scenario Lref_HL_swir in each band assuming a spectral resolution of 0.25 nm and a spectral sampling ratio of 3.

Table 6.12. LEO-SWIR SNR requirements.

Band ID Maximum radiance levels of

Lref_HL_swir [1012 photons /(s sr

cm2 nm)]

Max irradiance of

Lsun(1)

[1012 photons /(s cm2 nm)]

SNR (2, 3)

radiance

SNR (1, 2, 3)

irradiance

LEO-SWIR-1 1.1 220 (4) 5000

LEO-SWIR-2 0.5 (TBC) 140 (TBC) 100 (TBC) 5000

LEO-SWIR-3 0.4 85 (4) 5000

(1) Temporal averaging on-ground is allowed if necessary to meet the requirement. It is expected that this requirement will not drive the radiometric sizing of the instrument.

(2) The SNR at other wavelengths in a band is obtained by assuming that the SNR varies with the square-root of the radiance level.

(3) The SNR values apply at reference SSI defined by the spectral sampling ratio and threshold spectral resolution. The SNR at other SSI is obtained by assuming that the SNR varies with the square-root of the SSI.

(4) SNR requirements for radiance are specified in terms of reference SNR spectra shown in Figure 6.5.

/166

S-4/-5 MRTD



Figure 6.5. Reference SNR spectra for the SWIR-1 (top) and SWIR-3 (bottom) for sampling ratios 2.5 (red) and 3 (green).

/166

S-4/-5 MRTD



For radiance this requirement needs to be verified by on-ground radiometric calibration, which should demonstrate the accuracy in-flight (including corrections for long-term drifts and in-flight calibration hardware). For the reflectance it is applicable over the whole mission duration.

MR-LEO-SWIR-160. The absolute radiometric accuracy (ARA) of the Earth radiance and reflectance shall be better than specified in Table 6.13. This requirement applies in the reference dynamic range specified in Table 6.13. This error budget includes all errors, except radiometric noise and errors in radiometric standards. This requirement is applicable to polarised scenes with a degree of polarisation of 34% with any polarisation direction. This requirement is also applicable to spatial samples located in the dark zone outside of the transition zone in the contrast scene with the radiance levels of the scenarios Lref_TR_swir and Lmax_TR_swir (Llow and Lhigh in Figure 6.3). The width of the transition zone is 2 SSD (G)/10 SSD (T).

Table 6.13. LEO-SWIR absolute radiometric accuracy (ARA) of the signal S and reference dynamic range. The signal Smin is defined by the reference scenario High-latitude Dark (Lref_HL_swir). The signal Smax is defined by the reference scenario Tropical Bright (Lmax_TR_swir).

Band Reference dynamic range

ARA Min Max

LEO-SWIR-1 Smin(λ) Smax(λ) 4% S(λ) (G) / 6% S(λ) (T)

LEO-SWIR-2 Smin(λ) Smax(λ) TBD

LEO-SWIR-3 Smin(λ 2313 nm) Smax(λ) 1.5% S(λ) (G) / 3.5% S(λ) (T)

MR-LEO-SWIR-165. Radiometric errors in radiance and irradiance due to errors in radiometric standards shall be smaller than 0.5% (TBC).

MR-LEO-SWIR-170. The relative spatial radiometric accuracy of the (Earth) radiance and reflectance measurements within a swath shall be better than 0.25%

(1σ). This requirement applies in the reference dynamic range specified in Table 6.13.

Some optical components can introduce spectral features which are similar in shape and magnitude to the features used to retrieve certain trace gases. Therefore requirements are set for the relative spectral radiometric accuracy.

/166

S-4/-5 MRTD



MR-LEO-SWIR-185. The Effective Relative Spectral Radiometric Accuracy (ESRA) shall be better than the values specified in Table 6.14. The gain vectors to be used for evaluating ESRA are specified in the Appendix 4. This requirement is applicable to polarised scenes with a Degree Of Polarisation specified in Table 6.14 with any polarisation direction. This requirement is also applicable to spatial samples located in the dark zone outside of the transition zone in the contrast scene with the radiance levels of the scenarios Lref_TR_uvn and Lmax_TR_uvn (Llow and Lhigh in Figure 6.3). The width of the transition zone is 2 SSD (G) / 10 SSD (T). When evaluating this requirement on reflectance it shall be assumed that radiance and irradiance are sampled on same spectral grid and that there is no spectral shift between radiance and irradiance.

Table 6.14. ESRA requirement and Degree Of Polarisation (DOP).

Spectral window [nm]

ESRA [1/sr] DOP Gain

1590-1675 6.55E-5 0.15 Figure A4.9

2305-2385 2.30E-4 0.15 Figure A4.10

2315-2340 4.30E-5 0.15 Figure A4.11

MR-LEO-SWIR-198. Additive spectrally constant radiometric errors in the reflectance shall be smaller than 0.2% of the maximum reflectance in the bands SWIR-1 and SWIR-3. This may be achieved in-flight by a baseline correction scheme. Number for SWIR-2 band TBD.


Regular solar measurements are needed for calibration of Earth reflectance and for spectral calibration using Fraunhofer lines. For an impression of the irradiance levels in the SWIR see the reference solar spectrum Es_swir depicted in Figure 6.6. Note that the radiometric accuracy requirements have been derived making the following simplifying assumption: In case of requirements for Earth radiance no radiometric bias on the solar irradiance has been assumed, while in case of the requirements for solar irradiance no bias on the Earth radiance has been assumed.

MR-LEO-SWIR-230. Sun irradiance shall be measured in the same spectral channels as specified for Earth observation.

/166

S-4/-5 MRTD



Figure 6.6 Extra-terrestrial SWIR solar irradiance levels. The spectra are shown with a spectral resolution of 0.25 nm assuming a Gaussian ISRF.

MR-LEO-SWIR-240. The SNR of each band for solar irradiance measurements shall be larger than the values specified in Table 6.12 using the maximum irradiance levels given in the same table.

MR-LEO-SWIR-250. The absolute radiometric accuracy of the solar irradiance measurements shall be better than 2% excluding the uncertainty of the reference solar spectrum, errors on radiometric standards and radiometric noise. This requirement applies at irradiance levels defined by the reference solar spectrum Es_swir (see Fig. 6.6).

/166

S-4/-5 MRTD



MR-LEO-SWIR-255. The absolute radiometric accuracy of the solar irradiance measurements shall be better than 5% of the reference solar reference spectrum in the continuum, including the uncertainty of the reference solar spectrum but excluding errors on radiometric standards and radiometric noise. This requirement does not apply at stronger solar Fraunhofer lines.

MR-LEO-SWIR-260. The relative radiometric accuracy of the solar irradiance measurements within the full range of elevation and azimuth incidence angles of the sun (irradiance goniometry) shall be known to an accuracy of 0.5% (1σ).


Earth radiance in the SWIR can be significantly linearly polarised mainly due to surface reflection. It is recommended that the relative spectral variation of the polarization sensitivity is characterized on-ground.

MR-LEO-SWIR-280. The polarisation sensitivity within the SWIR shall be less than 0.2.

MR-LEO-SWIR-285. The instrument polarization sensitivity in the SWIR shall be known with an accuracy better than 0.05.

MR-LEO-SWIR-287. The deviation of the ratios of the Mueller matrix elements [M01 / M00] and [M02 / M00] of the instrument Mueller matrix M from a low order (≤2) polynomial function of the wavelength within any band in the SWIR shall be smaller than 0.01.

6.3.3.4 Radiometric Stability

Assuming a stable and spatially uniform scene, two requirements on the radiometric stability are made. They apply to Level-1b, including corrections for degradation.

MR-LEO-SWIR-290. The Earth spectral radiance and spectral reflectance shall be constant for any given spectral channel to within 0.5% over the day time part of the orbital period. This requirement applies in the reference dynamic range specified in Table 6.13.

MR-LEO-SWIR-300. The running orbital average Earth spectral radiance and spectral reflectance and Sun extraterrestrial spectral irradiance shall be constant for any given spectral channel to within 0.5% over the satellite in-orbit lifetime. This requirement applies in the reference dynamic range specified in Table 6.13.

/166

S-4/-5 MRTD



6.4 Thermal Infrared (LEO-TIR)

The Sentinel-5 system will utilise Thermal InfraRed (TIR) data from the EPS-SG Infrared Atmospheric Sounder (IAS). This section reflects the current understanding of the expected TIR performance, and is not meant to establish ESA mission requirements on IAS.

The required spectral bands, together with the target species, are listed in Table 6.15. An extended spectral coverage is strongly preferred over several narrower separate spectral windows. The geometrical, spectral and radiometric requirements specified in Sections 6.4.1, 6.4.2, and 6.4.3 are in line with the EPS-SG-EURD /EUM2013/.

6.4.1 LEO-TIR, Geometrical Requirements

TIR observations shall be made using a nadir geometry with the Earth and atmospheric thermal emission as a source. As it is not relying on solar radiation, data shall be acquired continuously during day and night.

MR-LEO-TIR-10. Data acquisition shall be performed continuously.

MR-LEO-TIR-20. TIR observations shall not exceed an OZA of 66o.

The highest priority shall be given to the spatial resolution. Note that this priority is also higher than harmonizing spatial resolution among the instruments.

MR-LEO-TIR-30. The spatial resolution at SSP shall be 12 by 12 km2 (T) / 5 by 5 km2 (G); this shall apply along and across track.

MR-LEO-TIR-35. The spatial resolution in all TIR bands shall be identical to within 5%. This shall apply along and across track.

MR-LEO-TIR-37. Intra-band spatial co-registration of each of the TIR bands shall be better than 0.05 of the spatial resolution.

MR-LEO-TIR-39. Inter-band spatial co-registration between bands in the TIR shall be known to better than 0.05 of the spatial resolution.

MR-LEO-TIR-40. The off-nadir spatial resolution shall be obtained by maintaining the same solid angle of the FOV.

MR-LEO-TIR-50. The IE over a squared area (resp. circle) covered by the spatial resolution shall be larger than 90%.

Spatial uniformity of the IPSF controls pseudo-noise with respect to the radiance emanating from a heterogeneous Earth scene, i.e., deviations due to a non-uniform spatial response (radiometric calibration pseudo-noise). A good knowledge of the spatial IPSF is also required to be able to quantify the cloud fraction in the sounder pixel used by cloud-clearing algorithms, i.e. almost all of the observed radiances should originate from the observed scene and in a homogenous manner.

MR-LEO-TIR-60. The IPSF shall be flat to within 5% peak-to-average within a diameter of 80% of the spatial resolution.

/166

S-4/-5 MRTD



MR-LEO-TIR-70. The IPSF shall be characterised within a diameter of 200% of the spatial resolution.

6.4.2 LEO-TIR, Spectral Requirements

In order to cover all target species in the TIR, several spectral bands and with specific resolutions are required as indicated in Table 6.16 (see Appendix 2 for tracability to Copernicus service requirements). Spectral bands are defined in terms of measurable species. The species highlighted in bold are primarily targeted in the spectral bands. An extended wavenumber coverage is very strongly preferred over narrower independent windows, allowing numerous species and also aerosol to be probed simultaneously.

The term threshold in this table indicates the minimum performance below which the data would have no value in supporting identified operational applications. Threshold requirements will not allow reaching some of the most stringent L2 requirements defined for S5 operation, namely the vertical resolution in the troposphere and the improved sensitivity to the surface.

Further study on band 8 is needed to decide on its merit for CH4 (with respect to SWIR band) and its potential to substitute band 3 for HNO3. For enhanced height-resolving capabilities in the low atmosphere during day and night, a high spectral resolution is required to use information from the pressure-broadened atmospheric lines.

MR-LEO-TIR-90. The TIR spectral bands shall cover the ranges specified in Table 6.16 taking into account the priorities and optional alternative bands.

MR-LEO-TIR-100. The unapodized spectral resolution defined as 0.6/OPD max shall be smaller than or equal to the values specified in Table 6.16. OPDmax refers to the maximum Optical Path Difference of the Fourier transform spectrometer. The self-apodized spectral resolution shall not exceed 1.7 times the unapodized spectral resolution.

MR-LEO-TIR-110. The spectral sampling shall be 0.5/OPDmax.

/166

S-4/-5 MRTD



Table 6.15. LEO-TIR data products, spectral ranges, and priorities. The species highlighted in bold are primarily targeted in the spectral bands.

Species Spectral range

(cm-1) Priority

C2H2, HCN 650-750 3

PAN (HONO, C2H6, NH3, CFC11) 750-850 2

HNO3 (NH3, CFC11) 850-920 2

NH3, C2H4 (CFC12) 920-980 1

O3 profile 1030-1080 1

CH3OH (NH3) 980-1080 1

HCOOH (SO2, NH3, CFC12) 1080-1130 1

SO2(1) (PAN, NH3, CFC12) 1130-1200 1

CH4, H2O(2) (N2O, HNO3, SO2-UT) 1200-1350 2

SO2(3) (H2O) 1350-1400 1

CO 2140-2180 1

CH4 2700-2760 3

CH4 2760-2900 4

(1) Within the planetary boundary layer and the free troposphere.

(2) Includes suitable information for heavier water isotopes.

(3) Within the upper troposphere. High priority justified by demonstrated operational aerial safety applications.

/166

S-4/-5 MRTD



Table 6.16. LEO-TIR spectral bands and requirements. Goal, Threshold and Breakthrough values (G/B/T) are provided for spectral resolution and NeDT.

Band ID Spectral range

(cm-1)

Unapodized spectral resolution(cm-1) G/B/T

NeDT (K @ 280 K)

G/B/T LEO-TIR-1 650-750 0.15/0.225/0.30 0.07/0.14/0.25

LEO-TIR-2 750-850 0.15/0.225/0.30 0.07/0.14/0.25

LEO-TIR-3 850-920 0.075/0.15/0.30 0.07/0.14/0.25

LEO-TIR-4 920-980 0.075/0.15/0.30 0.07/0.14/0.25

LEO-TIR-5a 1030-1080 0.075/0.15/0.30 0.07/0.14/0.25

LEO-TIR-5b 980-1080 0.075/0.15/0.30 0.07/0.14/0.25

LEO-TIR-6 1080-1130 0.15/0.225/0.30 0.07/0.14/0.25

LEO-TIR-7 1130-1200 0.15/0.225/0.30 0.07/0.14/0.25

LEO-TIR-8 1200-1350 0.15/0.225/0.30 0.07/0.14/0.25

LEO-TIR-9 1350-1400 0.15/0.225/0.30 0.07/0.14/0.25

LEO-TIR-10 2140-2180 0.075/0.15/0.30 0.07/0.14/0.25

LEO-TIR-11 2700-2760 0.15/0.225/0.30 0.14/0.25/0.3

LEO-TIR-12 2760-2900 0.15/0.225/0.30 0.14/0.25/0.3

The knowledge of the ISRF shall be such that for any spectral sample/channel the associated radiometric uncertainty satisfies

01.0~~ ISRF Equation 6.1,

for which the ISRF has been normalised according to

1~~ ISRF

Equation 6.2.

MR-LEO-TIR-120. For a spatially homogeneous scene, the ISRF of each spectral channel shall be known with an associated radiometric uncertainty such that Equation 6.1 and Equation 6.2 apply.

MR-LEO-TIR-130. The position of the spectral channels centre shall be known with an accuracy better than 110-6 (1) for /.

6.4.3 LEO-TIR, Radiometric Requirements

The radiometric requirements given in this section are specified per spectral band and are set for a (reference) surface temperature of 280 K. For temperatures T different from 280

K, the specified noise or accuracy shall be scaled by the factor

dT

TdB

dT

KdB ),()280,( 00 ,

/166

S-4/-5 MRTD



where B is the Planck’s function, 0 is the wavenumber of the channel and T is the target temperature in K.

The radiometric requirements given in this section shall be met over the dynamic range of the TIR sounder, which shall be optimised to cover the spectral radiances for nadir measurement scenes ranging from a clear-sky hot desert location to a cold icy place. Extreme situations such as clouds (~180 K) and fires events (>350 K) are not driving this requirement. For an impression of the TIR radiance levels see Figure 6.7.

The signal dynamic range for all spectral bands is 180 – 330 K brightness temperature.

MR-LEO-TIR-140. The radiometric noise for unapodized spectra specified in terms of NEdT shall be less than the values specified in Table 6.16.

MR-LEO-TIR-150. The absolute radiometric accuracy shall be better than 0.5 K at 280 K TOA brightness temperature.

MR-LEO-TIR-160. For radiometric stability, the data acquired in each spectral channel shall be constant to within at least 0.1 K (T) / 0.05 K (G) over one orbital period, assuming a stable and spatially uniform scene of 280 K TOA brightness temperature.

MR-LEO-TIR-170. The relative spectral radiometric accuracy over the complete spectral range within a TIR band shall be smaller than 0.1 K (RMS) assuming a stable and spatially uniform scene of 280 K TOA brightness temperature.

MR-LEO-TIR-180. The relative spatial radiometric accuracy shall be smaller than 0.1 K (RMS) assuming a stable and spatially uniform scene of 280 K TOA brightness temperature.

/166

S-4/-5 MRTD



Figure 6.7. LEO-TIR radiance levels with and without the influence of solar illumination on the observed atmosphere (dark blue and magenta line, respectively). <from file ‘LEO Radiance TIR.xls’>

Radiance TIR [W/(cm^2 sr cm^-1)]

1.E-09

1.E-08

1.E-07

1.E-06

1.E-05

1.E-04

800 1300 1800 2300 2800

Wavenumber [1/cm]

Radiance day

Radiance day, part 2

Radiance night

/166

S-4/-5 MRTD



6.5 Cloud Imagery Data

The Sentinel-5 system shall make use of imaging data from the EPS-SG VIS/IR imager (VII) as auxiliary information on clouds. These data are critical for cloud screening and for taking into account cloud contamination during Level-2 processing. For instrument requirements we refer to the EPS-SG End User Requirements Document /EUM2013/.

6.6 Aerosol Data

The Sentinel-5 system shall make use of data from the EPS-SG Multi-Viewing Multi-Channel Multi-Polarisation imager (3MI). These data provide essential aerosol information for the Copernicus Air Quality and Climate applications and are of integral importance for aerosol corrections of Level-2 trace gas products. This section reflects the current understanding of the expected 3MI performance, and is not meant to establish ESA mission requirements on 3MI.

On top of the requirements specified in /EUM2013/ the requirements listed below need to be satisfied by 3MI in order to meet the Copernicus Air Quality and Climate user requirements on aerosol optical thickness, and the user requirements on single scattering albedo, size distribution, refractive index formulated by Mishchenko et al. (2004) for the Advanced Polarimetric Sounder (APS) on the NASA-GLORY mission.

The absolute radiometric accuracy of Stokes fractions q=Q/I and u=U/I shall be better than 0.002 (G) / 0.005 (T). This specific requirement on Stokes fractions q and u is essential to meet the user requirements on single scattering albedo and refractive index;

The relative directional radiometric accuracy shall be better than 1% in order to achieve user requirements for aerosol optical thickness.

Retrievals from the 3MI should be supported by measurements of cloud height in the thermal infrared spectral range. This information is essential for cloud screening and may be used to perform simultaneous retrievals of aerosol and cloud properties for scenes with an elevated aerosol layer above the cloud top, and for partly clouded pixels with small cloud fractions.

6.6.1 LEO-3MI, Geometric Requirements

MR-LEO-3MI-10. For any spatial sample at least 10 (T) / 14 (G) directional measurements shall be acquired near simultaneously over the accessible directional space with an angular sampling in the order of 10°. The single scattering angles that are sampled shall cover the range from 90°-140° (G) / 100°-130° (T).

MR-LEO-3MI-20. The SSD shall be less than 4 km at SSP. MR-LEO-3MI-30. The spatial co-registration between polarised components and

different viewing angles shall be better than 10% SSD. The spatial co-

/166

S-4/-5 MRTD



registration between different spectral channels shall be better than 20% SSD.

6.6.2 LEO-3MI, Spectral Requirements

MR-LEO-3MI-40. The 3MI shall measure total radiance in the non-polarised spectral channels and polarized radiance in the polarised spectral channels, listed in Table 6.17.

Table 6.17. LEO-3MI channel specification in terms of centre wavelength, FWHM, polarization capability, and primary use.

Channel Centre Wavelength (μm)

FWHM (μm)

Polarization Primary Use

3MI-2b 0.410 0.02 Y Absorbing aerosol and ash cloud monitoring

3MI-3 0.443 0.02 Y Aerosol absorption and height indicators

3MI-4 0.490 0.02 Y Aerosol, surface albedo, cloud reflectance, cloud optical depth

3MI-5 0.555 0.02 Y Surface albedo 3MI-6 0.670 0.02 Y Aerosol properties 3MI-7 0.763 0.01 N Cloud and aerosol height

3MI-8 0.765 (T) 0.04 (T)

N Cloud and aerosol height 0.754 (B) 0.0075 (B)

3MI-9 0.865 0.04 Y Vegetation, aerosol, clouds, surface features

3MI-9a 0.910 0.02 N Water vapour, atmospheric correction

3MI-10 1.370 0.04 Y Cirrus clouds, water vapour imagery

3MI-11 1.650 0.04 Y Ground characterization for aerosol inversion

3MI-12 2.130 0.04 Y Cloud microphysics at cloud top, Vegetation, fire (effects) Ground characterization for aerosol inversion

MR-LEO-3MI-50. The spectral response of the channels shall be known, with 0.5 nm sampling, to 1% of their peak response at any wavelength.

MR-LEO-3MI-60. The spectral response stability shall be better than 0.1 nm.

6.6.3 LEO-3MI, Radiometric Requirements

MR-LEO-3MI-70. The Signal-to-noise ratio (SNR) shall be better than the values specified in Table 6.18, for the reference signal Lref specified in Table 6.18.

/166

S-4/-5 MRTD



Table 6.18. SNR requirements and reference signal per 3MI channel. Channel SNR Lref [W/m2/sr/μm] 3MI-2b 200 46.64

3MI-3 200 56.61

3MI-4 200 57.74

3MI-5 200 55.24

3MI-6 200 44.57

3MI-7 200 36.11

3MI-8 200 36.58

3MI-9 200 28.20

3MI-9a 200 25.18

3MI-10 200 10.73

3MI-11 200 6.80

3MI-12 200 2.50

MR-LEO-3MI-80. The absolute radiometric accuracy shall be better than 2% (G) / 3% (B)

/ 5% (T) for total radiance.

6.6.4 LEO-3MI, Auxiliary Data Requirements

Retrievals from the 3MI should be supported by measurements of cloud height in the thermal infrared spectral range. This information is essential for cloud screening and may be used to perform simultaneous retrievals of aerosol and cloud properties for scenes with an elevated aerosol layer above the cloud top, and for partly clouded pixels with small cloud fractions.

/166

S-4/-5 MRTD



6.7 Sentinel-5 Precursor (S5P)

MR-LEO-S5P-10. The payload of Sentinel-5 Precursor (S5P) shall satisfy the requirements provided above for the Sentinel-5 UV-VIS-NIR-SWIR spectrometers with the following exceptions:

the time frame of the precursor mission being 2017-2024 (see MR-SYSTEM-60);

the precursor shall include the UV-VIS-NIR and the SWIR-3 spectral bands. Considering the transitional nature of the precursor and the availability of TIR data - albeit with reduced quality - from IASI, it is acceptable not to consider the Sentinel-5 TIR instrument on the precursor.

auxiliary data from the VII and 3MI will not be available;

MR-LEO-S5P-20. The Sentinel-5 Precursor (S5P) shall be flown on a sun-synchronous low Earth orbit with an equator crossing mean local solar time of 13:30h.

The local time of the Sentinel-5 orbit is determined by the need for collocation with the IR sounder and the imager of the EPS-SG mission and will therefore be 9:30h. For S5P, an early afternoon orbit (13:30h equator crossing mean local solar time (MLST)) is required for the following reasons:

an afternoon orbit is favourable to pick up the main daily pollution signal which is the information needed for air quality forecast and protocol monitoring, and to support decisions on possible traffic or emission regulations for the next day;

an afternoon orbit will allow a start into observation of diurnal variability in synergy with Metop’s GOME-2 and IASI data; this would hardly be possible in the reverse case of a morning orbit and synergy with S-NPP/JPSS, due to the specification values of the relevant S-NPP/JPSS instruments which are not designed for chemical measurements in the troposphere;

early afternoon provides better observation conditions in solar channels than late afternoon, due to more favourable solar zenith angles;

in a 13:30h orbit, advantage can be taken by synergetic exploitation of simultaneous measurements of stratospheric ozone profiles by OMPS on the S-NPP/JPSS platform and of imager data of VIIRS on S-NPP/JPSS. The time delay between observations of a scene by S-NPP/JPSS and S5P is recommended to be less than 15 seconds (goal) / 5 minutes (threshold) (TBC).

an early afternoon orbit will be most suitable for continuation of the long term record of short-lived tropospheric species from OMI, which has been started in 2004, at 13:45 equator crossing MLST.

/166

S-4/-5 MRTD



7 LEVEL-1B PRODUCT REQUIREMENTS FOR GEO MISSION (SENTINEL-4)

The GEO mission Sentinel-4 shall contain UVN and TIR sounding capability according to requirements MR-SYSTEM-10 and MR-SYSTEM-20 and the implementation scenario outlined in Section 5.2.1. Note that requirements MR-GEO-SYS-50, MR-GEO-SYS-60, MR-GEO-SYS-70, and MR-GEO-SYS-80 apply only to the UVN bands.

7.1 GEO-SYSTEM

Spatial coverage and scan scenario

In addition to Western and Central Europe, the highest priorities are to observe Eastern Europe and North Africa. Measurements with an OZA > 75° are considered less useful.

MR-GEO-SYS-5. During the nominal operational phase (i.e. after commissioning), the system shall provide an availability of the Level-1b data of at least 95%.

MR-GEO-SYS-10. The nominal geographical coverage area shall comprise the longitude range from 30°W to 45°E at 40°N and the latitude range from 25°N (G) / 30°N (T) to 65°N at 0°E and shall be rectangular in view angles; areas with OZA>75° are not required.

A stepwise shift of the geographical coverage area in north-south direction is performed to extend the range of useful latitudes southward, during times when high latitude areas are dark.

MR-GEO-SYS-20. It shall be possible to shift the geographical coverage area southwards by 10° latitude (with respect to the Northern Edge) in two steps of each 5° in fall, and northwards by 10° latitude in two steps of each 5° in spring, if the threshold latitude range from MR-GEO-SYS-10 is achieved. If the goal latitude range of the geographical coverage area (25°N – 65°N) is achieved the shift of the geographical coverage area shall be 5° only. The geographical coverage area shall be identical with the nominal geographical coverage area in summer.

MR-GEO-SYS-30. The observed area of an individual scan shall cover a subset of the geographical coverage area with the full latitude range. The longitudinal range corresponding to the nominal repeat cycle of 1 h shall be 55° wide at 40°N. The scan shall start at the most eastern longitude within the geographical coverage area where the condition SZA<92° (MR-SYSTEM-10) is satisfied at 40°N. The scan shall move from East to West. In the default scan scenario, the longitude range is defined by condition MR-SYSTEM-10 and is limited by the nominal revisit time (MR-SYSTEM-20).

MR-GEO-SYS-40. The scan shall be programmable in a flexible way within the geographical coverage area.

/166

S-4/-5 MRTD



Geolocation knowledge

For a proper allocation of sources and sinks, accurate geolocation knowledge of the observations is required. This requirement is stringent over land, and over ocean mainly the relative knowledge from one acquisition to another is required. The requirement is evaluated per UV, VIS, or NIR band.

MR-GEO-SYS-50. Over land, the geolocation knowledge shall be better than 0.1 / 0.2 SSD (G / T) on any area of 50 SSD by 50 SSD within the observed area. Landmarks may be used to satisfy the requirement if the performance can be demonstrated. Over ocean, the requirement is 1 SSD.

MR-GEO-SYS-60. Over ocean, the geolocation knowledge in the scan direction, relative to the neighbouring pixel, shall be better than 0.5 SSD.

Spatial co-registration

The retrieval of data requires a good spatial co-registration, both within a spectral band and between different bands.

MR-GEO-SYS-70. Intra-band spatial co-registration of each of band shall be better than 0.1 SSD.

MR-GEO-SYS-80. Inter-band spatial co-registration between the UV and the VIS band shall be better than 0.1 SSD. Inter-band spatial co-registration between the UV and the NIR band, and between the VIS and the NIR band shall be better than 0.2 (T) / 0.1 (G) SSD.

Temporal co-registration

For consistent sampling of the atmosphere, measurements of a scene by the UVN and TIR bands need to be made within a time shorter than atmospheric variability, in particular cloud cover.

Good temporal co-registration (20 minutes (T),1 minute (G)) between measurements in the UV-VIS-NIR and the TIR would add value to the UVN measurement.

Data delivery time

Fast delivery services are required for air quality forecast applications, in particular for fast decision processes (traffic restrictions etc.).

MR-GEO-SYS-100. Level-1b data with full spectral coverage for the large data assimilation centres and Level-2 data for other NRT users shall be delivered regularly one hour after sensing with full spatial coverage.

/166

S-4/-5 MRTD



7.2 UV-VIS-NIR Requirements (GEO-UVN)

Specifications in this section apply to the implementation of Sentinel-4 according to the requirements on spatial coverage and revisit time for GEO specified in MR-SYSTEM-20 (see Table 5.1).

7.2.1 GEO-UVN, Geometrical and Temporal Requirements

For air quality purposes a high spatial resolution is required for almost all wavelength ranges. Spatial resolution is assumed to vary with target position with constant solid angle.

MR-GEO-UVN-10. The SSD at 45°N and longitude of the satellite shall be smaller than or equal to 8 km. The SSD at other positions shall be derived by maintaining a regular sampling of the SSA.

MR-GEO-UVN-15. Radiance data with an SSD in scan direction smaller than 0.2 times the nominal SSD in scan direction shall be available on-ground for the spectral samples located in the spectral ranges 310-315 nm, 338-343 nm, 415-420 nm, 460-465 nm, 750-755 nm, 760-763 nm, and 770-775 nm. Note that all other geometric, spectral, and radiometric requirements apply to the nominal SSD defined by MR-GEO-UVN-010.

MR-GEO-UVN-20. The SSD in all spectral channels shall be identical within 5%.

MR-GEO-UVN-30. The integrated energy (IE) shall be larger than 70% over an area of dimension 1.47 SSD in scan direction by

1.13 SSD in across scan direction, 90% over an area of dimension 1.72 SSD in scan direction by

1.72 SSD in across scan direction.

Requirement MR-GEO-UVN-30 poses an upper limit to the width of the IPSF and thus constrains the spatial overlap of adjacent spatial samples. Requirement MR-GEO-UVN-35 poses a lower limit to the width of the IPSF in order to exclude spatial undersampling.

MR-GEO-UVN-35. The IPSF in all channels in the UVN shall be sufficiently broad to allow spatially contiguous observations (exact formulation TBD).

MR-GEO-UVN-37. The integrated energy over an area of dimension 1.0 SSD in scan direction by 1.0 SSD in across scan direction at a given geo-location shall not vary spectrally by more than ± 5 %.

MR-GEO-UVN-40. The temporal co-registration of measurements of the same spatial sample for different wavelengths within the UVN shall be better than 15 s.

/166

S-4/-5 MRTD



7.2.2 GEO-UVN, Spectral Requirements

Spectral Coverage

The UVN band is used to measure several trace gas species, and gain information on aerosols and clouds. In Table 7.1 these data products are listed together with the required spectral ranges and the associated priorities in the context of Copernicus service requirements (see Appendix 2 for traceability).

Table 7.1. GEO-UVN data products, spectral ranges, and priorities.

Level-2 data product Wavelength range [nm] Priority

Tropospheric ozone 305 – 330 1

Sulphur dioxide 308 – 325 1 (1)

Albedo 310 – 775 2

Total ozone 325 – 337 1

Aerosol(2) 336 – 340 2

Formaldehyde 337 – 360 1


360 – 400 2, gaps may be acceptable between 375 nm and 380 nm

and between 384 nm and 400nm

Aerosol(2) 400 – 430 2

Nitrogen dioxide 405 – 500 1

Glyoxal 430 – 460 2 (3)

Aerosol (2) 440 – 460 2

Cloud (O2-O2) 460 – 490 1 (4)

Cloud (O2-A) 750 – 775 1

Aerosol profile (O2-A band) 750 – 775 1 (4, 5)

(1) Only observed in strongly polluted areas (volcanic eruptions, major powerplants)

(2) Column averaged aerosol optical thickness, and aerosol type.

(3) Considered as a by-product rather than a species driving the requirements. It has a weak spectral signature, only temporal averages will be retrieved.

(4) ’Priority 1’ not as geophysical product itself, but for use in correcting retrievals of other products.

(5) Used for estimating the aerosol profile for air quality and climate applications (see Section 4.2.2 and 4.2.3).

Overall this leads to the spectral bands and corresponding performance requirements as specified in Table 7.2 for the TOA radiance and the solar irradiance measurements. Some considerations and remarks that were used for the selection of the GEO UVN bands:

The spectral range 270 – 305 nm is excluded. This spectral range is mainly sensitive to stratospheric ozone, which is not the main target of the hourly observation

/166

S-4/-5 MRTD



capability of Sentinel-4. It is assumed that information on stratospheric ozone will be available on a daily basis from Sentinel-5.

Concomitant H2O data are assumed to be available from the infrared sounder on MTG. Main purpose of the O2-A band (750 – 775 nm) at a high spectral resolution (0.06nm) is to estimate a mean aerosol layer height, as investigated by /Roz1994/, /Tim1994/, /Tim1995/, /Koo1997/, and many other publications. The aerosol layer height is important to quantitatively determine tropospheric trace gas concentrations under polluted conditions. Aerosol from pollution is mostly concentrated within the PBL [/Ans2002/, /Wan2002/], and hence the aerosol effect on trace gas retrievals is in that height region.

Cloud top height and optical thickness can be determined alternatively from low spectral resolution (approx. 0.5 nm) O2-A band measurements.

The SWIR band for CO measurements has been dropped, taking into account results of technical studies done in preparation of MTG.

MR-GEO-UVN-50. The UVN sounder shall cover the spectral bands according to the ranges as specified in Table 7.2.

MR-GEO-UVN-60. The spectral resolution shall be smaller than or equal to the values specified in Table 7.2.


MR-GEO-UVN-70. The spectral sampling ratio shall be larger than or equal to the values specified in Table 7.2.

Table 7.2. GEO-UVN spectral bands and spectral requirements.

Band ID

Spectral range [nm]

Spectral resolution [nm]

Spectral sampling

ratio

Priority

GEO-UV 305-400(1) 0.5 3 1

GEO-VIS 400-500 0.5 3 1

GEO-NIR 750-775 0.5 (T) / 0.2 (B) / 0.12 (G)(2) 3 1

(1) Gaps may be acceptable in the region 375-380 nm and 384-400 nm.

(2) The spectral resolution of 0.12 nm is insufficient for aerosol profile retrievals in sun-satellite geometries associated with observations around noon. A spectral resolution of 0.06 nm would be needed for aerosol profile retrievals at all times of the day.

Spectral accuracy and stability

Except for the NIR range, the instrument spectral calibration can be achieved using adequate on-ground and in-flight spectral calibration techniques, such as analysis of solar Fraunhofer lines and atmospheric absorption features.

/166

S-4/-5 MRTD



MR-GEO-UVN-80. The position of the spectral channel centres shall be known with an accuracy equal to or better than 0.01 SSI (UV-VIS) or 0.05 (NIR). Note that this knowledge requirement can be fulfilled by analysis of in-flight data relying on the specified spectral stability.

MR-GEO-UVN-90. The position of the spectral channels in the radiances shall not vary by more than 0.05 SSI within a day.

Spectral response function

For a spatially homogeneous scene the ISRF shall be known sufficiently accurately such that the following requirements are met for all applicable wavelengths and viewing angles. Spectral features of the ISRF with spectral scales as small as 0.1 times the FWHM shall be resolved.

MR-GEO-UVN-100. The ISRF shall be known to an accuracy of 1% of the peak value of the ISRF in the spectral range where the ISRF is at least 1% of the peak value. The FWHM of the ISRF shall be known to an accuracy of 1%.

MR-GEO-UVN-104. The shape of the ISRF shall be sufficiently peaked such that the integrated areas satisfy


MR-GEO-UVN-106. The skewness of the slit function shall be sufficiently small such that the integrated areas in the wavelengths ranges andsatisfy

9

10

)(ISRF

)(ISRF

10

9

2

1

d

d

.

The wavelength ranges andrefer to the parts of above and below the average of the wavelengths defined by the FWHM of the ISRF, respectively.

The wavelength range where MR-GEO-UVN-100 applies is chosen such that localized spectral stray light (ghosts) and spatial stray light do not dominate compliance. Outside this wavelength range the ISRF is constrained by stray light requirements. MR-GEO-UVN-104 and MR-GEO-UVN-106 exclude slit functions with a stronlgy skewed shape or with strongly pronounced wings.

7.2.3 GEO-UVN, Radiometric Requirements


The radiometric requirements are applicable in the dynamic range defined by the reference radiance scenarios L_GEO_dyn_bright and L_GEO_dyn_dark, except for requirements on instrument noise. Requirements on noise are specified in terms of SNR associated with the reference radiance scenario L_GEO_SNR. Straylight requirements are applicable to a contrast scenario defined by the L_GEO_straylight_bright and the

/166

S-4/-5 MRTD



L_GEO_straylight_dark scenario. All reference scenarios used for GEO (Table 7.4) are calculated for a standard atmosphere (US Standard 1976) assuming background aerosol, no clouds, no precipitation, and a sub-satellite point at 0°N, 0°E.

Table 7.3. GEO-UVN SNR requirements.

Wavelength [nm] SNR (1)

radiance irradiance

305 200 (G) / 160 (T) 1000 (T) / 8000 (G)

310 400 (G) / 320 (T) 1600 (T) / 8000 (G)

315 700 (G) / 630 (T) 2000 (T) / 10000 (G)

320 900 2700 (T) / 10000 (G)

350 1040 3120 (T) / 10000 (G)

400 1800 (G) / 1240 (T) -

450 1800 (G) / 1440 (T) 4320 (T) / 15000 (G)

500 1800 (G) / 1440 (T) 4320 (T) / 15000 (G)

775 1200/250 at 0.5nm (2)

700/130 at 0.2nm (2)

566/85 at 0.12nm (2)

3 (T) / 10 (G) times SNR of radiance in continuum

(1) The SNR values apply per spectral sampling element.

(2) SNR requirement in continuum / deepest absorption minimum at given spectral resolution

Table 7.4. GEO-UVN reference scenarios.

Scenario ID Location Local Time Albedo [%] UV, VIS, NIR

Altitude [km)

L_GEO_SNR 50N, 0E 23. Sep, 15:00 5, 5, 15 0

L_GEO_dyn_bright 30N, 0E 21. Jun, 12:00 95, 95, 95 6

L_GEO_dyn_dark 61N, 0E 23. Sep, 16:30 5, 5, 15 0

L_GEO_straylight_bright 50N, 0E 23. Sep, 15:00 60, 60 , 60 6

L_GEO_straylight_dark 50N, 0E 23. Sep, 15:00 5, 5, 15 0

/166

S-4/-5 MRTD



MR-GEO-UVN-110. For the radiance levels in the reference scenario L_GEO_SNR, the SNR of each spectral sample for reflectance measurements shall be larger than or equal to the values in Table 7.3. Note that this requirement applies per spectral sample. Requirement values between the specified spectral points shall be obtained by linear interpolation. Around the lines located at 393.5 nm, 397 nm, 431 nm and 486.3 nm, the SNR requirement applies to the radiance interpolated between the wavelength pairs [392.5 nm, 394.5 nm], [396 nm, 398 nm], [429.5 nm, 431.5 nm], and [485.5, 487.5 nm] respectively. Degraded performance may be acceptable in the region 375–380 nm and 384–400 nm.

MR-GEO-UVN-120. The absolute radiometric accuracy (ARA) of the Earth radiance and reflectance shall be better than 2% (G) / 3% (T). This error budget includes all instrumental errors, but excludes radiometric noise and errors in radiometric standards. The dynamic range of radiance or reflectance is defined at each channel by the signal (radiance or corresponding reflectance) range covered by the reference scenarios L_GEO_dyn_dark and L_GEO_dyn_bright (see Table 7.4). This requirement is applicable to polarized scenes with a degree of polarization of 100% with any polarization direction. This requirement is also applicable to spatial samples located in the dark zone outside of the transition zone in the contrast scene with the radiance levels of the scenarios L_GEO_straylight_dark and L_GEO_straylight_bright (Llow and Lhigh in Figure 6.3). The width of the transition zone is 2 SSD (G) /10 SSD(T).

MR-GEO-UVN-125. Radiometric errors in radiance and irradiance due to errors in radiometric standards shall be lower than 1% (TBC).

MR-GEO-UVN-130. The relative spatial radiometric accuracy (RSRA) of the (Earth) radiance and reflectance measurements shall be smaller than 0.25% (1σ). This requirement is applicable for scenes with linearly polarised TOA spectral radiance with a degree of polarisation (DOP) of 50%.

/166

S-4/-5 MRTD



Figure 7.1. Radiance spectra of the Sentinel-4 UVN reference scenarios L_GEO_SNR (labelled Lref), L_GEO_dyn_bright (labelled Lmax) and L_GEO_dyn_dark (labelled Lmin) are shown for the UV-VIS (top) and the NIR (bottom) spectral range.

Sentinel-4 TOA Spectral Radiances

1.E+09

1.E+10

1.E+11

1.E+12

1.E+13

1.E+14

1.E+15

300 320 340 360 380 400 420 440 460 480 500

Wavelength [nm]

Ra

dia

nce

[p

ho

ton

s/s

/nm

/cm

2/sr

]

Lmin

Lref

Lmax

Sentinel-4 TOA Spectral Radiances

1.E+09

1.E+10

1.E+11

1.E+12

1.E+13

1.E+14

1.E+15

750 755 760 765 770 775

Wavelength [nm]

Ra

dia

nce

[p

ho

ton

s/s

/nm

/cm

2/sr

]

Lmin

Lref

Lmax

/166

S-4/-5 MRTD



MR-GEO-UVN-140. The peak-to-peak relative spectral radiometric accuracy (RSRA) of the (Earth) reflectance measurements shall be better than specified hereafter. This requirement is applicable to polarized scenes with a degree of polarization of 100% with any polarization direction. This requirement is also applicable to spatial samples located in the dark zone outside of the transition zone in the contrast scene with the radiance levels of the scenarios L_GEO_straylight_dark and L_GEO_straylight_bright (Llow and Lhigh in Figure 6.3). The width of the transition zone is 2 SSD (G) / 10 SSD (T).

• 0.1% for any spectral window with a width between 0.5 - 3 nm and below 315 nm

• 0.5% over any spectral window with a width less than 0.5 nm and below 315 nm.

• 0.05% (T) / 0.01% (G) over any spectral window in the UV-VIS above 315 nm with a width between 0.5 - 3 nm

• 0.25% over any spectral window in the UV-VIS above 315 nm with a width less than 0.5 nm.

• 0.05% (T) / 0.01% (G) over any spectral window in the NIR with a width between 0.1 – 7.5 nm if the spectral resolution (FWHM) = 0.2 or 0.5 nm

• 0.5%(T) / 0.1% (G) over any spectral window in the NIR with a width between 0.05 – 7.5 nm if the spectral resolution (FWHM) = 0.12 nm

MR-GEO-UVN-150. The peak-to-peak relative spectral radiometric accuracy (RSRA) of the (Earth) reflectance measurements shall be smaller than 2% over any spectral range in the UVVIS of width 3 nm to 100 nm. In the NIR, this shall be smaller than 1% (for spectral resolution 0.2 or 0.5 nm) / 3% (for spectral resolution 0.12 nm) over any spectral window of width 7.5 nm to 100 nm. This requirement is applicable to polarized scenes with a degree of polarization of 100% with any polarization direction. This requirement is also applicable to spatial samples located in the dark zone outside of the transition zone in the contrast scene with the radiance levels of the scenarios L_GEO_straylight_dark and L_GEO_straylight_bright (Llow and Lhigh in Figure 6.3). The width of the transition zone is 2 SSD (G) / 10 SSD (T)


Regular solar measurements are needed for calibration of Earth reflectance and for spectral calibration using Fraunhofer lines.

MR-GEO-UVN-180. Sun irradiance shall be measured in the same spectral channels as specified for Earth observation.

/166

S-4/-5 MRTD



The SNR for the solar irradiance is specified for the reference irradiance. Between the spectral points specified, the values need to be interpolated. See Fig. 6.4 for an impression of the irradiance levels over the whole wavelength range.

MR-GEO-UVN-190. The SNR of each spectral channel for solar irradiance measurements shall be larger than or equal to the values specified in Table 7.3. If the goal SNR cannot be achieved for individual solar measurements, the SNR in irradiance shall be sufficiently large that irradiance spectra with the goal SNR can be generated by co-adding consecutive solar measurements.

MR-GEO-UVN-200. The absolute radiometric accuracy (ARA) of the solar irradiance measurements shall be better than 3% (T) / 2 % (G) excluding the uncertainty of the reference solar spectrum, errors on radiometric standards and radiometric noise. This requirement applies at irradiance levels defined by the reference solar spectrum Es_GEO_uvn (see Fig. 6.4).

MR-GEO-UVN-205. The absolute radiometric accuracy (ARA) of the solar irradiance measurements shall be better than 6% (T) / 5% (G) of the reference solar reference spectrum in the continuum, including the uncertainty of the reference solar spectrum but excluding errors on radiometric standards and radiometric noise. This requirement does not apply at stronger solar Fraunhofer lines.

MR-GEO-UVN-210. The relative radiometric accuracy of the solar irradiance within the full range of elevation and azimuth incidence angles of the sun (irradiance goniometry) shall be known to an accuracy of 0.5%.


The light reflected by the Earth and the atmosphere can be strongly linearly polarised. The relative spectral variation of the polarization sensitivity should be characterized on-ground to an accuracy of (TBS).

MR-GEO-UVN-235. The polarisation sensitivity of each spectral channel shall be less than 0.005 (G) / 0.01 (T).

/166

S-4/-5 MRTD



7.3 Thermal Infrared (GEO-TIR)

The Sentinel-4 system shall make use of data from the TIR sounder on the MTG-S platform. The required spectral bands, together with the target species, are listed in Table 7.6. An extended wave number coverage is strongly preferred over narrower independent spectral windows. For instrument requirements we refer to the MTG-MRD /EUM2007/.

Table 7.6. GEO-TIR data products, and required spectral ranges.

Species Spectral Range [cm-1] H2O, O3, CO2, Surface, Clouds, Aerosols, HNO3, NH3,

CH3OH, HCOOH700 - 1210

H2O, CO, N2O 1600 - 2175

/166

S-4/-5 MRTD



8 DATA DELIVERY REQUIREMENTS

Application areas have been discussed in chapters 3 and 4. It should be noted that near real time delivery of data is required for all three environmental issues, supporting

UV dose and UV index forecast: near real time data delivery for this application implies that the data needs to be available to an operational modelling environment within a couple of hours after observation. In that case a significant part of today’s observations can still be used for the analysis on which the required forecast for tomorrow (etc.) will be based.

weather forecast and climate monitoring: The main users for near-real time data within the climate theme are the NWP centres. These centres need near-real-time information on numerous aspects of the land-atmosphere-ocean-cryosphere system that all contribute to the analysis of the atmosphere and therefore to the initial state on which the weather prediction is based, and on which climate monitoring relies. Atmospheric composition is one of the key elements for the monitoring of the climate system.

air quality forecast: An important requirement for health and safety is further near-real time source detection and attribution of the emissions of aerosols and aerosol and ozone precursors (NO2, SO2, and CO). Additional information on methane (CH4), water vapour (H2O), formaldehyde (CH2O) as well as the UV-VIS photolysis rates is important for the forecasting of the photochemical activity. These observations are needed to constrain the chemical conversion rates and help to determine the atmospheric residence time of pollutants. Finally, near real time data delivery for this application implies that the data needs to be available to an operational modelling environment within a couple of hours after observation. In that case a significant part of today’s observations can still be used for the analysis on which the required forecast for tomorrow (etc.) will be based. It should be noted that current practice of data time handling at ECMWF is not favourable for Air Quality forecasts. Data are collected twice a day (till 3 am and 3 pm) to provide forecasts in the morning and evening. For Air Quality forecasts it would likely make sense to include also the late afternoon observations of today in the Air Quality forecast for tomorrow, which should be available to operational agencies in the very early morning of the day to come.

Quantitative delivery time requirements are given in Chapters 6 and 7.

The distribution baseline documents for Sentinel-4 /S4DB/ and for Sentinel-5 /S5DB/ give an overview of the distribution mechanisms for the mission products that will be generated at the EUMETSAT Central Application Facility.

/166

S-4/-5 MRTD



9 SYNERGIES AND INTERNATIONAL CONTEXT

The only meaningful space-borne complement to Sentinel-4 and -5 can be another mission supporting operational applications. Research missions cannot be considered as reliable providers of long-term continuous data as needed for the Sentinel system /GAS2009/. Their data might be exploited on opportunistic basis - however, there are no firm plans for relevant research missions in the post-Envisat timeframe.

As outlined in Section 4.3, the Eumetsat and NOAA operational weather satellites include instruments for atmospheric composition measurements, and also provide relevant auxiliary information, such as temperature and water vapour profiles and imager data informing on cloud coverage and aerosol abundance. Section 5.2.1 explains the current scenario of integration of Sentinel-4 and -5 into this system and the resulting synergies.

In addition to the meteorological system, there will be aerosol data available from Sentinel-3, albeit at different local time and not collocated.

/166

S-4/-5 MRTD



10 REFERENCE DOCUMENTS

/Ans2002/ Ansmann, A., Wandlinger, U., Wiedensohler, A., Leiterer, U., Lindeberg Aerosol Charactersisation Experiment 1998 (LACE 98): Overview, JGR 107 D21 (2002) 8129, doi:10.1029/2000JD000233, 2002.

/Bar2004/ Barrie, L. and Langen, J. (Eds.) (2004). An Integrated Global Atmospheric Chemistry Observation Theme for the IGOS Partnership (IGACO), ESA SP-1282, WMO GAW No. 159 e.g. from ftp://ftp.wmo.int/Documents/PublicWeb/arep/gaw/gaw159.pdf

/Bar2005/ Barret, B., S. Turquety, D. Hurtman, C. Clerbaux, J. Hadji-Lazao, I. Bey, M. Auvray, P.F. Coheur, Global carbon monoxide vertical distributions from spaceborne high-resolution FTIR nadir measurements, ACPD, 5, 4599 – 4639, 2005.

/Bee2001/ de Beek, R, M. Vountas, V. Rozanov, A. Richter, J. Burrows, The Ring Effect in the cloudy atmosphere, Geophys. Res. Lett., 28, 2001.

/Ber2003/ Bergamaschi et al. (ed), Inverse modelling of national and EU greenhouse gas emission inventories – Report of the workshop “Inverse modelling for potential verification of national and EU bottom-up GHG inventories” under the mandate of the Monitoring Mechanism Committee WG-1, 23-24, JRC, Ispra, pp.144, EUR 21099 EN / ISBN 92-894-7455-6, October 2003.

/Bre2003/ F. M. Bréon et al, The Potential of Remote Sensing to Contribute to the Quantification of Anthropogenic Emissions in the Frame of the Kyoto Protocol, Final Report and Technical Notes of ESA Study Contract no. 15427/01/NL/MM, May 2003.

/Cha2010/ K. Chance, R.L.Kurucz, An improved high-resolution solar reference spectrum for earth’s atmosphere measurements in the ultraviolet, visible, and near infrared, Journal of Quantitative Spectroscopy & Radiative Transfer, 111, 1289–1295 , 2010.

/CLRTAP/ United Nations Economic Commission for Europe, Convention on Long-Range Transboundary Air Pollution, http://www.unece.org/env/lrtap/

/Din 2004/ van Dingenen et al., A European Aerosol Phenomenology – 1: Physical Characteristics of Particulate Matter at Kerbside, Rural, Urban and Background Sites in Europe, Atmos Environ., 38, 2561-2577, 2004.

/Dob2008/ M. Dobber, R. Voors, R. Dirksen, Q. Kleipool, and P. Levelt, Solar Physics volume 249, no. 2, 281-291, DOI 10.1007/s11207-008-9187-7, June 2008.

/ESA2001/ ACECHEM – Atmospheric Composition Explorer for Chemistry and Climate Interaction, Report for Assessment, ESA SP-1257(4), ISBN 92-9092-628-7, September 2001.

/ESA2007/ GMES Sentinels 4 and 5 Mission Requirements Document, EOP-SMA/1507/JL-dr.

/ESA2012/ Procedure for Mission Requirements Management, QMS-PR-MMAN-2050-EOP.

/EU2007/ European Commission, FP7 Cooperation Work Programme Space, C(2007)2460, June 2007.

/EUM2003a/ EUMETSAT position paper on Observation Requirements for Now Casting and Very Short Range Forecasting in 2015-2025. B W Golding, S Senesi, K. Browning, B Bizzarri, W Benesch, D Rosenfeld, V Levizzani, H Roesli, U Platt, T E Nordeng, J T Carmona, P Ambrosetti, P Pagano, M Kurz. VII.02 05/12/2003, 28 February 2003.

/EUM2003b/ Geo-stationary Satellite Observations for Monitoring Atmospheric Composition and Chemistry Applications, by Jos Lelieveld, Mainz, January 2003. EUMETSAT study for Meteosat Third Generation 2015-2025.

/166

S-4/-5 MRTD



/EUM2006/ Kelder, H., et al., Requirements for operational atmospheric chemistry monitoring in the post-EPS timeframe beyond 2020, Position paper on post-EPS atmospheric chemistry data user requirements, issue 0 draft H, 6 March 2006.

/EUM2007/ Eumetsat, MTG Mission Requirements Document, v2C, Dec. 2007.

/EUM2013/ Eumetsat, EPS-SG End User Requirements Document [EURD], v3C, June 2013.

/GAS2006/ GMES Atmospheric Service Orientation Paper, l, Nov. 2006.

/GAS2009/ GMES Atmospheric Service Implementation Group, Space Observation Infrastructure for the GMES Atmosphere Service – Summary for the Implementation Group: Conclusions and Recommendations, April 2009.

/GAT2004/ GMES-GATO Strategy Report. Global Monitoring for Environment and Security-Global ATmospheric Observations, March 2004.

/GCO2010/ GCOS Implementation Plan for the Global Observing System for Climate in Support of the UNFCCC, GCOS-138, (GOOS-184, GTOS-76, WMO-TD/No. 1523) http://www.wmo.int/pages/prog/gcos/Publications/gcos-138.pdf

/GEO2004/ Group on Earth Observation (GEO), Report of the Subgroup on User Requirements and Outreach, 22 March 2004.

/Hol2008/ Hollingsworth, A., et al., Toward a monitoring and forecasting system for atmospheric composition. The GEMS Project. Bulletin of the American Meteorological Society, 89, 1147 - 1164, doi:10.1175/2008BAMS2355.1, 2002.

/Hun2010/ Hungershoefer, K., F.-M. Breon, P. Peylin, F. Chevallier, P. Rayner1, A. Klonecki, S. Houweling, and J. Marshall, Evaluation of various observing systems for the global monitoring of CO2 surface fluxes, Atmos. Chem. Phys., 10, 10503–10520, 2010.

/Kel2005/ Kelder, H., et al., Operational Atmospheric Chemistry Monitoring Missions, Final Report and Technical Notes of ESA Study, Contract number 17237/03/NL/GS, October 2005. also http://www.knmi.nl/capacity/

/Ker2001/ Kerridge, B. et al, Definition of Mission Objectives and Observational Requirements for an Atmospheric Chemistry Explorer Mission, Final Report of ESA contract 13048/98/NL/GD. April 2001.

/Koo1997/ Koppers, G.A.A., Jansson, J., Murtagh, D.P., Aerosol optical thickness retrieval from GOME data in the oxygen A-band, ESA EP. Proceedings of the 3rd ERS symposium, Florence, Italy, 1997.

/Lev2009/ Levelt, P. et al., CAMELOT (Observation Techniques and Mission Concepts for Atmospheric Chemistry), Final Report of the ESA study, Contract nr. 20533/06/NL/HE, 2009.

/Liu2005/ Liu, X., K. Chance, C.E. Sioris, R.J.D. Spurr, T.P. Kurosu, R.V. Martin, M.J. Newchurch, Ozone Profile and Tropospheric Ozone Retrievals from Global Ozone Monitoring Experiment: Algorithm Description and Validation, JGR, vol. 110, noD20, pp. D20307.1-D20307.19 (1 p.1/4), 2005.

/Mon1987/ UNEP Ozone Secretariat, The Montreal Protocol on Substances that Deplete the Ozone Layer as either adjusted and/or amended in London 1990, Copenhagen 1992, Vienna 1995, Montreal 1997, Beijing 1999, 2000, ISBN 92-807-1888-6, also available at http://ozone.unep.org/pdfs/Montreal-Protocol2000.pdf

/Mun1998/ Munro, R., R. Siddans, W. J. Reburn, and B. J. Kerridge, Direct measurement of tropospheric ozone distributions from space, Nature, 392, March, 1998.

/ON2009/ TRAQ – Performance Analysis and Requirements Consolidation, Final Report of the ESA study, contract nr. 21509/07/NL/CB.

/166

S-4/-5 MRTD



/PRO2009/ Protocol Monitoring for the GMES Service Element, Atmosphere Service, Final Report, 2009. http://www.gse-promote.org/docs/FinalReport_WebVersion.pdf

/Roz1994/ Rozanov, V.V., Timofeev, Yu.M., Calculation of vertical profiles of atmospheric aerosol optical characteristics using satellite measurements of outgoing radiation in 0.76 micrometer absorption band. (in Russian) Fizika Atmosphery i Okeana, 30, 779-785. 1994.

/S4DB/ Copernicus Sentinel-4 ESA-EUM Product Distribution Baseline (S4HQ), ESA reference: S4-ESA-SY-LI-0142, EUMETSAT reference: EUM/MTG/SPE/12/0767, Issue 1E, 26 August 2016.

/S5DB/ Copernicus Sentinel-5 Data Distribution Baseline, S5-LI-ESA-SY-0085, Issue 1, Revision 2, 29 Sept 2016.

/Thu2003/ Thuillier, G., Hersé, M., Labs, D., Foujols, T., Peetermans, W., Gillotay, D., Simon, P.C., Mandel, H., The Solar Spectral Irradiance from 200 to 2400 nm as Measured by the SOLSPEC Spectrometer from the Atlas and Eureca Missions, Solar Physics 214(1): 1-22; May 2003.

/Tim1995/ Timofeev, Yu.M., Vasilyev, A.A., Rozanov, V.V., Information content of the spectral measurements of the 0.76 µm O2 outgoing radiation with respect to the vertical aerosol optical properties. Adv. Space Res., 16, (10)91-(10)94, 1995.

/UN1992/ United Nations Framework Convention on Climate Change, 1992. http://unfccc.int/resource/docs/convkp/conveng.pdf

/UN1998/ United Nations, The Kyoto Protocol to the United Nations Framework Convention on Climate Change, 1998. http://unfccc.int/resource/docs/convkp/kpeng.pdf

/Vie2006/ UNEP Ozone Secretariat, The Vienna Convention for the Protection of the Ozone Layer, http://ozone.unep.org/Publications/VC_Handbook/Section_1_The_Vienna_Convention/index.shtml

/Wan2002/ Wandlinger, U., Müller, D., Böckmann, C., Althausen, D., Matthias, V., Bösenberg, J., Weiss, V., Fiebig, M., Wendisch, M., Stohl, A., Ansmann, A.: Optical and microphysical charactrization of biomass burning and industrial pollution aerosol from multi-wavelength lidar and aircraft measurements. JGR, 107, D21 (2002) 8125, doi:10.1029/2000JD000202, 2002.

/166

S-4/-5 MRTD



LIST OF ACRONYMS AND ABBREVIATIONS

ACE Atmospheric Chemistry Explorer ACECHEM Atmospheric Composition Explorer for CHEMistry and climate interaction ADM Atmospheric Dynamics Mission AOT Aerosol Optical Thickness AQ Air Quality BT Brightness Temperature CAFE Clean Air For Europe CAMELOT Composition of the Atmosphere: Mission concEpts and sentinel Observation Techniques CAPACITY Composition of the Atmosphere: Progress to Applications in the user Community CFC Chloro-Fluoro-Carbon CLRTAP Convention on the Long-Range Transboundary Air Pollution CRIS Cross-track Infrared Sounder EAP Environmental Action Programme EC European Commission ECE Economic Commission for Europe ECMWF European Centre for Medium-range Weather Forecast EEA European Environmental Agency EMEP Co-operative Programme for Monitoring and Evaluation of the Long-Range Transmission of Air pollutants in Europe EOS Earth Observing System EPS EUMETSAT Polar System ERS European Remote Sensing (satellite) EUMETSAT European Organisation for the Exploitation of Meteorological Satellites FOV Field-of-View FT Free Troposphere FTIR Fourier-Transform Infrared (instrument) FWHM Full Width at Half Maximum G Goal (requirement) GAS GMES Atmospheric Service GATO Global Atmospheric Observations GEMS Global and regional Earth-system (Atmosphere) Monitoring using Satellite and in-situ data GEO Geostationary Orbit GHG Greenhouse Gas GMES Global Monitoring for Environment and Security GOES Geostationary Operational Environmental Satellite GOME Global Ozone Monitoring Experiment (instrument) GPS Global Positioning System GRG Global Reactive Gases HCFC Halogenated Chloro-Fluoro-Carbon IASI Infrared Atmospheric Sounding Interferometer (instrument) ICAO International Civil Aviation Organisation IE Integrated Energy IFOV Instantaneous Field Of View IG Implementation Group IGACO Integrated Global Atmospheric Chemistry Observations IGOS Integrated Global Observing Strategy IPCC Intergovernmental Panel on Climate Change IPSF Instrument Point Spread Function IR Infrared ISRF Instrument Spectral Response Function ITCZ Inter-Tropical Convergence Zone

/166

S-4/-5 MRTD



JPSS Joint Polar Satellite System LEO Low Earth Orbit LIDAR Light Detection And Ranging LS Lower Stratosphere MACC Monitoring Atmospheric Composition and Climate METOP Meteorological Operational (satellite) MODTRAN Moderate Resolution Transmission Code MOPD Maximum Optical Path Difference MR Mission Requirement MRD Mission Requirements Document M Mesosphere MS Middle Stratosphere MSG Meteosat Second Generation (geostationary satellites) MTF Modulation Transfer Function MTG Meteosat Third Generation (geostationary satellites) NEdL Noise Equivalent differential Radiance NEdR Noise Equivalent differential Reflectance NEdT Noise Equivalent differential Temperature NIR Near Infrared NOAA National Oceanic and Atmospheric AdministrationNWP Numerical Weather Prediction ODS Ozone Depleting Substance OMI Ozone Monitoring Instrument OMPS Ozone Monitoring and Profiling Suite OPD Optical Path Difference OZA Observation Zenith Angle PAH Polycyclic Aromatic Hydrocarbon PAN Peroxy-Acetyl-Nitrate PBL Planetary Boundary Layer PSF Point Spread Function PM Particulate Matter POP Persistent Organic Pollutant PROMOTE Protocol Monitoring for the GMES Service Element: Atmosphere PSC Polar Stratospheric Cloud RAQ Regional Air Quality S/C Spacecraft S-NPP Suomi-National Polar-orbiting Partnership SG Second Generation SNR Signal-to-Noise–Ratio SSD Spatial Sampling Distance SSI Spectral Sampling Interval SSA Spatial Sampling Angle SSP Sub-Satellite Point ST Stratosphere-Troposphere SWIR Short-Wave Infrared (instrument) SZA Solar Zenith Angle T Threshold (requirement) TBC To Be Confirmed TBD To Be Determined TIR Thermal Infrared (instrument) TOA Top-of-Atmosphere TTL Tropical Tropopause Layer UN United Nations

/166

S-4/-5 MRTD



UNEP United Nations Environmental Programme UNFCCC United Nations Framework Convention on Climate Change US Upper Stratosphere UT Upper Troposphere UTLS Upper Troposphere / Lower Stratosphere UV Ultraviolet UVN UV-VIS-NIR (instrument) UVNS UV-VIS-NIR-SWIR (instrument) VAAC Volcanic Ash Advisory Center VIS Visible VOC Volatile Organic Compound WMO World Meteorological Organisation ZA Zonal Average (longitudinal average over a latitude band and time period)

/166

S-4/-5 MRTD



APPENDIX 1: REQUIREMENT TERMS, CONVENTIONS AND DEFINITIONS

Absolute Accuracy The absolute accuracy is the root mean square (RMS) difference between the measurement and truth including both random and bias errors. Error distributions are to be understood as Gaussian and the values given refer to one standard deviation, unless otherwise stated.

Additive radiometric error Radiometric bias I that is independent from the signal viz.:

a Iref

The parameter a defines the amplitude of the additive error relative to a reference signal Iref.

Band Width Width of spectral band.

Channel A channel is defined and delineated by its spectral boundaries.

Coverage The coverage is defined as the geographical region covered by a set of measurements.

Co-Registration See spatial or temporal co-registration.

Data Level The following definitions of data levels are used:

Observation Level-0: Raw data after restoration of the chronological data sequence for the instrument(s) operating in observation mode.

Level-1a: Level-0 data with corresponding radiometric, spectral and geometric (i.e. Earth location) correction and calibration computed and appended, but not applied.

Level-1b: As Level-1a data but radiometrically and spectrally calibrated and geo-located. Both radiance and irradiance data are comprised in the Level-1b data.

Level-1c: Level-1b data re-sampled to a specified rectangular grid.

Level-2: Earth located pixel values converted to geophysical parameters.

Level-3 – Level-2 data re-sampled to given spatial / temporal grid.

Level-4 – observed measurements combined with a model (e.g. via data assimilation or inverse modelling).

Dwell Time

/166

S-4/-5 MRTD



The dwell time is the time period required to acquire data for all spectral channels for a given spatial sample.

Effective Relative Spectral Radiometric Accuracy (ESRA)

The Effective Spectral Radiometric Accuracy (ESRA) is defined as the scalar product of the radiometric error vector ΔS with the gain-vector G for a given spectral window:

ESRA = ∑Gk .∆Sk

k

whereby k indicates the spectral channel within a given spectral window, for which ESRA is evaluated. The ESRA has the same dimension as the signal it is applicable to. The radiometric error ΔSk of a spectral channel k is defined as the systematic error of the detected signal

ΔSk = Smeas – Strue

where Strue is the true, error-free signal. The gain Gk is a dimensionless number quantifying the impact of the radiometric error of spectral channel k, on the accuracy of the Level-2 product.

Full Width at Half Maximum (FWHM) The full width at half maximum is the full width of a function at half-maximum points.

Gain Matrix The Gain matrix G is the generalized inverse of the linearized forward model K. The matrix G links spectral features y with retrieval bias x viz.:

x = G y,

where x is the retrieval state vector, and y is the reflectance spectrum.

Goal The term “goal” denotes a non-mandatory requirement, the implementation of which shall be studied to allow for an assessment of the system impacts. The implementation or not of the goal requirements will be decided by the Agency after proper analysis of the implications

Image An image is defined as the ensemble of spatial samples acquired for a given channel.

Instantantaneous Field of View (IFOV) Width of an area of sensitivity that is sampled, which represents an individual measurement.

Instrument Point Spread Function (IPSF)

/166

S-4/-5 MRTD



For a spatial sample i, observing in spectral channel/sample, j, with spectral centroid, λ0, a stable scene of spectrally integrated radiance Lij(λ0, x, y), the measured spectral radiance L’ij is given by:

dxdyyxLyxL ijijij ,,, 00 ,

where: Ψij(x,y) is the instrument Point Spread Function (PSF) for detector i in a channel j. The instrument PSF is the integral over dwell time of the instantaneous PSF, including optical effects, detector characteristics, electronic fluctuations and nominal pointing variation during the dwell time, but excluding effects due to unknown LOS instability. The spatial integral of Ψij(x,y) is normalised to 1.

Instrument Spectral Response Function (ISRF) The ISRF (or Φ) relates the radiometrically calibrated, spectrally integrated radiance L’(υ0) measured by a detector element i, in a spectral band j, with the spectral radiance L(υ) emanating from a spatially homogeneous scene. The ISRF Φij is normalised such that its spectral integral yields 1. It is defined by:

0

00 , dLL ijij ,

Integrated Energy (IE) The Energy, E(x,y), is the spatial integral of the IPSF (Ψ) of a detector element i and a spectral channel j over an area D of dimensions x × y centred around the IPSF barycentre position (x0,y0).

D

ijij dxdyyyxxdIE 00 , ,

IE is the ratio of the energy E measured by an instrument over an area D to the energy measured by the same instrument from the entire large and uniform scene.

Irradiance Irradiance is defined as the signal when the instrument points to the Sun.

Knowledge Requirement A requirement which is met when the parameter is known to within the specified range.

LOS Pointing Stability The line-of-sight (LOS) pointing stability is defined as the temporal variation of the position of the spatial sample within a given time period.

Modulation Transfer Function (MTF) The modulation transfer function describes the ratio of the image spatial frequency spectrum to the object spatial frequency spectrum and such defines the limit to which

/166

S-4/-5 MRTD



spatial frequencies within the object can be resolved. The MTF includes the optical transfer function as well as the rest of the measurement system’s transfer function.

Multiplicative radiometric error Radiometric bias I that is proportional to the signal viz.:

I = mI().

The parameter m defines the amplitude of the bias relative to the signal at each wavelength I(). Examples are, but are not limited to: radiometric corrections, non-linearity in the electronics or detectors, residual polarisation effects, spectral structures of all frequencies, ADC conversion, electronic conversion, pixel-to-pixel non-uniformity sensitivity, exposure time, potential electronic gain.

Nadir Angle Angle between nadir and actual instrument’s viewing direction.

Noise Equivalent Brightness Temperature (NEdT) See radiometric resolution.

Noise Equivalent Differential Radiance (NEdL) See radiometric resolution.

Nyquist frequency Nyquist frequency is given by 1/(2*SSD).

Observation Zenith Angle (OZA) Angle between the satellite viewing direction and the local zenith defined in the target reference frame (i.e., zenith – target – satellite).

Point Spread Function (PSF) See Instrument Point Spread Function.

Polarisation Sensitivity Assuming measurement of a stable, spatially uniform, linearly polarised Lambertian scene, the polarisation sensitivity is defined as

minmax

minmax

SS

SSP

,

where Smax and Smin are the maximum and minimum sample values, respectively, obtained when the polarisation is gradually rotated over 180°.

Principal Plane Plane defined by the local zenith and the sun direction.

Radiance Radiance is defined as the signal when the instrument points to Earth and observes radiation backscattered from the surface or atmosphere.

/166

S-4/-5 MRTD



PM Particulate matter

Radiometric Resolution (NEdL, NEdR, NEdT, SNR) or Radiometric Noise The ‘radiometric resolution’ or noise is the standard deviation of a series of radiance measurements for a specific channel on a stable and spatially uniform scene. For the thermal channels, the radiometric resolution is either specified in terms of noise-equivalent differential temperature (NEdT) associated with a reference (surface) temperature (usually 280 K unless otherwise specified), or as Noise Equivalent differential Radiance (NEdL). NEdL is the RMS deviation of the retrieved radiance associated to the samples of an image acquired over a stable and spatially uniform scene. For the solar channels, the radiometric noise is often expressed by the signal-to-noise-ratio (SNR) or the noise-equivalent reflectance difference (NEdR) for a given reference scenario.

SNR

LNEdL ,

SNR

RNEdR ,

)cos(.. ssER

L ,

dTdLNEdL

NEdT

where • Es is the TOA solar irradiance, • θs is the solar zenith angle, • R is the reflectance.

For thermal channels the radiance L is given by the Planck function for the corresponding temperature.

The definition applies to radiometrically calibrated data, which means that the contribution of the noise of the calibration coefficients is included. Error distributions are to be understood as Gaussian and the values given refer to one standard deviation, unless otherwise stated.

Reference Scene Radiances (UVN and SWIR) Four scenarios shall be considered to evaluate the scene radiance which are specified in the appropriate sections. For the solar channels the radiance L at TOA shall be estimated according to the following formula:

)cos(.. ssER

L

Where: L is the spectral radiance TOA [W.m-2.sr-1.nm-1], R is the Reflectance, Es is the sun extraterrestrial spectral irradiance given in [W.m-2.nm-1], s is the SZA.

The reference TOA spectral radiances for the four cases have been calculated with the radiative transfer model DAK v.2.3 at a resolution of 0.1 nm. The results are available as ASCII files.

Reflectance

/166

S-4/-5 MRTD



Reflectance R() is derived from measured Earth radiance Lmeas() and solar irradiance Emeas() as follows: measured solar irradiance and Earth radiance spectra are spectrally and radiometrically calibrated. Then the solar irradiance spectrum is interpolated to the wavelength grid of the radiance spectrum using the high-sampling interpolation method. Reflectance is hence defined by the ratio of calibrated Earth radiance Lcal() and calibrated and interpolated solar irradiance Ecalint() viz.:

)(

)()(

E

LR .

High sampling interpolation method The high-sampling interpolation method is used to interpolate measured solar irradiance spectra to the wavelength grid of measured radiance spectra. This method minimizes interpolation errors by exploiting prior information about the spectral structure of the solar spectrum using a reference solar spectrum with a high spectral resolution and sampling Ehs(Solar spectra Ehs_ori and Ehs_new are obtained by convolution of Ehs with the ISRF and sampling to the original wavelength grid (Ehs_ori) or the new wavelength grid (Ehs_new). The spectral sample j of the interpolated solar irradiance Enew is obtained from the nearest neighbor spectral sample k of the original solar irradiance Eori by

orihsk

newhsjori

knewj E

EEE _

_

.

Revisit time The revisit time is defined as the time elapsed between two consecutive measurements observing the same spatial sample.

Radiometric Stability The radiometric stability defines the range of variability of the measured radiances L(t), or brightness temperature Tb(t), allowed over a fixed time period τ, for a stable scene such that:

stability cradiometri100

k

jk

tL

tLtL, in [%] for the solar channels,

and stability cradiometri jbkb tTtT , in [K] for the thermal channels,

where tj , tk are any two fixed times within the specified time period τ.

Relative Spatial Radiometric Accuracy The relative spatial radiometric accuracy is the relative variation of the unknown systematic error of the samples acquired in an image.

Relative Spectral Radiometric Accuracy The relative spectral radiometric accuracy is the difference in radiometric bias between given spectral channels.

/166

S-4/-5 MRTD



Sample A radiometric measurement taken at some location.

Sigma The one sigma (1σ) specified in some requirements reflects, unless stated otherwise, the confidence on that requirement is 1σ.

Signal-to-Noise Ratio See radiometric resolution.

Solar Channels Channels with a centre wavelength shorter than 3.0 μm.

Spatial Co-Registration Maximum equivalent ground distance between the centres of all pairs of spatial samples acquired in two spectral channels and related to the same target on Earth.

Spatial Resolution The spatial resolution ΔXij of a detector i in a spectral channel j is defined by the dimensions of a rectangular area centred at the instrument PSF barycentre that contains 70% of the integrated energy.

Spatial Sample A spatial sample is a Level-1b measurement associated with instrument PSF. The centre of the spatial sample is the instrument PSF barycentre.

Spatial Sampling Distance (SSD) The Spatial Sampling Distance (SSD) is defined as the barycentre-to-barycentre distance between adjacent spatial samples on the Earth’s surface.

Spectral Band A spectral band is a contiguous sub-region of the whole spectral range covered by an instrument.

Spectral Channel A spectral channel is the defined by the ISRF. The centre of the spectral channel is the system ISRF spectral barycentre.

Spectral Resolution The spectral resolution, in Δλ or in Δν, is defined as the FWHM of the ISRF. In case of a Fourier Transform Spectrometer, a strongly apodised ISRF shall be assumed (1/OPDmax).

Spatial Sampling Angle The angle subtended by a spatial sampling distance as seen from the satellite.

Spectral Sampling Interval (SSI)

/166

S-4/-5 MRTD



The Spectral Sampling Interval is the spectral distance between the centroid of the ISRF of two adjacent channels.

Spectral Sampling Ratio Defined as the ratio between the spectral resolution and the spectral sampling interval.

Spectral stability Spectral stability is the temporal variation of the spectral position within a given time period. In case of a Fourier Transform Spectrometer, it is the maximum drift of spectral reference source during one interferogram duration.

Sub-Satellite Point (SSP) The sub-satellite point (SSP) is the point on the Earth’s surface that intersects the line between the satellite and the geocentric centre of the Earth.

Swath Width The swath width is defined as the length of area on the Earth’s surface, perpendicular to the satellite velocity, that is imaged during a given satellite pass.

Temporal Co-Registration Maximum time difference between the average moment of acquisition of all pairs of spatial samples acquired in two spectral channels and related to the same target on Earth.

Thermal channels Channels with a centre wavelength longer than 3.0 μm.

Unapodised spectral resolution

Unapodised spectral resolution -1 (cm )na has to be understood as max

0.6 na OPD

for a

Fourier transform spectrometer. OPDmax refers to the maximum Optical Path Difference of the Fourier transform spectrometer.

Self-apodisation In the case of a Fourier transform spectrometer, the self-apodisation effect shall be minimised. The FWHM of the instrument line shape (including self-apodisation but excluding any apodisation function applied in post-processing of the data) for any spatial resolution element within the FOV shall not exceed the unapodised spectral resolution by a factor larger than about 1.4 throughout the measured spectral range.

Unapodised radiometric noise The requirements on unapodized readiometric noise apply to spectra without external apodisation, but with self-apodisation.

Spectral sampling for TIR Spectral sampling is specified as = 0.5/OPDmax for a Fourier transform spectrometer. OPDmax refers to the maximum Optical Path Difference of the Fourier transform spectrometer.

/166

S-4/-5 MRTD

Date 19.05.2014 Issue 1 Rev 1


APPENDIX 2: COPERNICUS ATMOSPHERIC SERVICE REQUIREMENTS TRACEABILITY

Sentinel-4&5 traceability table of mission requirements to Copernicus service requirements (Chapter 3). The main characteristics of the Sentinel-5 mission basically reflect the requirements inherent to all data products and can be summarised by 1) high spatial resolution, 2) global coverage and 3) daily revisit time. With the Sentinel-4 the air quality related services are further supported in order to serve the need for higher temporal resolution on a regional scale. Products required to support the retrieval of other products are denoted with an (x) in lowercase.

Level-2 data product

Wavelength range

Sentinel-4

Sentinel-5

Spectral regime

Comment Services (as introduced in chapter 3)

GRG GRG GRG and RAQ

-GHG Aerosol

Promote-Strat. O3

Promote-UV

Promote-AQ

Promote-Climate

Promote-Volcanic

Stratospheric ozone (O3)

270–300 nm X X

UVB

Lower spectral and spatial resolution than other bands. Required to support the retrieval of tropospheric O3.

X

x

x

1030–1080 cm-1

X X TIR Less sensitive in stratosphere than UVB observations.

Tropospheric ozone (O3)

300–330 nm X X UVB For optimal sensitivity, the retrieval is performed in synergy with the TIR band.

X

X

1030–1080 cm-1

X X TIR For optimal sensitivity, the retrieval is performed in synergy with the UVB band.

/166

S-4/-5 MRTD



Sulphur dioxide (SO2)

300–325 nm X X UVB X X

Volcanic SO2 300–325 nm X X UVB X X

1280–1360 cm-1

X X TIR Priority 2

Total ozone (O3) 325–337 nm X X UVA X X

Aerosol several small bands from

336–775 nm

X X UVA-VIS Priority 2.Sentinel-3 supplementary aerosol data

X X

Formaldehyde (CH2O)

337–360 nm X X UVA X

Bromine monoxide (BrO)

345–360 nm X X UVA Priority 2 / by-product

CAPACITY theme/category: A3

X


360–400 nm X X UVA Priority 2.Sentinel-3 supplementary aerosol data.Required to support the retrieval of most products

x x X x X

Nitrogen dioxide (NO2)

405–500 nm X X VIS X

Glyoxal (CHOCHO)

430–460 nm X X VIS Priority 2 / by-product X

Water vapour (H2O)

710–750 nm X NIR Priority 2 X

/166

S-4/-5 MRTD



Cloud information (effective scattering height, covered fraction) from O4 and O2-A band

460–490 nm,

710–775 nm

X X VIS-NIR Required to support the retrieval of most products

x x x x x

Methane (CH4)

1.6 and 2.3 µm band

X SWIR For optimal sensitivity, the retrieval is performed in synergy with the TIR band.

X

1280–1360 cm-1

X TIR For optimal sensitivity, the retrieval is performed in synergy with the SWIR band.

Carbon monoxide (CO)

2.3 µm band X SWIR For optimal sensitivity, the retrieval is performed in synergy with the TIR band.

X

X

2140–2180 cm-1

X X TIR In case of Sentinel-4, for optimal sensitivity, the retrieval is performed in synergy with the SWIR band.

Nitric acid (HNO3)

860–900 cm-1, 1280–1360

cm-1

X X TIR Priority 2 X X

/166

S-4/-5 MRTD



APPENDIX 3: DATA REQUIREMENTS TABLES

The Level-2 data requirements presented in this Appendix are taken from the CAPACITY study \Kel2005\

A1 Theme:Category:

Ozone LayerProtocol Monitoring

Requirement Data Product

Driver

Height Range

Horizontal resolution (km)

Vertical resolution (km)

Revisit Time (hours)

Uncertainty

O3 Trend Total column 50 / 100 -- 24 / 24*3 3%

Spectral UV surface albedo Surface UV Trend Surface 10 / 50 -- 24 / 24*3 0.1

Spectral UV solar irradiance Surface UV Trend Top of Atmosphere -- -- Daily / Monthly 2%

UV Aerosol Optical Depth Surface UV Trend Total column 10 / 50 -- 24 / 24*3 0.1

UV Aerosol Absorption

Optical Depth

Surface UV Trend Total column 10 / 50 -- 24 / 24*3 0.02

A2 Theme:Category:

Ozone LayerNear-Real Time Data

/166

S-4/-5 MRTD




Driver

Height Range




Uncertainty

O3 Ozone and UV

Forecast

UT

LS

MS

US+M

Troph.

column

Total column

20 / 100

50 / 100

100 / 200

100 / 200

10 / 50

50 / 100

0.5 / 2

0.5 / 2

2 / 3

3 / 5

--

--

6 / 24*3

6 / 24*3

6 / 24*3

12 / 24*7

6 / 24*3

6 / 24*3

20%

20%

20%

20%

20%

5%

Spectral UV surface albedo UV Forecast Surface 10 / 50 -- 6 / 24*3 0.1

Spectral UV solar irradiance UV Forecast Top of

Atmosphere

-- -- Daily / Monthly 2%

UV Aerosol Optical Depth UV Forecast Total column 10 / 50 -- 6 / 24*3 0.1

UV Aerosol Absorption

Optical Depth

UV Forecast Total column 10 / 50 -- 6 / 24*3 0.02

Strat. Aerosol Optical Depth Ozone loss LS

MS

Stratosphere

50 / 100

50 / 200

50 / 200

0.5 / 2

1 / 3

--

6 / 24*3

12 / 24*7

6 / 24*7

0.05

0.05

0.05

ClO Ozone loss LS

MS

Stratosphere

50 / 200

100 / 200

50 / 200

2 / part. column

2 / part. column

--

24 / 24*7

24 / 24*7

24 / 24*7

50%

50%

50%

NO2 Ozone loss LS

MS

Stratosphere

50 / 200

100 / 200

50 / 200

2 / part. column

2 / part. column

--

24 / 24*7

24 / 24*7

24 / 24*7

20%

20%

20%

PSC occurrence Ozone loss LS 50 / 100 0.5 / 2 6 / 24*3 < 10% mis-

assignments

SF6 Tracer LS

MS

50 / 200

100 / 200

1 / 2

2 / 3

6 / 24*3

12 / 24*7

10%

10%

CO2 (as tracer alternative to

SF6)

Tracer; Radiation

budget

LS

MS

50 / 200

100 / 200

1 / 2

2 / 3

6 / 24*3

12 / 24*7

10%

10%

/166

S-4/-5 MRTD



H2O Radiation budget;

ST exchange

UT

LS

MS

20 / 100

50 / 100

100 / 200

0.5 / 2

1 / 2

2 / 3

6 / 24*3

6 / 24*3

12 / 24*7

20%

20%

20%

N2O (as tracer alternative to

SF6)

Tracer; Radiation

budget

LS

MS

US

50 / 100

50 / 200

50 / 200

1 / 2

2 / 3

3 / 5

6 / 24*3

12 / 24*7

12 / 24*7

20%

20%

20%

CH4 (as tracer alternative to

SF6)

Tracer; Radiation

budget

LS

MS

50 / 200

100 / 200

1 / 2

2 / 3

6 / 24*3

12 / 24*7

20%

20%

HCl ST exchange LS Co-located with O3 Co-located with O3 Co-located with O3 20%

HNO3 ST exchange LS Co-located with O3 Co-located with O3 Co-located with O3 20%

CO ST exchange UT+LS Co-located with O3 Co-located with O3 Co-located with O3 20%

/166

S-4/-5 MRTD



A3 Theme:Category:

Ozone LayerAssessment


Driver

Height Range




Uncertainty

O3 Ozone and UV Trend; Ozone

loss; Surface UV, Ozone-

Climate interaction

UT

LS

MS

US+M

Troph. column

Total column

20 / 100

50 / 100

100 / 200

100 / 200

10 / 50

50 / 100

1 / 3

1 / 3

2 / 3

3 / 5

--

--

6 / 24*3

6 / 24*3

6 / 24*3

6 / 24*7

6 / 24*3

6 / 24*3

20%

10%

20%

20%

20%

10%

Spectral UV surface albedo Surface UV Surface 10 / 50 -- 6 / 24*3 0.1

UV Aerosol Optical Depth Surface UV Total column 10 / 50 -- 6 / 24*3 0.1

UV Aerosol Absorpton

Optical Depth

Surface UV Total column 10 / 50 -- 6 / 24*3 0.02

Spectral UV solar irradiance Surface UV Top of Atmosphere -- -- monthly 2%

H2O Ozone-Climate interaction LS

MS

US

Stratosphere

50 / 100

100 / 200

100 / 200

50 / 200

1 / 3

2 / 3

3 / 5

--

12 / 24*3

12 / 24*7

12 / 24*7

12 / 24*7

15% (1000 km)

15% (1000 km)

15% (1000 km)

15% (1000 km)

N2O Ozone-Climate interaction LS

MS

US

Stratosphere

50 / 100

100 / 200

100 / 200

50 / 200

1 / 3

2 / 3

3 / 5

--

12 / 24*3

12 / 24*7

12 / 24*7

12 / 24*7

10% (ZA)

10% (ZA)

10% (ZA)

10% (ZA)

CH4 Ozone-Climate interaction LS

MS

US

Stratosphere

50 / 100

100 / 200

100 / 200

50 / 200

1 / 3

2 / 3

3 / 5

--

12 / 24*3

12 / 24*7

12 / 24*7

12 / 24*7

10% (ZA)

10% (ZA)

10% (ZA)

10% (ZA)

HN O3 Ozone Trend; Dinitrification LS 50 / 100 1 / 3 12 / 24*3 30% (1000 km)

/166

S-4/-5 MRTD



MS

Stratosphere

100 / 200

50 / 200

2 / 3

--

12 / 24*7

12 / 24*7

30% (1000 km)

30% (1000 km)

CFC-11 Ozone trend LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

5% (ZA)

5% (ZA)

5% (ZA)

CFC-12 Ozone trend LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

5% (ZA)

5% (ZA)

5% (ZA)

HCFC-22 Ozone trend LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

20% (ZA)

20% (ZA)

20% (ZA)

ClO Ozone loss LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

30% (1000 km)

30% (1000 km)

30% (1000 km)

BrO Ozone loss LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

30% (1000 km)

30% (1000 km)

30% (1000 km)

NO2 Ozone loss LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

30% (1000 km)

30% (1000 km)

30% (1000 km)

Aerosol surface density Ozone loss LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

100% (1000 km)

100% (1000 km)

100% (1000 km)

PSC occurrence Ozone loss LS 50 / 100 1 / 3 12 / 24*3 < 10% mis-

assignments

HCl Chlorine trend LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

30% (1000 km)

30% (1000 km)

30% (1000 km)

ClONO2 Chlorine trend LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

30% (1000 km)

30% (1000 km)

30% (1000 km)

/166

S-4/-5 MRTD



HBr Bromine trend LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

30% (1000 km)

30% (1000 km)

30% (1000 km)

BrONO2 Bromine trend LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

30%

30%

30%

CH3Cl Bromine trend LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

30%

30%

30%

CH3Br Bromine trend LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

5% (ZA)

5% (ZA)

5% (ZA)

SO2 enhanced Ozone loss LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

50%

50%

50%

Volcanic aerosol Ozone loss LS

MS

Stratosphere

50 / 100

100 / 200

50 / 200

1 / 3

2 / 3

--

12 / 24*3

12 / 24*7

12 / 24*7

< 10% mis-

assignments

/166

S-4/-5 MRTD



B1 Theme:Category:

Air QualityProtocol Monitoring


Driver

Height Range




Uncertainty

O3 Interpolation of Surface

network; Boundary

condition; UV actinic fluxes

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

50 / 100

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

24 / 24*3

10%

20%

25%

3%

NO2 Interpolation of Surface

network; Emissions;

Boundary condition

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

10%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

CO Interpolation of Surface

network; Emissions;

Boundary condition

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

25%

25%

SO2 Interpolation of Surface

network; Emissions;

Boundary condition

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

CH2O Interpolation of Surface

network; VOC Emissions;

Boundary condition

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

Aerosol OD Interpolation of Surface PBL 5 / 20 -- 0.5 / 2 0.05

/166

S-4/-5 MRTD



network; Emissions;

Boundary condition; UV

actinic fluxes

FT

Tropospheric Column

Total Column

5 / 50

5 / 20

5 / 20

--

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.05

0.05

0.05

Aerosol Type Translation Aerosol OD to

PM surface concentrations

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

--

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

< 10% mis-

assignments

< 10% mis-

assignments

< 10% mis-

assignments

< 10% mis-

assignments

/166

S-4/-5 MRTD



B2 Theme:Category:

Air QualityNear-Real Time


Driver

Height Range




Uncertainty

O3 Air Quality Forecast; UV

actinic fluxes

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

50 / 100

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

12 / 24*3

10%

20%

25%

5%

NO2 Air Quality Forecast PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

10%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

CO Air Quality Forecast PBL

FT

Ttropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

25%

25%

Aerosol OD Air Quality Forecast; UV

actinic fluxes

PBL

FT


Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

--

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

0.05

0.05

0.05

0.05

Aerosol Type Air Quality Forecast PBL

FT


Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

--

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

< 10% mis-

assignments

< 10% mis-

assignments

< 10% mis-

assignments

< 10% mis-

assignments

/166

S-4/-5 MRTD



H2O Air Quality Forecast PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

10%

20%

10%

10%

SO2 Air Quality Forecast PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

CH2O Air Quality Forecast PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

HN O3 Air Quality Forecast PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

N2O5 (night) Air Quality Forecast PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

50%

1.3e15 molec cm-2

1.3e15 molec cm-2

PAN Air Quality Forecast PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

Spectral UV surface albedo UV actinic fluxes Surface 5 / 20 -- 24 / 24*3 0.1

/166

S-4/-5 MRTD



B3 Theme:Category:

Air QualityAssessment


Driver

Height Range




Uncertainty

O3 Phot. Acitivity; Oxidising

Capacity; Background

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

10%

20%

25%

3%

NO2 Emissions; Phot. Acitivity;

Oxidising Capacity

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

10%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

CO ( + isotopes) Oxidising Capacity;

Emissions; Background

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

25%

25%

Aerosol OD Emissions; UV actinic fluxes PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

--

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

0.05

0.05

0.05

0.05

Aerosol Type Emissions PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

--

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

< 10% mis-

assignments

(for all altitude

ranges)

H2O Oxidising Capacity PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

10%

20%

10%

10%

SO2 Emissions PBL 5 / 20 -- 0.5 / 2 20%

/166

S-4/-5 MRTD



FT

Tropospheric Column

Total Column

5 / 50

5 / 20

5 / 20

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

CH2O Phot. Activity; VOC

emissions

PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

HN O3 Oxidising Capacity PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

N2O5

(nighttime)

Oxidising Capacity PBL

FT

Tropospheric Column

Total Column

5 / 20

5 / 50

5 / 20

5 / 20

--

1 / 3

--

--

0.5 / 2

0.5 / 2

0.5 / 2

0.5 / 2

20%

20%

1.3e15 molec cm-2

1.3e15 molec cm-2

Organic Nitrates Oxidising Capacity PBL 5 / 20 -- 0.5 / 2 30%

Spectral UV surface albedo UV actinic fluxes Surface 5 / 20 -- 24 / 24*3 0.1

/166

S-4/-5 MRTD



C1 Theme:Category:

ClimateProtocol Monitoring


Driver

Height Range




Uncertainty

CO2 (PBL sensitive) Emissions Tropospheric column

Total column

10 / 50

10 / 50

--

--

6 / 12

6 / 12

0.5%

0.5%

CH4 (PBL sensitive) Emissions Tropospheric column

Total column

10 / 50

10 / 50

--

--

24 / 24*3

24 / 24*3

2%

2%

O3 Radiative forcing Troposphere

Tropospheric column

Total column

10 / 50

10 / 50

50 / 100

2 / 5

--

--

12 / 24*3

12 / 24*3

24 / 24*3

20%

25%

3%

NO2 (PBL sensitive) Emissions Troposphere

Tropospheric column

Total column

10 / 50

10 / 50

10 / 50

2 / 5

--

--

12 / 24*3

12 / 24*3

12 / 24*3

50%

1.3·(10)15 cm-

1.3·(10)15 cm-2

CO (PBL sensitive) Emissions Troposphere

Tropospheric column

Total column

10 / 50

10 / 50

10 / 50

2 / 5

--

--

12 / 24*3

12 / 24*3

12 / 24*3

20%

25%

25%

Aerosol OD Emissions; Radiative

forcing

Troposphere

LS

MS

Total column

10 / 50

50 / 100

50 / 200

10 / 50

--

1 / part. column

2 / part. column

--

6 / 24*3

12 / 24*3

12 / 24*3

12 / 24*3

0.05

0.05

0.05

0.05

Aerosol absorption OD Radiative forcing Troposphere

Total column

10 / 50

10 / 50

--

--

6 / 24*3

6 / 24*3

0.01

0.01

/166

S-4/-5 MRTD



C2 Theme:Category:

ClimateNear-Real Time


Driver

Height Range




Uncertainty

O3 Radiation; Dynamics PBL

Tropospheric column

LS

MS

US+M

Total column

5 / 50

10 / 50

50 / 100

50 / 200

50 / 200

50 / 100

--

--

0.5 / 2

1 / 3

3 / 5

--

6 / 24

6 / 24*3

6 / 24*3

6 / 24*7

6 / 24*7

6 / 24*3

30%

25%

10%

20%

20%

5%

H2O Radiation; Dynamics;

Hydrological cycle;

Stratospheric H2O

PBL

FT

UT

LS

MS

US

Total column

5 / 50

10 / 50

10 / 100

50 / 100

50 / 200

50 / 200

10 / 50

--

0.5 / 2

0.5 / 2

0.5 / 2

1 / 3

3 / 5

--

1 / 6

1 / 6

1 / 6

3 / 24

6 / 24*7

6 / 24*7

6 / 24*3

50%

30%

30%

20%

20%

20%

5%

CO2 Radiation; Tracer PBL

MS

US

Total column

5 / 50

50 / 200

50 / 200

1 / 20

--

1 / 3

1 / 3

--

6 / 12

12 / 24*7

12 / 24*7

1 / 12

10%

10%

10%

2%

CH4 Radiation; Tracer LS

MS

Total column

50 / 100

50 / 200

10 / 50

1 / 3

1 / 3

--

12 / 24*3

12 / 24*3

12 / 24*3

20%

20%

2%

N2O Radiation; Tracer LS

MS

US

Total Column

50 / 100

50 / 200

50 / 200

10 / 50

1 / 3

1 / 3

3 / 5

--

12 / 24*3

12 / 24*3

12 / 24*3

12 / 24*3

20%

20%

20%

2%

/166

S-4/-5 MRTD



Aerosol OD Radiation PBL

Troposphere

LS

MS

5 / 10

5 / 50

50 / 100

50 / 200

--

--

1 / part. column

1 / part. column

1 / 6

3 / 24

12 / 24*3

12 / 24*3

0.05

0.05

0.05

0.05

Aerosol absorption OD Radiation PBL

Troposphere

5 / 10

5 / 50

--

--

1 / 6

3 / 24

0.01

0.01

Cirrus OD Radiation UT 10 / 100 -- 6 / 24 100%

SF6 Tracer LS

MS

US

50 / 100

50 / 200

50 / 200

1 / 3

1 / 3

3 / 5

12 / 24*7

12 / 24*7

12 / 24*7

10%

10%

10%

HDO Tracer; Stratospheric H2O LS

MS

US

50 / 100

50 / 200

50 / 200

1 / 3

1 / 3

3 / 5

12 / 24*7

12 / 24*7

12 / 24*7

10%

10%

10%

HF (alternative tracer) Tracer LS

MS

US

50 / 100

50 / 200

50 / 200

1 / 3

1 / 3

3 / 5

12 / 24*7

12 / 24*7

12 / 24*7

10%

10%

10%

Aerosol phase function Radiation PBL

Troposphere

5 / 10

5 / 50

--

--

1 / 6

3 / 24

0.1 on asymmetry

factor

0.1 on asymmetry

factor

Cirrus phase function Radiation UT 10 / 100 -- 6 / 24 0.1 on a. factor

/166

S-4/-5 MRTD



C3 Theme:Category:

ClimateAssessment


Driver

Height Range




Uncertainty

O3 Radiative Forcing; Oxidising

Capacity; Tracer; Ozone

recovery

Troposphere

Tropospheric Column

UT

LS

MS

US+M

Total Column

10 / 50

10 / 50

20 / 100

50 / 100

50 / 100

100 / 200

50 / 100

1 / 3

--

0.5 / 2

0.5 / 2

2 / 3

3 / 5

--

6 / 24*3

6 / 24*3

6 / 24*3

6 / 24*3

6 / 24*3

6 / 24*7

6 / 24*3

30%

25%

20%

20%

20%

20%

3%

H2O Radiative Forcing; Oxidising

Capacity; Tracer; O3

recovery; Stratospheric H2O

PBL

Troposphere

Tropospheric Column

UT

LS

MS

US+M

Total Column

1 / 20

10 / 50

10 / 50

20 / 100

50 / 100

50 / 100

100 / 200

50 / 100

--

1 / 3

--

0.5 / 2

0.5 / 2

2 / 3

3 / 5

--

6 / 24

6 / 24*3

6 / 24*3

6 / 24*3

6 / 24*3

6 / 24*7

6 / 24*7

6 / 24*3

30%

30%

10%

20%

20%

20%

20%

10%

CO2 Radiative Forcing; Tracer MS

Total Col. (PBL sensitive)

50 / 100

10 / 50

2 / 3

--

12 / 24*3

1 / 12

10%

0.5%

CH4 Radiative Forcing; Oxidising

Capacity; Tracer;

Stratospheric H2O

LS

MS


50 / 100

50 / 100

10 / 50

1 / 3

2 / 3

--

12 / 24*3

12 / 24*3

12 / 24*3

20%

20%

2%

N2O Radiative Forcing; Tracer; N

budget

LS

MS

US


50 / 100

50 / 100

50 / 100

10 / 50

1 / 3

2 / 3

3 / 5

--

12 / 24*3

12 / 24*3

12 / 24*7

12 / 24*3

20%

20%

20%

2%

/166

S-4/-5 MRTD



CO Ozone and CO2 precursor Troposphere

Trop. Col. (PBL sensitive)

UT

LS

10 / 50

10 / 50

20 / 100

50 / 100

1 / 3

--

1 / 3

1 / 3

12 / 24*3

12 / 24*3

12 / 24*3

12 / 24*3

30%

25%

20%

20%

NO2 Ozone and Aerosol

precursor

Troposphere


UT

LS

MS

Total Column

10 / 50

10 / 50

20 / 100

50 / 100

50 / 200

50 / 100

1 / 3

--

1 / 3

1 / 3

2 / 3

--

6 / 24*3

12 / 24*3

6 / 24*3

12 / 24*3

12 / 24*3

12 / 24*3

30%

1.3·(10)15 cm-2

50%

50%

30%

10%

CH2O Oxidising Capacity Troposphere


UT


10 / 50

10 / 50

20 / 100

10 / 50

1 / 3

--

1 / 3

--

6 / 24*3

12 / 24*3

6 / 24*3

12 / 24*3

30%

1.3·(10)15 cm-2

30%

1.3·(10)15 cm-2

HN O3 N budget Troposphere

UT

LS

MS

Total Column

10 / 50

20 / 100

50 / 100

50 / 200

10 / 50

1 / 3

1 / 3

1 / 3

2 / 3

--

6 / 24*3

6 / 24*3

12 / 24*3

12 / 24*3

12 / 24*3

30%

20%

20%

20%

20%

Cirrus OD Radiative Forcing UT 10 / 100 -- 6 / 24 100%

PSC occurrence (day +

night)

Radiative Forcing LS 50 / 100 0.5 / 2 6 / 24*3 < 10% mis-assignments

Aerosol OD Radiative Forcing PBL

Troposphere

LS

MS

Total Column

5 / 20

10 / 50

50 / 100

50 / 200

10 / 50

--

--

1 / part. column

2 / part. column

--

6 / 24

6 / 24

12 / 24*3

12 / 24*3

12 / 24*3

0.05

0.05

0.05

0.05

0.05

Aerosol absorption OD Radiative Forcing Troposphere

Total Column

5 / 50

5 / 50

--

--

6 / 24

6 / 24

0.01

0.01

Spectral solar irradiance Radiative Forcing Top of Atmosphere -- -- 24 / 24*7 2%

HCl Ozone recovery LS 50 / 100 1 / 3 12 / 24*3 20%

/166

S-4/-5 MRTD



CH3Cl Ozone recovery LS

MS

Stratosphere

50 / 100

50 / 200

50 / 100

1 / 3

2 / 3

--

12 / 24*3

12 / 24*3

12 / 24*3

20%

20%

20%

CH3Br Ozone recovery LS

MS

Stratosphere

50 / 100

50 / 200

50 / 100

1 / 3

2 / 3

--

12 / 24*3

12 / 24*3

12 / 24*3

20%

20%

20%

SF6 Tracer LS

MS

US

50 / 100

50 / 200

50 / 200

1 / 3

2 / 3

3 / 5

12 / 24*7

12 / 24*7

12 / 24*7

10%

10%

10%

HDO Tracer; Stratospheric H2O LS

MS

US

Stratosphere

50 / 100

50 / 200

50 / 200

50 / 100

1 / 3

2 / 3

3 / 5

--

12 / 24*7

12 / 24*7

12 / 24*7

12 / 24*7

10%

10%

10%

10%

HF Tracer LS

MS

US

50 / 100

50 / 200

50 / 200

1 / 3

2 / 3

3 / 5

12 / 24*7

12 / 24*7

12 / 24*7

10%

10%

10%

CFC-11 Radiative Forcing LS

MS

Stratosphere

50 / 100

50 / 200

50 / 100

1 / 3

2 / 3

--

12 / 24*7

12 / 24*7

12 / 24*7

20%

20%

20%

CFC-12 Radiative Forcing LS

MS

Stratosphere

50 / 100

50 / 200

50 / 100

1 / 3

2 / 3

--

12 / 24*7

12 / 24*7

12 / 24*7

20%

20%

20%

HCFC-22 Radiative Forcing UT

LS

MS

Stratosphere

20 / 100

50 / 100

50 / 200

50 / 100

1 / 3

1 / 3

2 / 3

--

12 / 24*3

12 / 24*3

12 / 24*3

12 / 24*3

20%

20%

20%

20%

H2O2 Oxidising Capacity Troposphere

UT

10 / 50

20 / 100

1 / 3

1 / 3

6 / 24*3

6 / 24*3

30%

30%

/166

S-4/-5 MRTD



N2O5 N budget Troposphere

UT

LS

MS

Stratosphere

10 / 50

20 / 100

50 / 100

50 / 200

50 / 100

--

1 / 3

1 / 3

1 / 3

--

6 / 24*3

6 / 24*3

12 / 24*3

12 / 24*3

12 / 24*3

30%

30%

50%

50%

50%

PAN N budget Troposphere

UT

Total column

10 / 50

20 / 100

10 / 50

--

1 / 3

--

6 / 24*3

6 / 24*3

6 / 24*3

30%

30%

30%

CH3COCH3 Oxidising Capacity Troposphere

UT

Total column

10 / 50

20 / 100

10 / 50

--

1 / 3

--

6 / 24*3

6 / 24*3

6 / 24*3

30%

30%

30%

C2H6 Oxidising Capacity Troposphere

UT

Total column

10 / 50

20 / 100

10 / 50

--

1 / 3

--

6 / 24*3

6 / 24*3

6 / 24*3

50%

50%

50%

ClO (for enhanced levels) Ozone Recovery LS

MS

Stratosphere

50 / 100

50 / 200

50 / 100

1 / 3

2 / 3

--

12 / 24*3

12 / 24*3

12 / 24*3

20%

20%

20%

ClONO2 Ozone Recovery LS

MS

Stratosphere

50 / 100

50 / 200

50 / 100

1 / 3

2 / 3

--

12 / 24*3

12 / 24*3

12 / 24*3

20%

20%

20%

SO2 (for enhanced levels) Volcanoes Troposphere

LS

MS

Total column

10 / 50

50 / 100

50 / 200

10 / 50

1 / 3

1 / 3

2 / 3

--

6 / 24*3

12 / 24*3

12 / 24*3

6 / 24*3

50%

50%

50%

50%

/166

S-4/-5 MRTD



Aerosol phase function Radiative Forcing;

Volcanoes

Troposphere

LS

MS

Total column

10 / 50

50 / 100

50 / 200

10 / 50

--

1 / part. column

2 / part. column

--

6 / 24

12 / 24*3

12 / 24*3

6 / 24

0.1 on asymmetry

factor

0.1 on asymmetry

factor

0.1 on asymmetry

factor

0.1 on asymmetry

factor

Cirrus phase function Radiative Forcing UT 10 / 100 -- 6 / 24 0.1 on asymmetry

factor

/166

S-4/-5 MRTD



APPENDIX 4 GAIN VECTORS

In this Appendix, gain vectors are depicted to be used for the evaluation of the ESRA requirements MR-LEO-UVN-185 and MR-LEO-SWIR-185. Where applicable, also the spectrally varying Degree Of Polarisation (DOP) to be assumed for this evaluation is depicted.

Figure A4.1. Gain Vector for the spectral range 310-325 nm for the evaluation of the requirement MR-LEO-UVN-185.

/166

S-4/-5 MRTD




/166

S-4/-5 MRTD



Figure A4.5. Gain Vector for the spectral range 685-710 and 755-773 with 0.4 nm spectral resolution for the evaluation of the requirement MR-LEO-UVN-185 for an cirrus cloud scenario.

/166

S-4/-5 MRTD



Figure A4.6. Gain Vector for the spectral range 685-710 and 755-773 with 0.4 nm spectral resolution for the evaluation of the requirement MR-LEO-UVN-185 for an aerosol scenario.

/166

S-4/-5 MRTD



Figure A4.7. Gain Vector for the spectral range 755-773 nm with 0.12 nm spectral resolution for the evaluation of the requirement MR-LEO-UVN-185.

/166

S-4/-5 MRTD



Figure A4.8. Degree of Polarization for the spectral range 755-773 nm with 0.12 nm spectral resolution for the evaluation of the requirement MR-LEO-UVN-185.

/166

S-4/-5 MRTD



Figure A4.9. Gain Vector for the spectral range 1590-1675 nm spectral resolution for the evaluation of the requirement MR-LEO-SWIR-185.

/166

S-4/-5 MRTD




/166

S-4/-5 MRTD



APPENDIX 5 Detailed Change Record

Changes of the MRTD v2.0 with respect to v1.0

Requirement ID Change Comment

MR-LEO-SYS-35 simplification In the Sentinel-5 requirement on across-track spatial sampling, the reference has been removed to which across track spatial sampling the radiometric requirements apply. Replaced by a general comment that all requirements apply to the achieved spatial sampling, as agreed at MAG#3

MR-LEO-SYS-37 removed The Sentinel-5 requirement indicating specifc complaince for spatial requirements has been eplaced by a general comment that all requirements apply to the achieved spatial sampling. C18

MR-LEO-SYS-39 relaxation The Sentinel-5 requirement on the number of along-track spatial smaples for the high spatial sampling ranges has now also a threshold value of 0.33 SSD and the original value of 0.2 SSD became a goal value

MR-LEO-SYS-40 relaxation The Sentinel-5 requirements on intra-band co-registration have been relaxed. The values are presented in a table without goal values and with relaxed threshold values. Agreed at MAG meeting #7.

MR-LEO-SYS-60 relaxation The Sentinel-5 requirements on inter-band co-registration have been relaxed. The values are presented in a table without goal values. The new values are relaxed in combination with NIR-2 and not applicable to all combinations of different bands. Agreed at MAG meeting #7.

MR-LEO-SYS-61 new This new Sentinel-5 requirement on integrated enegery replaces the former MR-LEO-UVN-80 and -87, and MR-LEO-SWIR-50 and -57. The requirement is now applicable only for the system integrated energy (SIE) and at nadir

MR-LEO-SYS-62 new This new Sentinel-5 requirement on integrated enegery replaces the former MR-LEO-UVN-80 and -87. The requirement is now only applicable for the spectral variation of the system integrated energy (SIE) between UV2-VIS-NIR1-NIR2 at nadir

MR-LEO-SYS-63 new This new Sentinel-5 requirement on integrated enegery replaces the former MR-LEO-UVN-80 and -87, and MR-LEO-SWIR-50 and -57. The requirement is now only applicable for the spectral variation of the system integrated energy (SIE) between NIR2-SWIR1-SWIR3 at nadir

MR-LEO-SYS-64 new This new Sentinel-5 requirement on integrated enegery replaces the former MR-LEO-UVN-80 and -87. The requirement is now only applicable for the spectral variation of the instrument integrated energy (IIE) between UV2-VIS-NIR1-NIR2 for any

/166

S-4/-5 MRTD



spatial sample

MR-LEO-SYS-65 new This new Sentinel-5 requirement on integrated enegery replaces the former MR-LEO-UVN-80 and -87, and MR-LEO-SWIR-50 and -57. The requirement is now only applicable for the spectral variation of the instrument integrated energy (IIE) between NIR2-SWIR1-SWIR3 for any spatial sample

MR-LEO-SYS-130

tightening The Sentinel-5 requirement on timeliness is now to make level-0 data available in real time to local users during overpass.

MR-LEO-UVN-80

removed The Sentinel-5 UVN requirement on integrated energy has been merged with MR-LEO-UVN-87, MR-LEO-SWIR-50 and MR-LEO-SWIR-57, and was subsequently redefined and replaced by the new requirements MR-LEO-SYS-61, -62, -63, -64 and -65.

MR-LEO-UVN-87

removed The Sentinel-5 UVN requirement on integrated energy has been merged with MR-LEO-UVN-80, MR-LEO-SWIR-50 and MR-LEO-SWIR-57, and was subsequently redefined and replaced by the new requirements MR-LEO-SYS-61, -62, -63, -64 and -65.

MR-LEO-UVN-88

new This new Sentinel-5 UVN requirement is introduced to constrain the spatial variation of the IIE over all channels in the UVN

MR-LEO-UVN-90

clarification The Sentinel-5 UVN requirement defining the spectral ranges to be covered in the NIR has been updated in the NIR clarifying goal and threshold ranges. The threshold range in the NIR-1 has slightly from 710-750 nm to 710-745 nm. In NIR-2 there is a new threshold range from 755-773 nm besides the goal of 750-775 nm. A new footnote in Table 6.3 specifies which combination of spectral requirements are tolerable for compliance

MR-LEO-UVN-100

clarification The Sentinel-5 UVN requirement on spectral resolution has been updated adding in the NIR a new breakthrough value and clarifying in a new footnote (of Table 6.3) which combination of spectral requirements are tolerable for compliance

MR-LEO-UVN-120

tightening and relaxation

The Sentinel-5 UVN requirement on has been updated adding the VIS band (i.e., tightening) and changing the value in the NIR from 0.01 SSI to 0.05 SSI (i.e., relaxation). Agreed at MAG meeting #5.

MR-LEO-UVN-130

relaxation The Sentinel-5 UVN requirement on spectral stability at level-0 has been relaxed from 0.05 SSI to 0.1 SSI. Agreed at MAG meeting #7.

/166

S-4/-5 MRTD



MR-LEO-UVN-150

relaxation The Sentinel-5 UVN requirement on signal-to-noise ratios of radiance has been slightly relaxed in the VIS range between 430 and 460 nm. Threshold value was 1800 and is now 1500. Goal value was 2500 and is now 2130. Agreed at MAG meeting #4.

MR-LEO-UVN-160

simplification The Sentinel-5 UVN requirement on absolute radiometric accuracy has been simplified. Previously it was formulated in a complex way with signal-dependent and signal-independent parts, it now applies to the reference dynamic range with the lower limit is defined by continuum level of Lmin. Agreed at MAG meeting #7.

MR-LEO-UVN-180

relaxation The Sentinel-5 UVN requirement on relative spectral radiometric accuracy (RSRA) has been relaxed in view of new MR-LEO-UVN-185 requirement (on ESRA) and is now applicable in ranges outside those covered by the ESRA. Agreed at MAG meeting #8.

MR-LEO-UVN-185

new This new Sentinel-5 UVN requirement on effective spectral radiometric accuracy (ESRA) has been introduced to constrain only the spectral features with an impact on the level-2 performance by using, so-called, gain vectors. Agreed at MAG meeting #8.

MR-LEO-UVN-190

removed The Sentinel-5 UVN requirement on relative spectral radiometric accuracy (RSRA) of reflectance has been replaced by the new requirement MR-LEO-UVN-185 (ESRA). Agreed at MAG meeting #8.

MR-LEO-UVN-220

relaxation The Sentinel-5 UVN requirement on signal-to-noise ratios of irradiance has been slightly relaxed in the UV range below 325 nm from 10000 to 8000 (in Table 6.4). Agreed at MAG meeting #3.

MR-LEO-UVN-257

relaxation The Sentinel-5 UVN requirement on radiometric stability over an orbit is now only applicable to the reference dynamic range with the lower limit is defined by continuum level of Lmin. Agreed at MAG meeting #7.

MR-LEO-UVN-258

relaxation The Sentinel-5 UVN requirement on radiometric stability over its lifetime is now only applicable to the reference dynamic range with the lower limit is defined by continuum level of Lmin. Agreed at MAG meeting #7.

MR-LEO-UVN-260

relaxation The Sentinel-5 UVN requirement on polarisation sensitivity in the NIR-2 range has been relaxed from 0.005 to 0.007 outside the central part of the swath. Agreed at MAG meeting #7.

MR-LEO-SWIR-50

removed The Sentinel-5 SWIR requirement on integrated energy has been merged with MR-LEO-UVN-80, MR-LEO-UVN-87 and MR-LEO-SWIR-57, and was subsequently redefined and replaced by the new requirements MR-LEO-SYS-61, -62, -63, -64 and -65.

/166

S-4/-5 MRTD



MR-LEO-SWIR-57

removed The Sentinel-5 SWIR requirement on integrated energy has been merged with MR-LEO-UVN-80, MR-LEO-UVN-87 and MR-LEO-SWIR-50, and was subsequently redefined and replaced by the new requirements MR-LEO-SYS-61, -62, -63, -64 and -65.

MR-LEO-SWIR-110

relaxation The Sentinel-5 SWIR requirement on spectral sampling ratio has been relaxed by introduction of a threshold value of 2.5 besides the (original) goal value of 3.0.

MR-LEO-SWIR-120

relaxation The Sentinel-5 SWIR requirement on spectral calibration in SWIR-3 has been relaxed from 0.002 to 0.05 nm. This is now in line with the MR-LEO-UVN-120 requirement.

MR-LEO-SWIR-130

relaxation The Sentinel-5 SWIR requirement on spectral stability at level-0 in SWIR-3 has been relaxed from 0.05 nm to 0.1 nm.

MR-LEO-SWIR-140

clarification In the Sentinel-5 SWIR requirement on ISRF knowledge, it has been clarified that this requirement applies to spatially homogeneous scenes. Agreed at MAG meeting #5.

MR-LEO-SWIR-150

modification The Sentinel-5 SWIR requirement on signal-to-noise ratios (SNR) of radiance for the SWIR bands is now specified in terns of a SNR spectrum.

MR-LEO-SWIR-160

simplification The Sentinel-5 SWIR requirement on absolute radiometric accuracy has been simplified. Previously it was formulated in a complex way with signal-dependent and signal-independent parts, it now applies to the reference dynamic range. Agreed at MAG meeting #7.

MR-LEO-SWIR-170

clarification In the Sentinel-5 SWIR requirement on on absolute radiometric accuracy, it has been clarified that this requirement applies to the reference dynamic range. Agreed at MAG meeting #7.

MR-LEO-SWIR-180

removed The Sentinel-5 SWIR requirement on relative spectral radiometric accuracy (RSRA) of reflectance has been replaced by the new requirement MR-LEO-SWIR-185 (ESRA). Agreed at MAG meeting #8.

MR-LEO-SWIR-185

new This new Sentinel-5 SWIR requirement on effective spectral radiometric accuracy (ESRA) has been introduced to constrain only the spectral features with an impact on the level-2 performance by using, so-called, gain vectors. Agreed at MAG meeting #8.

MR-LEO-SWIR- relaxation The Sentinel-5 SWIR requirement on additive offset in reflectance has been relaxed from 0.1% to 0.2%.

/166

S-4/-5 MRTD



198

MR-LEO-SWIR-287

simplification The Sentinel-5 SWIR requirement constraining the deviation of the polarisation sensitivity has been specified for specific elements of the Müller matrix. Agreed at MAG meeting #5.

MR-LEO-SWIR-290


MR-LEO-SWIR-300


MR-GEO-SYS-5 new The requirement on Sentinel-4 Level-1b data availability has been introduced for consistency with the corresponding Sentinel-5 requirement MR-LEO-SYS-5, and in line with with the Sentinel-4 system requirement S4-SA-AVA-034.

MR-GEO-SYS-20 relaxation A dedicated Sentinel-4 data acquisition mode with a south shifted geographic coverage was dropped from the operations baseline, as agreed by the MAG (meeting 7, June 2013) and in line with with the Sentinel-4 system requirement PL-OB-150. The reformulated requirement calls for the capability for the Sentinel-4 mission to be operated with south shifted geographic coverage, in view of an operations mode that might be defined in the future.

MR-GEO-SYS-30 clarification The explicit specification of the Sentinel-4 nominal repeat cycle of 1 h has been added, in the context of the associated E/W scan range in line with with the Sentinel-4 system requirement S4-PL-OB-160. Previously the nominal repeat cycle was only defined in MR-SYSTEM-20, where the associated E/W scan range is not defined.

MR-GEO-SYS-80 tightening The Sentinel-4 requirement on inter-band spatial co-registration, previously formulated for any pair of bands including the UV, Vis and NIR, has been split. The mis-registration between UV and VIS is now constrained more tightly than the mis-registration between between pairs of bands including the NIR, in line with users needs and in line with with the Sentinel-4 system requirement PL-OB-340.

MR-GEO-UVN-30

relaxation The Sentinel-4 requirement on spatial resolution has been relaxed by extending the spatial reference area across which the integrated energy of the spatial response function is to be reached. The relaxation in EW direction was needed after changing the acquisition strategy from step-and-stare to continuous scanning, agreed by the MAG (meeting 1, August 2010). An additional small relaxation was made in all directions to reach feasibility and tracing to the Sentinel-4 system requirements PL-OB-310 and -320.

/166

S-4/-5 MRTD



MR-GEO-UVN-37

new A constraint of the Sentinel-4 spectral variation of the Integrated Energy has been introduced in order to complement the inter-band spatial co-registration requirements, in line with with the Sentinel-4 system requirement PL-OB-330.

MR-GEO-UVN-110

relaxation It was clarified that the SNR performance for Sentinel-4 is to be evaluated per spectral sampling element and spectral ranges where degraded performance is allowed have been introduced, in line with with the Sentinel-4 system requirement S4-PL-OB-710.

MR-GEO-UVN-120

simplification The Sentinel-4 requirement on absolute radiometric accuracy, previously formulated in a complex way with signal-dependent and signal-independent parts, has been simplified, in line with with the Sentinel-4 system requirement PL-OB-440.

MR-GEO-UVN-130

clarification It has been clarified that the Sentinel-4 requirement on relative spatial radiometric accuracy applies to partially polarised scenes and the reference degree of polarisation has been specified, in line with the Sentinel-4 system requirement PL-OB-450.

MR-GEO-UVN-190

correction The required SNR for the Sentinel-4 solar irradiance has been increased in the UV. The previous specification erroneously accounted for a decrease of the signal with decreasing wavelength as it applies to Earth radiance spectra. The new specification accounts for the spectral dependence of the solar irradiance signal, in line with the Sentinel-4 system requirement S4-RS PL-OB-600.