Results of NAMAP (see much more in the CPC Atlas and BAMS paper)

NAMAP2A multi-model assessment of North American Monsoon simulations

IntroductionNAMAP2 is a coordinated exercise in global and regional atmospheric modeling of the North American Monsoon System (NAMS). The summer season of 2004 (during which the NAME 2004 field campaign took place) is the simulation target. A preliminary discussion among potential participants to develop simulation protocols has just been completed and runs will be carried out over the next several months with analysis to follow immediately. The NAMAP2 project timeline is shown in the box to the right.

The first NAME Model Assessment Project (NAMAP) was an attempt to engage the modeling community in advance of the NAME 2004 field campaign. NAMAP provided an indication of the ability of numerical models to simulate atmospheric variability across southwestern North America during the summer season. For NAMAP, numerical simulations of atmospheric variability across southwestern North America during a single summer (1990) were carried out independently by six modeling groups. The NAMAP analysis focused on the ability of the models to simulate the observed seasonal and (where observations permitted) diurnal cycles. The results provided motivation for enhanced observations in data-poor areas during the NAME 2004 Field Campaign and led to the formulation of metrics to quantify model simulation quality and improvement (CPC Atlas online 2004; Gutzler et al. 2005, in press in BAMS).

NAMAP2 will re-examine the metrics proposed by NAMAP, extend the NAMAP analysis to transient variability, and exploit the extensive observational database provided by NAME 2004 to analyze simulation targets of special interest. A centralized comparative analysis of model output will be carried out at the NOAA Climate Prediction Center once the simulations are complete, and we hope that the simulations will be used by individual modeling groups as Control Runs for their own analysis purposes.

Oceanic temperature data and (for regional models) lateral boundary conditions will be specified. The first phase of NAMAP, and other studies of the North American Monsoon, have indicated the sensitivity of the results to proper specification of SSTs in the Gulf of California. Motivated by this need, a new SST analysis has been developed for NAMAP2, as described in the lower right box.

New participants are welcomed! Online registration at the NAMAP2 web page

For More Information: contact Dave Gutzler NAMAP CPC Atlas: http://www.cpc.ncep.noaa.gov/research_papers/ncep_cpc_atlas/11/index.htmlNAMAP2 web page: http://www.joss.ucar.edu/name/namap2

David S. Gutzler (and NAMAP2 collaborators)University of New Mexico

gutzler@unm.edu

Pingping Xie and Wanqiu WangNOAA/NCEP/Climate Prediction Center

pingping.xie@noaa.gov wanqiu.wang@noaa.gov

A new Multi-Platform-Merged (MPM) SST analysis for NAME and NAMAP2 (Xie & Wang)The first phase of NAMAP, and other studies of the North American Monsoon, have indicated the sensitivity of the results to proper specification of SSTs in the Gulf of California. Motivated by this need, a new SST analysis has been developed for NAMAP2 and other NAME research.

Resolution: 0.25° in space, 3-hour in time Input data: In situ observations (buoys and ships)Domain: 180°–30°W, 30°S–60°N (embedded within global OI analysis) Satellites: GOES, TMI, AMSR, NOAA-16, NOAA17Target Period: 2001–present (summer 2004 completed) Fields merged via Optimum Interpolation after Bias Correction

Mean differences (K)between analyses

Jun 1-Aug 31, 2004Input SST fields

Apr-May-Jun 2004

8/dayclear sky

2/dayclear sky

2/dayclear sky

2/dayall sky

2/dayall sky

Mean difference (K)between analyses and

in situ observationsJun 1-Aug 31, 2004

smallest bias!

operational OI analysis is too cold in the Gulf of CA

Real time global analysis tends to be too warm in the Gulf of CA

NAMAP2 Protocols and Timeline (see NAMAP2 web page for details)

Time Series of 3-Hourly SST

109.875oW; 26.125oN

Seasonal progression ofmonthly precipitation in the CORE and AZNM subdomains

Monthly-average diurnal cycleof precipitation in the

CORE and AZNM subdomains

NAMAP was initiated to provide benchmark simulations of a representative

warm season across Tier 1, prior to the NAME 2004 field campaign

Both global models exhibit delays in monsoon onset and peak precip Big discrepancies in nocturnal precipitation as well as daytime convective peak

Six modeling groups(running 4 regional models

and 2 global models) carried out the simulations of the 1990 warm season

Observed monthly precipitationestimated from in situ observations

Monthly precipitation for thepeak month of each simulation

Huge range of simulated precipitation No obvious systematic biases associated with model resolution, and repeated initialization in one of the simulations does not guarantee “best” results

All NAMAP output is freely available at http://www.joss.ucar.edu/name/namap

Monthly-average diurnal cycleof latent flux in AZNM Huge range, poorly constrained by observations. Similar results obtained for sensible flux.

NAMAP-derived goals for model improvement include:* Simulate observed monsoon onset within a week* Simulate monthly mean precipitation to within 20%* Simulate large scale surface fluxes to within 20%* Correctly simulate the diurnal cycle of convective and

nonconvective precip

Output Archiving Protocols (8 Sept 05 revision)Simulation Timeline (22 June 2005 revision)

Simulation Guidelines (22 June 2005 revision)

site lat lon site lat lon

NAME sounding sites US, Cent Amer. raobs

Puerto Penasco (ISS 2) 31.18N 113.33W Tucson (NWS) 32.12N 110.92W

Bahia Kino (ISS 3) 28.81N 111.93W Las Vegas (NWS) 36.62N 116.02W

Los Mochis (ISS 4) 25.41N 109.05W San Diego (NWS) 32.85N 117.12W

Loreto (GLASS) 26.01N 111.21W Flagstaff (NWS) 35.23N 111.82W

RV Altair (CSU) 21.49N 106.07W Albuquerque (NWS) 35.05N 106.62W

SMN sites El Paso (NWS) 31.87N 106.70W

Empalme 27.95N 110.77W Amarillo (NWS) 35.23N 101.70W

Mazatlan 23.20N 106.42W Midland (NWS) 31.95N 102.18W

Chihuahua 28.63N 106.08W Del Rio (NWS) 29.37N 100.92W

Torreon 25.53N 103.45W Yuma (ARMY) 32.51N 114.00W

Monterrey 25.87N 100.23W Phoenix (SRP) 33.45N 111.95W

Zacatecas/Guadalupe 22.75N 102.51W Belize City, Belize 17.53N 88.3W

La Paz 24.17N 110.30W

Mexico City 19.4N 99.2W

a) For spatial analysis:Archive lat-lon fields covering the NAMAP2domain every 3 hours (8/day) during the simulation period. Fields to archive:

b) For high-resolution temporal analysis:Archive "MOLTS"-style time series (at least hourly in time and full vertical resolution). We will consider surface fluxes and profiles of humidity, T, u, v, w, p, resolved and convective precipitation, cloud fraction, radiation, and turbulence at model grid points corresponding to the following NAME sounding sites: 

abbrv field abbrv field Ms sfc soil moisture Ts temperature (surface) Msub subsfc soil moisture T2m temperature (2m) Msca sfc soil moisture cap T850 temperature (850 hPa) Msub subsfc soil mois. cap T500 temperature (500 hPa) Rs sfc runoff T300 temperature (300 hPa) Rsub subsfc runoff Zs geopot ht (surface) Veg Vegetation stress Z850 geopot ht (850 hPa) SWs net sfc SW flux (+dn) Z500 geopot ht (500 hPa) LWs net sfc LW flux (+up) Z300 geopot ht (300 hPa) LHs sfc latent flux (+up) Q10m specific humid (2m) SHs sfc sensible flux (+up) Q850 specific humid (850 hPa) As albedo (surface) Q500 specific humid (500 hPa) Ap albedo (planetary) Q300 specific humid (300 hPa) SLP sea level pressure U10m zonal wind (10m) OLR outgoing LW flux (TOA) U850 zonal wind (850 hPa) Topo surface elevation (ASL) U500 zonal wind (500 hPa) CWco column condensed H2O U300 zonal wind (300 hPa) Qco column specific humid V10m merid wind (10m) QUco vert integrated QU V850 merid wind (850 hPa) QVco vert integrated QV V500 merid wind (500 hPa) CWlt sfc- 700hPa cond H2O V300 merid wind (300 hPa) Qlt sfc- 700hPa spec hum Pcon precip (convective) QUlt sfc- 700hPa cond QU Pres precip (resolved) QVlt sfc- 700hPa cond QV CAPE convective available PE CWut 700hPa- top cond H2O Cin convective inhibition Qut 700hPa- top spec hum QUut 700hPa- top cond QU QVut 700hPa- top cond QV

NAMAP2 spatial analysis will focus on uncertainties identified in NAMAP, with additional emphasis on verificationusing enhanced observations from theNAME 2004 field campaign. Moisture transport and land surface hydrologywill be assessed in detail.

NAMAP2 will extend the descriptiveanalysis of NAMAP with a process-oriented analysis of model output atNAME 2004 sounding sites. Thiscomponent of the analysis is focusedon cloud and precipitation processes.

With the recent completion of the MPM SST analysis for summer 2004 (below), we are ready to begin the NAMAP2 simulations. Analysis of the simulations will begin in several months. NAMAP2-based publications will include full joint authorship (as in NAMAP).

** Please join us! **

Time Period Event

Spring- Summer 2005 Solicit participation and develop NAMAP2 simulation protocols

Summer- Autumn 2005 Develop the prescribed global SST fields [activity at NOAA/CPC]

Autumn 2005 I ndividual modeling groups carry out NAMAP2 simulations

October 2005 Report on NAMAP2 progress at the Climate Diagnostics Workshop

Winter 2005/6 NAMAP2 analysis to be carried out at NOAA/CPC

Simulation Period 15 May- 30 September 2004 Domain of I nterest 15N- 45N 125W- 75W Lateral Boundary Conditions (for regional models) NOAA CDAS2 (to be supplied)

Surface Boundary Conditions/ oceanic

Multiple- Platform- Merged Analysis (to be supplied; see description below)

Surface Boundary Conditions/ continental Chosen by each modeling group

Participants have expressed tentative interest in repeating the analysis for the 2005 season

These differences are currently being diagnosed, but at first glance they seem quite consistent with what we would expect from an improved analysis ... as suggested by the small bias of MPM relative to in situ data.