Michael Buchwitz Institute of Environmental Physics (IUP)

ESA Earth Observation Summer School, ESRIN, 4-14 August 2014

Greenhouse gas observations from space: Results from ESA‘s GHG-CCI project

1 www.iup.uni-bremen.de

Greenhouse gas observations from space 1. Why and how ?

2. Results from ESA‘s GHG-CCI project

3. Proposed future mission CarbonSat

Overview Talks 1 - 3

2

Greenhouse gas observations from space: Results from ESA‘s GHG-CCI project

• Project overview:

• GHG-CCI: A project to deliver the Essential Climate Variable „Greenhouse Gases“ (short)

• CO2 and CH4 sources & sinks:

• What has already been learned using satellite data ? (focus)

Overview Talk 2

3

Important note for ESA

4

This is NOT the public version of this talk !

Contains unpublished material submitted to peer-reviewed journals

with „press embargo“ and related restrictions prior to publication

A public version pdf file can be made available on request

ESA Climate Change Initiative (CCI) to generate Essential Climate Variables (ECVs)

www.esa-ghg-cci.org/ ESA programme led by Mark Doherty, ESA/ESRIN

Currently 13 ECV projects: • Aerosol-CCI • Cloud-CCI • Fire-CCI • GHG-CCI - CO2 & CH4 • Glaciers-CCI • LandCover-CCI • OceanColour-CCI • Ozone-CCI • SeaLevel-CCI • SST-CCI • SoilMoisture-CCI • SeaIce-CCI • IceSheets-CCI

+ CMUG (Climate Modelling User Group) • Lead: Roger Saunders (Met Office Hadley Centre) • Met Office Hadley Centre, ECMWF, MPI-Meteorology, Météo France

5

Essential Climate Variable (ECV) Greenhouse Gases (GHG)

6

ECV GHG (GCOS-154*)): “Retrievals of greenhouse gases, such as CO2 and CH4, of sufficient quality to estimate regional sources and sinks.”

*) „SYSTEMATIC OBSERVATION REQUIREMENTS FOR SATELLITE-BASED DATA PRODUCTS FOR CLIMATE“

Reliable climate prediction requires a good understanding of the natural and anthropogenic (surface) sources and sinks of CO2 and CH4.

Important questions are, for example:

• Where are they ? • How strong are they ? • How do they respond to a changing climate ?

A better understanding requires appropriate global observations and (inverse) modelling.

CO2 and CH4 are the two most important anthropogenic greenhouse gases and increasing concentrations

result in global warming.

Observed and predicted temperature change (AR5)

Future?

Existing and planed GHG (*) satellite missions

From: CEOS Strategy for Carbon Observations from Space, v.2.0

2016 GHG-CCI time series

2002 2012 2014 2021 2009

SCIAMACHY / ENVISAT ESA/DLR Uni Bremen

CarbonSat ESA / Uni Bremen

GOSAT JAXA / NIES / MOE OCO-2 NASA

(*) Near-surface-sensitive CO2 and/or CH4 missions 7

…

GHG-CCI project www.esa-ghg-cci.org

SCIAMACHY/ENVISAT

Global satellite observations Global information on near-surface CO2 & CH4

TANSO/GOSAT

Upper layer CO2 & CH4

IASI, MIPAS, SCIA/occ, AIRS, ACE-FTS, …

Global observations

Calibrated radiances Calibration (L 0-1)

Reference observations Validation

Inverse modelling

(L 2-4)

Improved information on GHG sources & sinks

Atmospheric GHG distributions

Retrieval (L 1-2)

8

User Requirements

~0.12%

~0.5%

Available on http://www.esa-ghg-cci.org/

9

GHG-CCI Carbon dioxide

GHG-CCI Methane

10

GHG-CCI: GHG data sets

GHG-CCI: XCO2 Animation

11

SCIAMACHY/ENVISAT TANSO/GOSAT

GHG-CCI Phase 1 (2010-2013): CRDP#1

Level 1: ESA/DLR JAXA

Level 2: IUP, Univ. Bremen Univ. Leicester SRON / KIT

This image cannot currently be displayed.

Animation: http://www.esa-ghg-cci.org/sites/default/files/documents/public/images/co2scigos_crdp1_ani_v2sm.gif

http://www.bookperhead.com/images/articles/round-robin-parlays.jpg

Round Robin

Algo 1

12

GHG-CCI: Overview incl. RR

13 Link to this & more: www.esa-ghg-cci.org -> Publications

CCI Integration Meeting, ECMWF, 14-16 March 2011

URD DARD

AIECAR

ASR

PSD

PVASR

RREP

PVIR

CAR

SPD

SRD

ATBD

… and many more …

All publicly available on www.esa-ghg-cci.org -> Documents

14

GHG-CCI: Documents

GHG-CCI: XCO2 time series

15

To be extended in Phase 2

http://www.northpacificmusic.com/ensemble.east.west.jpg

Ensemble: Key to success

16

• Multiple satellite algorithms / products

• Multiple models / inverse models

Ensemble algorithm EMMA

17

„Monitoring“ of improvement of global products

GOSAT

SCIA + GOSAT SCIA + GOSAT

Reuter et al., ACP, 2013

a b c

EMMA: Latest results (v1.6a)

18

Algo N coloc. StdDev of Δ [ppm]

Station-to-station

bias [ppm]

WFMD 20173 3.93 1.06

BESD 8606 2.17 0.85

RemoTeC 1621 2.07 0.61

ACOS 1812 1.79 0.64

UoL-FP 2396 2.34 0.80

NIES PPDF 1355 1.73 0.69

NIES op. 1407 2.18 0.89

EMMA 2087 2.04 0.73

Validation using TCCON (11 sites):

Method & details: Reuter et al., ACP, 2013

19

XCO2: Ensemble results: Seasonal cycle amplitude

Buchwitz et al., RSE, 2013

SCIAMACHY CO2

De-trended CO2 time series

Peak-to-peak amplitude

GHG-CCI: Validation: XCO2 Product Validation and Intercomparison Report

(PVIR), v2

Available from: http://www.esa-ghg-cci.org/

-> Documents

PVIR v2

20

GHG-CCI: Validation: XCH4 Product Validation and Intercomparison Report

(PVIR), v2

Available from: http://www.esa-ghg-cci.org/

-> Documents

PVIR v2

21

Variable(*) Resolution Accuracy Stability

XCO2 Temporal:

GCOS: 4 hours Achieved: Days

No existing nor any planned mission meets the GCOS temporal resolution requirement.

Spatial: GCOS: 5-10 km

Achieved($): 10 km ($) for GOSAT. SCIAMACHY: 30x60 km2.

URD: SCIAMACHY and GOSAT are useful to generate the ECV GHG. Note: GCOS requirements are target (maximum) requirements but URD requirements listed here are thresold (minumum) requirements.

GCOS: 1 ppm URD(#): 0.5 ppm

Achieved(#): ~1 ppm

GCOS: 0.2 ppm/yr URD: 0.5 ppm/yr

Achieved: ~0.2 ppm/yr(+)

(+) for SCIAMACHY; for GOSAT: Not yet quantified (time period too short)

XCH4 GCOS: 10 ppb URD(#): 10 ppb

Achieved(#): ~6 ppb(§)

(§) for GOSAT; for SCIAMACHY 8-18 ppb depending on time period

GCOS: 2 ppb/yr URD: 10 ppb/yr Achieved: (?)

(?) GOSAT: Not yet quantified (time period too short); SCIAMACHY: Not met due to degradation issues

(#) Relative accuracy

(*) Requirements for column-averaged mixing ratios (= normalized vertical columns) as required by URD; it is assumed here that this corresponds to GCOS variables „Tropospheric CO2 column“ and „Tropospheric CH4 column“ References: Requirements for ECV Greenhouse Gases (GHG): • GCOS-154: „SYSTEMATIC OBSERVATION REQUIREMENTS FOR SATELLITE-BASED DATA PRODUCTS FOR CLIMATE“ • URD: “GHG-CCI User Requirements Document”, v1.0 Definition: ECV GHG (GCOS-154): • Product A.8.1: Retrievals of CO2 and CH4 of sufficient quality to estimate regional sources and sinks

Phase 1 achievements: Comparison with GCOS requirements

22

Results

23

Satellite-derived GHG source/sink-related results:

Carbon dioxide

CO2 seasonal cycle and growth rate

Schneising et al., ACP, 2014

24

SCIA WFMD&BESD CO2: Terrestrial carbon sink

Inter-annual variability of CO2 growth rate vs Temperature

Inter-annual variability of CO2 seasonal cycle amplitude vs Temperature

Terrestrial carbon uptake variability correlated with / driven by near-surface temperature changes:

SCIAMACHY: • 1.25 +/ 0.32 ppm /yr /K • -> approx. 2.7 +/- 0.7 GtC /yr /K

CarbonTracker vs. SCIAMACHY: Good agreement

Schneising et al., ACP, 2014

Less carbon uptake (= higher atmospheric growth rate) in warmer years 25

XCO2: Comparison with Models

JFM AMJ

JAS OND

NOAA/CT

LSCE/MACC MPI-BGC

-8 +8 0 Model data: F. Chevallier, LSCE; C. Rödenbeck, MPI-BGC; NOAA

SCIA

26 Who is right and who is wrong ? Assessment ongoing …

CO2 emissions -> Temperature change

Emitted until 2011: 531 (446-616) GtC

IPCC 2013, AR5 „Summary for Policy Makers“

27

Future: Large uncertainty, e.g., response

of terrestrial biosphere to changing climate ?

Natural CO2: Terrestrial C sinks

28

Houghton, Biologist, 2002: “Strangely, the difference between the net terrestrial sink and the emissions from land-use change suggests that there is a residual terrestrial sink, not well understood, that locked away as much as 3.0 PgC/yr during the last two decades. … The exact magnitude, location and cause of this residual terrestrial sink are uncertain, …”

Natural CO2 fluxes from in-situ obs.: Gurney et al., Nature, 2002

Gurney et al., Nature, 2002

TransCom 3 regional CO2 flux inversions Observartions:

Very accurate but sparse

Information content sources & sinks (excluding fossil fuel fluxes):

Large regions only (continents, ocean basins)

Large uncertainties (often +/- 100%)

A priori land

Within model uncertainty

Inversions:

Left / right: different inversions

Mean flux x

Data: GLOBALVIEW-2000: 1992-1996

30

NH land: Weaker sink?

(+1 GtC/yr)

Tropics: Weaker source? Net approx. zero ?

(-2 GtC/yr)

Stephens et al., Science , 2007

Natural CO2 fluxes from in-situ obs. incl. aircraft: Stephens et al., Science, 2007

31

CarboEurope findings (2009) Executive Summary of the terrestrial carbon balance (CarboEurope-IP)

• The land surface of continental Europe (the geographic region between the Atlantic coast and the Ural Mountains) is a carbon sink for CO2 of 300 Tg C/yr (0.3 GtC/yr) (as indicated by atmospheric and ground-based measurements). The estimated sink has almost doubled since 2003, mainly due to additional processes understanding.

• … • Almost 60% of the continental CO2 sink is located

outside the EU-25 in eastern Europe, mainly European Russia. …

• … • The uncertainty in the magnitude of the terrestrial

sink remains high. This is a consequence of the heterogenous landscape of Europe, and the diversity of management practices at small scale.

• …

First global regional-scale CO2 surface fluxes from GOSAT/RemoTeC

Basu et al., ACP, 2013

Chevallier et al., GRL, 2011: • TCCON-only inversion • Consistent with flask-only

but larger uncertainties

Adding GOSAT:

Shift of terrestrial net carbon uptake from tropics to (northern) extra tropics

But: 1 year only, still bias issues (e.g., land/ocean), …

Natural fluxes only as fossil fuel emissions prescribed

32 NAM EUR trASI trSAM

CO2 flux inversions using different GOSAT XCO2 products and models

Regional natural CO2 fluxes for 2010 Method: • 3 inversion methods (2x LSCE (LMDZ 19&39), 1x Univ.

Edinburgh (UoE)) • CO2 surface observations and x2 GOSAT satellite

XCO2 products: • GHG-CCI UoL (OCFP) v4 • NASA ACOS v3.3

Conclusions: Regional flux time series: • Good agreement for phase but NOT amplitude Annual regional fluxes: • Not considered realistic for all regions, e.g.,

Europe: inferred sink „significantly too large“ Possible issues / to be improved: Inversion method incl. prior fluxes and transport models, satellite data (biases to be further reduced)

33

Chevallier et al., GRL , 2014

European terrestrial carbon fluxes from SCIAMACHY and GOSAT

Reuter et al., in review

„Europe only“ inversion using STILT-based short range (days) particle dispersion modelling using an ensemble of satellite XCO2 retrievals:

• 2 satellites

• 5 retrieval algorithms / products

• New flux inversion method insensitive to observations outside Europe, large-range transport & other errors

• Various sensitivity studies -> stability & error analysis

Satellite data suggest a (TransCom region) European C sink of 1.02 +/- 0.3 GtC/yr (for 2010)

34

Satellite CT

Reduced carbon uptake in the summer of 2010 most likely due to Eurasian heat wave driving biospheric fluxes and fire emissions.

Joint inversion GOSAT & flasks: Biospheric & fire CO2 emission anomaly April–September 2010: 0.89±0.20 PgC over Eurasia

35

GOSAT/RemoTeC CO2: Northern hemisphere summer 2010 carbon fluxes

Guerlet et al., GRL, 2013

Carbon fluxes (GPP, NPP, NEP, NEE, NBP, …)

Source: IPCC

GPP RES(auto) RES(hetero) Fires, …

NEP = (minus) long-term mean of NEE NEP = GPP – RES = (-)NEE

Plants Soils (bacteria, fungi, …)

36

GHG-CCI CAR: CCDAS Carbon Cycle Data Assimilation System (CCDAS)

37

CAR, v1.1: Initial assessment by FastOpt (T. Kaminski and M. Scholze) using the reported (reliable) uncertainties as given in the CRDP#1 SCIAMACHY BESD (SCIAMACHY) and EMMA (SCIAMACHY and GOSAT merged) XCO2 products

Approach: • Optimization of biosphere model parameters

Advantage w.r.t. direct flux inversion: • May lead to improved biosphere models ->

Better climate prediction

Assessed target quantities: • regional Net Primary Production (NPP) • regional heterotrophic RESpiration (RES) • regional Net Ecosystem Production (NEP)

Findings: • Very high uncertainty reduction:

• > 50% at model grid scale • > 70% for aggregated regions

• To be assessed: impact of biases

Potential for high uncertainty reduction of NEP even when using only 1 year of SCIAMACHY XCO2

0% 50% 100%

CAR v1.1

Prior: Scholze et al., 2007 Model: BETHY-TM3

38

GOSAT/UoL CH4/CO2: Wildfire emission ratios

First fire emission ratios, ER CH4/CO2, from satellite observations: • boreal forest: 0.00603 mol/mol • tropical forest: 0.00527 mol/mol • savanna fires: 0.00395 mol/mol (uncertainties ~ +/- 0.0003)

Ross et al., GRL, 2013

Fires near Moscow, 8 Aug 2010

GOSAT FTS & CAI • Boreal • Tropical • Savanna

Anthropogenic CO2

39

SCIAMACHY XCO2 EDGAR CO2 emissions

Schneising et al., 2013

Europe

China

US

Trend [%CO2/yr]

EDGAR emissions consistent with SCIAMACHY

Regional enhancement = Source - Background

SCIAMACHY EDGAR

SCIAMACHY CO2 over anthropogenic source regions

40

SCIAMACHY CO2 & NO2 over anthropogenic source regions

Satellite XCO2e („expected

XCO2 enhancement from nearby anthropogenic emissions“) via co-located SCIAMACHY NO2 & XCO2 during 2003-2011.

Satellite XCO2e

& NO2 trends versus EDGAR emission trends

CO2-to-NO2 emission ratio trend [%/year]:

4.2 +/ 1.7 %/yr Less NO2 per CO2: trend

towards cleaner technology

East Asia

Reuter et al., in review

41

From SCIAMACHY to CarbonSat

Berlin

Germany

New capabilities: Cities, power plants, oil & gas fields, geological „point“ sources, … 42

LSCE EDGAR IUP-UB

IUP-UB

2 x 3 km2

Results

43

Satellite-derived GHG source/sink-related results:

Methane

Methane

44

Coal mining

Landfills

Termites

Wastewater

Wetlands Rice

Ruminants

Hydrates

Natural gas

Energy

Findings: • Increase ~7-9 ppb/yr (0.4-0.5%/yr) (2007-2009 relative to 2003-2006) • Mainly tropics & NH mid latitudes • No “local / regional hot spot” found • Analysis complicated by detector degradation

45

SCIAMACHY: Renewed methane growth

Schneising et al., 2011

Frankenberg et al., 2011

Tropics

NH Tropics

NH (~0- 60o)

46

SCIAMACHY & NOAA/flasks: Renewed methane growth

Findings: • Methane emissions 2007-2010: +16-20 TgCH4/yr higher compared to 2003-2005 • Atmospheric increase 2007-2010: on average ~6+/-1 ppb/yr (0.3-0.4%/yr) (relative to 2003-2006; update of global means from Dlugokencky et al., 2009) • Where?: Mainly tropics & NH mid latitudes, no significant trend for arctic latitudes • Reason for increase: Mainly increasing anthropogenic emissions • Interannual variations: Mainly wetlands & biomass burning

Bergamaschi et al., 2013

Total emissions Anthropogenic Wetlands

2007

SCIAMACHY & NOAA/flasks: Renewed methane growth

Houweling et al., ACP, 2014

47

Addresses which region contributed most to the CH4 increase since 2007:

• Two 2-year periods before and after July 2006 analyzed

• Global difference varies between 27 and 35 Tg/yr most of which is attributed to the tropics with the northern hemispheric part of this zone contributing most

• Splitting the tropics: largest portion south-east Asia (9+/-13 Tg/yr) consistent with growing demand for energy and food and rapidly growing economies (but large uncertainty)

Fraser et al., ACP, 2013

First GOSAT methane emissions published in peer-reviewed journal derived using GHG-CCI XCH4

48

GOSAT/UoL-PR CH4: Regional surface fluxes

Kirschke, Bousquet, Ciais, et al., „Global Methan Budget 2013“. 2013

Renewed methane growth: Anthropogenic or wetlands ?

Kirschke, Bousquet, Ciais, et al., 2013

49

Assessement of Climate-Chemistry Model using SCIAMACHY methane - I

Shindell et al., 2013

SCIAMACHY – (ER-2 + 1.8%)

0% 2% 4%

SCIAMACHY WFMD 2003-2005 (Schneising et al., 2011)

ER-2 Model

50 ER-2: New generation GISS climate model

Hayman et al., 2014 We found that the annual cycles observed in the SCIAMACHY measurements and at many of the surface sites influenced by non-wetland sources could not be reproduced in these HadGEM2 runs. This suggests that the emissions over certain regions (e.g., India and China) are possibly too high and/or the monthly emission patterns for specific sectors are incorrect.

Assessement of Climate-Chemistry Model using SCIAMACHY methane - II

51

SCIAMACHY: fugitive methane emissions from oil and gas production in

North American tight geologic formations

52

Schneising et al., submitted, 2014

GHG-CCI: Publications www.esa-ghg-cci.org/

Interested to see more results?

• Publications

53

Many thanks for your attention !

54