NEWS NEWSWorld Climate Research Programme__WCRP

Vol. 13, No. 4 November 2003

Global Energy and Water Cycle Experiment

FIRST USE OF MODIS DATA TO CROSS-CALIBRATE WITHGEWEX/SRB DATA SETS

GLACE RESULTSSHOW THAT

SOIL MOISTURE'SIMPACT ON PRECIPITATION

VARIES WITH AGCM

Degree to which subsurface soil moisture varia-tions influence precipitation in nine AtmosphericGeneral Circulation Models (AGCMs), as deter-mined in the Global Land Atmosphere CouplingExperiment (GLACE). Ten panels are shownbecause one AGCM employed two different landsurface schemes. See article on Page 6.

What's New• Rick Lawford Selected as the New Director of IGPO (Page 2).

• GABLS 1D and LES Comparisons for Stable BL Provide Future Direction for Climate Models (Page 11).

• New NCEP 24-Year Regional Reanalysis Now Available (Page 16).

The estimated surface downward shortwave fluxes using MODIS information for July 2001, as well as those derivedwith the GEWEX/Surface Radiation Budget (SRB) Project model with International Satellite Cloud ClimatologyProject (ISCCP) D1 data. The comparison shows that the two approaches driven with independent input data yieldsimilar global patterns. See article on Page 4.

November 20032

Contents PAGE

Commentary: Richard G. Lawford is the 2New Director of IGPO

Update on the GEWEX Americas Prediction 3Project (GAPP)

First Use of Modis Data to Cross-Calibrate 4with GEWEX/SRB Data Sets

First Results from GLACE 6

African Monsoon Multidisciplinary Analysis 8(AMMA) Project

Workshop/Meeting Summaries:GLASS Science Panel Meeting 9GABLS Workshop on Stable Boundary 11 Layers9th Annual Meeting of GHP and Five 13 Focused Workshops

New GLASS Chair 11

GEWEX Flies Aboard the Shuttle 15

Meetings Calendar 15

New 24-Year Regional Reanalysis 16

COMMENTARY

RICHARD G. LAWFORD IS THE NEWDIRECTOR OF IGPO

David Carson, DirectorWorld Climate Research Programme

Soroosh Sorooshian, ChairmanGEWEX Scientific Steering Group

We are pleased to an-nounce tha t Richard G.Lawford has been selectedas the new Director of theInternational GEWEX ProjectOffice (IGPO). We con-sider the GEWEX researchcommunity fortunate to haveRick step in after the out-standing leadership of Paul

Try as IGPO director for over a decade.

Rick has many years of experience with GEWEX,beginning in 1991 as Chief of the HydrologicalProcesses Division at the Canadian Climate Centre,when he helped launch the Mackenzie GEWEXStudy (MAGS), which has led to findings aboutthe water balance in the Mackenzie River Basin andthe role of freshwater fluxes into the Arctic Ocean.

In 1995, Rick moved to NOAA’s Office ofGlobal Programs in Silver Spring, Maryland, USAto become Program Manager of the GEWEX Con-tinental-scale International Project (GCIP). ThroughRick’s oversight, GCIP advanced the forecast ca-pabilities of the National Weather Service and broughttogether the atmospheric and hydrological commu-nities to work on coupled modeling studies. Heoversaw the successful transition of GCIP into theGEWEX Americas Prediction Project (GAPP).

Since 2001, Rick has also served as part-timeDirector of the Global Water Cycle Program Of-fice, an interagency water cycle program that hehelped establish, and which serves as a US liaisonto a number of international organizations, includ-ing UNESCO and the Integrated Global ObservingStrategy (IGOS)-Partnership (P). As Director, hecoordinated the development of the IGOS-P WaterCycle Theme of which the Coordinated EnhancedObserving Period (CEOP) is the first element.

We both have known and worked with Rick onGEWEX related activities for nearly 8 years andlook forward to working with him as IGPO Direc-tor. With over 30 years of experience in research,science policy, applied and operational meteorol-ogy, and his extensive knowledge of GEWEX andinternational and interagency science programs, Rickwill ensure strong leadership for IGPO.

We are also delighted to report that DawnErlich will continue to support GEWEX as theEditor of GEWEX News, webmaster, publicationsspecialist, and coordinator of project activities andmeetings. Joseph Trowell will also continue toprovide graphic design, publication, meeting, andadministrative support.

We will continue to benefit from the part-timeinvolvement of Paul Try and Robert Schiffer forscientific and programmatic support, and SamBenedict with primary responsibilities as Interna-tional Coordinator for CEOP. Gilles Sommeria,Senior Scientific Officer of the WCRP Joint Plan-ning Staff, continues to be the primary focal pointfor GEWEX at WCRP in Geneva. The range ofGEWEX activities is a long one and we are happythat with the appointment of Rick as IGPO direc-tor, we have a very strong team to face thechallenges of GEWEX Phase II.

3November 2003

UPDATE ON THE GEWEX AMERICASPREDICTION PROJECT (GAPP)

Rick Lawford1, Jin Huang2, Jared Entin3

1UCAR/NOAA Office of Global Programs2NOAA Office of Global Programs

3NASA Earth Science Enterprise

The GEWEX Americas Prediction Project (GAPP)has been gaining momentum as a number of projectshave been funded over the past 2 years throughthe NOAA Office of Global Programs (OGP) andthe NASA Earth Science Enterprise to support thisinitiative. The GAPP initiative has two primaryobjectives, namely:

• Develop and demonstrate a capability to makereliable monthly and seasonal predictions ofprecipitation and land surface variables throughimproved understanding and representationof the land surface and related hydrometeo-rological and boundary layer processes inclimate prediction models.

• Interpret and transfer the results of improvedseasonal predictions for the optimal manage-ment of water resources.

GAPP is steadily gaining the stature and namerecognition of its predecessor, the GEWEX Conti-nental-scale International Project (GCIP), and isproducing data sets and model improvements thatwill be a legacy for it and its sponsors. WhileGAPP is proceeding with a focus on land-atmo-sphere interactions and hydrology similar to GCIP,it has a larger geographical domain (the contiguousUSA) and a clearer focus on addressing intraseasonalto interseasonal prediction. It encompasses processstudies in complex regions involving land and oceanforcing of the atmosphere and in the complex ter-rain of the western Cordillera. New opportunitiesare being developed for linking GAPP with globaldomain studies through the Coordinated EnhancedObserving Period (CEOP) and to areas such as theLa Plata River Basin through transferability studies.In addition, the heritage data sets and model devel-opments realized through GCIP are being used inGAPP for model development. GAPP continuesto stress the development of products for applica-tion in the water resource sector. Within GAPP,a NOAA Core project is playing a central role infacilitating science developments to reach opera-tional prediction services.

One of the newest developments is the releaseof the Regional Reanalysis products (see figure at

the top of Page 16). The Regional Reanalysis Projectis using a recent version of the NOAH land surfacemodel, along with the Eta model to produce a 25-year data set of reanalysis products (1979–2004).This product is being generated at 32 km resolutionfor the domain of North America. A unique featureof this product is the precipitation assimilation schemethat allows both surface and atmospheric conditionsto be updated by precipitation observations ratherthan the model precipitation output. Preliminary as-sessments indicate that this approach will produceprecipitation products (and related fields) that will besubstantial improvements over the Global Reanalysisprecipitation and other water cycle products. Just asthe Global Reanalysis products were widely used, itis anticipated that these products will be used exten-sively and will lead to many new research findingsabout interseasonal and interannual variability overNorth America.

GAPP is expected to make major contributionsto the US Climate Change Science Program throughits Global Water Cycle activity. The GAPP imple-mentation plan will include a number of initiativesthat will directly address water cycle priorities. Forexample, special mountain hydrometeorology activi-ties being planned for the western Cordillera willhelp to address issues related to complex terrainand its effects on precipitation and runoff. Moregenerally, GAPP will be a major contributor inaddressing questions related to “uncertainties inseasonal and interannual predictions of water cyclevariables” and “improvements to global and regionalmodels to reduce these uncertainties.”

In addition to the on-going opportunities GAPPoffers for working with other Continental ScaleExperiments through the the GEWEX Hydrometeo-rology Panel (Baltic Sea Experiment, GEWEX AsianMonsoon Experiment, Large Scale Biosphere-Atmosphere Experiment in Amazonia, MackenzieGEWEX Study, Murray-Darling Basin), CEOP isproviding new opportunities for involvement ofthe GAPP science community in both domesticand international priorities. This effort, which cur-rently involves the assemblage of data sets for theperiod of July 2001 to December 2004, is involvinga number of US institutions, including some thatare not associated with GAPP and some agenciessupported directly by GAPP. GAPP’s contribu-tions to this program include support for theproduction of regional and global model outputs(National Centers for Environmental Prediction) andfor the CEOP data management module, and coor-dination and formatting of reference site data from

November 20034

FIRST USE OF MODIS DATA TOCROSS-CALIBRATE WITHGEWEX/SRB DATA SETS

Rachel Pinker1, Hengmao Wang1,Michael King2, and Steven Platnick2

1Department of Meteorology, University ofMaryland, 2NASA/Goddard Space Flight Center

Many attempts have been made to use satellitedata to estimate surface shortwave fluxes at bothregional and global scales. The emphasis has been onthe use of geostationary satellites to capture the diur-nal variability in cloud distributions that determines theamount of energy received at the surface during thecourse of a day. Due to their instrument configura-tions, many of these satellites are limited in theircapability to accurately detect cloud or aerosol opticalproperties that are important elements of the radiationbudget. Polar orbiting satellites tend to have higherspatial resolution than the geostationary satellites, aswell as more spectrally resolving bands. The Moder-ate Resolution Imaging Spectroradiometer (MODIS)instrument onboard the Terra and Aqua satellites, astate-of-the-art sensor with 36 spectral bands andonboard calibration of both solar and infrared bands,has the capability for measuring atmospheric and sur-face properties with higher accuracy and consistencythan previous Earth observation imagers. Therefore,it is important to utilize such observations to evaluatethe performance of algorithms applied to satellites thatdo not provide direct information on cloud and aerosoloptical characteristics.

To enable the use of independently derived op-tical parameters from multi-spectral satellite observations,the GEWEX/Surface Radiation Budget (SRB) Projectmodel (Pinker et al., 1995; 2003) has been modifiedfor use with observations from the Advanced EarthObservation Satellite (ADEOS II) Midori II and testedpreviously with proxy data (Wang et al., 2002). Inthe present study, it was implemented with data fromMODIS, which provide parameters similar to thosethat will be forthcoming from GLI onboard Midori II,and that have already been extensively tested andreprocessed (King et al., 2003; Platnick et al., 2003).

For the modified version of the inference scheme,the parameters needed to drive the SRB are: viewinggeometry, column water vapor, column ozone amount,cloud fraction, cloud optical thickness, aerosol opti-cal depth and spectral surface albedo. In the currentstudy, monthly mean MODIS data for July, Augustand September of 2001 were used. The monthlymean Cloud Fraction Total, Cloud Optical ThicknessCombined, Optical Depth Land And Ocean aerosol,

the GAPP area of study (University Corporationfor Atmospheric Research). Some of these effortsalso support the Integrated Global Observing Strategy(IGOS)-Partnership (P) as CEOP is the initial projectwithin the emerging IGOS-P Water Cycle theme.

One of the new developments within GAPP hasbeen the initiation of cooperative activities with theClimate Variability and Predictability/Pan AmericanClimate Studies Program and the contributions ofthose new projects to understanding the NorthAmerican Monsoon System. The North AmericanMonsoon Experiment is a concerted effort to ad-dress the monsoon system by studying three tiersacross North America. GAPP goals are best ad-dressed through Tiers II (southern USA) and III(covers entire contiguous USA) although there arealso potential benefits from process studies fromthe field phase for Tier I (Gulf of California andenvirons). In addition, a new thrust within NOAAOGP, known as the Climate Prediction Programfor the Americas, is aimed at consolidating effortsrelated to Intraseasonal-Seasonal Predictions, andwill influence some of GAPP’s future priorities.

GAPP programmatics have become quite com-plex due to collaborations with other programs,coordination of activities with multiple agenciesand commitments to global scale WCRP programs.For this reason GAPP program management hasfelt an increasing need for stronger scientific guid-ance in managing the program. To this end aScientific Advisory Group (SAG), chaired by Dr.Paul Houser of NASA’s Goddard Space FlightCenter, has been formed to provide advice anddirection to the program managers for GAPP. TheSAG will also interface with the US Water CycleScientific Steering Committee on relevant issues.

GAPP offers the US hydrometeorological sci-ence community an opportunity to become involvedin addressing major climate issues that will becentral to the interseasonal and interannual predic-tion activities of NOAA and also to support relatedNASA goals. In addition to funding meritoriousprojects, GAPP management wants to work withscientific groups and individuals who wish to ana-lyze GAPP data and utilize GAPP infrastructure ina way that contributes to GAPP’s goals. Anyonewith an interest in contributing to this ambitiouseffort is invited to dialogue with Drs. Jin Huangand Jared Entin about the possibilities.

5November 2003

the latest collection from four algorithms. The miss-ing aerosol optical depths over arid areas were filledfrom the MODIS-Global Ozone Chemistry AerosolRadiation Transport (GOCART) integrated monthlyaerosol optical depth data provided by school ofEarth and Atmosphere Sciences, Georgia Institute ofTechnology (Yu et al., 2003). Missing cloud opticalthickness values were replaced by interpolated values.

The spectral surface parameters used to calcu-late spectral surface albedo were taken from theMODIS Bidirectional Reflectance Distribution Func-tion (BRDF) and albedo product (MOD43B) at the0.25° resolution (Lucht et al., 2000; Schaaf et al.,2002). The three weighting parameters associatedwith the RossThickLiSparse Reciprocal BRDF modelthat best describes the anisotropy of each pixel areprovided for each of the MODIS spectral bands aswell as for the three broad bands (0.3–0.7 µm,0.7–5.0 µm, and 0.3–5.0 µm). For the two broadbands (0.3–0.7 µm, 0.7–5.0 µm), these parameterswere used with simple polynomials to estimate thewhite sky albedo and the black sky albedo for themonthly mean solar zenith angle.

The estimated surface downward shortwave fluxesusing the above MODIS information, as well as thosederived with the GEWEX/SRB model using the Inter-national Satellite Cloud Climatology Project (ISCCP)D1 data (Rossow and Schiffer, 1991; 1999), areshown for July 2001 in the figure at the top of Page1. For the ISCCP D1 data, the monthly mean fluxeswere obtained by averaging the daily means that wereobtained by integration of 3-hourly instantaneous fluxes.To make the results from the two data sets coherent,when using the MODIS parameters, the monthly meanviewing geometry was used with the instantaneousobservations, multiplied by day length and divided by24 hours to get the monthly mean values. Thecomparison shows that the two approaches drivenwith independent input data, yield similar globalpatterns. Over land, especially over Africa, esti-mates of fluxes from MODIS are higher than thosefrom ISCCP D1, whereas over oceans, fluxes fromISCCP D1 tend to be larger than those from MO-DIS. This may be attributed to the cloud amountdifferences between these two data sets, as well asdifferences in cloud optical depth, as illustrated in thefigure at the bottom of Page 16. In general, surfaceshortwave fluxes from ISCCP D1 are larger thanthose from MODIS. The lower temporal resolutionof polar orbit satellites as compared to geostationarysatellites may also contribute to these differences.Additional work is needed to reconcile the two.

To benefit from the detailed cloud informationprovided by MODIS, cloud types and their corre-sponding optical depth will be incorporated into thenext version of the inference scheme.

The ability to derive global radiative fluxes fromthe same high quality instrument is important formany new research initiatives. For example, underthe Coordinated Enhanced Observing Period (CEOP)activity (Grassl, 2002) that started in the summer of2001, issues to be addressed include those related towater and energy fluxes over land areas, monsoonalcirculations, extension of derived products (e.g., ISCCP,GPCP, SRB, GVaP, ISLSCP) from operational sat-ellites and validation from new satellite systems. UnderCEOP, simultaneous observations will be collectedover several GEWEX Continental Scale Experiment(CSE) sites, as well as over additional regions ofclimatic significance. This will allow testing thetransfer of techniques and models between differentcontinental-scale regions and predicting water-relatedparameters at scales suitable for assimilation intoglobal NWP models. CEOP will also provide anopportunity to evaluate the usefulness of ongoingoperational satellites and new generation experi-mental satellites in hydrological research, and toimprove NWP and climate predictions. The ra-diative fluxes derived from MODIS could serveas a calibration reference for estimates derivedfrom other satellites, as well as a means ofcross-calibrating between the many different sen-sors in use.

Acknowledgments

The work on surface radiative fluxes was sup-ported under grants NAG59634 from NASA EOD/IDS and RDC101GC1 from the National Space De-velopment Agency of Japan to the University ofMaryland. The ISCCP D1 data were obtained fromthe NASA Langley Research Center AtmosphericSciences Data Center. The authors acknowledge thedevelopers of the GrADS software.

References

Grassl, H., 2002. CEOP-Global Monitoring for Improved Predic-tion. CEOP Newsletter No. 2.

King, M. D., W. P. Menzel, Y. J. Kaufman, D. Tanré, et al.,2003. Cloud and Aerosol Properties, Perceptible Water, and Pro-files of Temperature and Humidity from MODIS. IEEE Trans.Geosci. Remote Sens., 41, 442-458.

Lucht, W., C. B. Schaaf, and A. H. Strahler, 2000. An Algorithmfor the Retrieval of Albedo from Space Using Semi-empiricalBRDF Models. IEEE Trans. Geosci. Remote Sens., 38, 977-998.

Total Ozone, and Atmospheric Water Vapor weretaken from the 1° x 1° resolution Level-4 MODISatmosphere monthly global product processed with

(Contined on Page 15)

November 20036

FIRST RESULTS FROM GLACE(Impact of Soil Moisture on Precip Prediction)

Randal Koster1, Zhichang Guo2,and Paul Dirmeyer2

1NASA/Goddard Space Flight Center2Center for Ocean-Land-Atmosphere Studies

To what extent do land surface moistureand temperature states affect the evolution ofweather and the generation of precipitation?How does a human-induced change in landcover affect local and remote weather, if atall? Such questions lie at the heart of muchrecent climate research. Atmospheric General Cir-culation Models (AGCM) are often used to addressthese questions, and the results can be enlighten-ing. Nevertheless, the results are also oftenmodel-dependent, strongly reducing their signifi-cance. In particular, AGCMs are known to havedifferent degrees of “land-atmosphere couplingstrength” (Koster et al., 2002), that is, differentdegrees to which the atmosphere responds to anoma-lies in land surface state. Models with differentinherent coupling strengths can lead to vastly dif-ferent conclusions about climate sensitivity, to theconsternation and confusion of the scientific community.

The strength of land-atmosphere coupling in anAGCM is central to land impact studies but isneither well understood nor well documented. Thevast majority of AGCM studies appear to acceptthe model’s implicit land-atmosphere coupling strengthon faith, not addressing either its realism or how itcompares with that in other models. This is argu-ably a major deficiency in the current state of thescience. The quantification and documentation ofthe coupling strength across a broad range of mod-els would be valuable, if only to serve as a frameof reference when interpreting the experimental resultsof any particular model. This quantification anddocumentation is indeed the goal of the Global LandAtmosphere Coupling Experiment (GLACE), anexperiment jointly sponsored by the GEWEX GlobalLand Atmosphere System Study and the ClimateVariability and Predictability (CLIVAR) WorkingGroup on Seasonal-to-Interannual Prediction.

Each AGCM group participating in GLACEperforms the same experiment, a series of borealsummer simulations designed to isolate the in-herent coupling strength in the model used.Three ensembles of simulations are produced,as follows:

• Ensemble W: Sixteen 92-day simulationsspanning June 1– August 31 using prescribedSea Surface Temperatures (SST) from a par-ticular year of interest.

• Ensemble R: Sixteen simulations spanningthe same time period and using the sameSSTs, but with the following twist: all simu-lations are forced to maintain the samegeographically varying time series of landsurface prognostic variables (e.g., soil mois-tures, surface and subsurface temperatures).This is achieved by replacing, at every timestep, the prognostic variables’ values withthose produced at that time step by a par-ticular member of Ensemble W.

• Ensemble S: The same as Ensemble R, ex-cept that only the subsurface soil moistureprognostic variables are forced to be iden-tical amongst the member simulations.

Note the implications of this experimental de-sign. In Ensemble R, the atmospheres in all membersimulations see the same time-varying, spatially varyinganomalies of temperature and moisture at the landsurface. If all members of Ensemble R produce,for example, a large rainfall in Spain in the middleof July and very little rainfall there during the restof the period, and if similar coherence amongstsimulations is absent in Ensemble W, then we canconclude that the rainfall signal there is stronglytied to – and indeed was induced by – the landsurface state. In other words, the degree of land-atmosphere coupling (from either local or remotestates) is very high. Conversely, if the membersof Ensemble R show no coherence in their precipi-tation signal, we can conclude that atmospheric chaosoverwhelms the land surface signal and thus thatthe land-atmosphere coupling strength in the modelis very low.

Ensemble S has important implications for sea-sonal forecasting. Subsurface soil moisture is theland state that, during summer, has the greatestmemory and thus the greatest potential for contrib-uting to seasonal forecasts. Ensemble S is designedto quantify the responsiveness of the atmosphereto this potentially predictable land variable.

Fifteen AGCM groups expressed an interest inparticipating in GLACE when it was introduced inFebruary of 2003. At present, nine groups havecompleted the simulations. Two others are ac-tively running them, giving an expected total of atleast 11 participating models.

7November 2003

The coherence of precipitation among the mem-bers of a simulation is measured here as Ω, aderived parameter that varies from 0 (implying acomplete lack of coherence) to 1 (implying maxi-mal coherence, such that all simulations produceexactly the same time series of precipitation). SeeKoster et al., 2002, for details on its calculation.In essence, Ω is a measure of the ratio of “signal”to “total” variance. By subtracting the Ω value forEnsemble W from that of Ensemble R, we isolatethe impact of the land state on this coherence fromthat of all other influences, such as variations inthe SST field. In other words, we objectivelyquantify the land-atmosphere coupling strength.

Similarly, by subtracting the Ω value for En-semble W from that for Ensemble S, we quantifythe impact of subsurface moisture on the evolutionof precipitation. The figure on the bottom of Page1 shows the Ω(S) – Ω(W) differences over NorthAmerica for nine of the participating GLACE mod-els. Precipitation data were aggregated to 6-daytotals prior to computing the statistic. The plottedparameter thus effectively shows the fraction ofthe total precipitation variance (for 6-day totals)explained by subsurface soil moisture anomalies.

Notice the strong model dependence in thedegree to which soil moisture controls precipi-tation. The NASA Seasonal-to-Interannual PredictionProgram (NSIPP) model appears to be an outlier,having a somewhat stronger soil moisture–atmo-sphere connection in the central United States thanthe other models. The Australia Bureau of Meteo-rology Research Centre (BMRC) and NOAA NationalCenters for Environmental Prediction (NCEP) models,on the other hand, have a very weak coupling –for these models, atmospheric chaos overwhelmsthe soil moisture contribution to the precipitationsignal. (Note that the blue shading in the panelsindicates a certain level of sampling error in thegeneration of the statistic; a roughly equivalentamount of light yellow shading can be ignored.)The feedback strength at high latitudes in the AustraliaCommonwealth Scientific and Industrial ResearchOrganization (CSIRO) model appears to be sensi-tive to the land surface scheme used. Work isproceeding to explain the spatial patterns of themodels’ signals, relating them, for example, tovariations in hydrological regime.

Which model best represents nature? We can-not say. The GLACE project unfortunately cannotaddress the realism of simulated coupling strength,

since direct measurements of land-atmosphere in-teraction at large scales simply do not exist. Again,GLACE’s main aim is to document land-atmo-sphere coupling strength across a broad rangeof models, to allow individual models to becharacterized as having a relatively strong, in-termediate, or weak coupling. Only when thisfundamental characteristic of an AGCM is quan-tified can a land impact modeling study beproperly interpreted and understood in thecontext of parallel modeling studies. Note thatas models change and evolve, the GLACE experi-ments can be re-run easily, and the inherent couplingstrength of the newer model version can be putimmediately into context.

GLACE results (which, by the way, will alsofocus on the land’s connection to air temperature)indeed highlight a very uncertain aspect of AGCMmodeling, an aspect that demands attention frommodel developers. By improving the realism of themechanisms controlling land-atmosphere couplingstrength (e.g., moist convection, boundary layerstructure, and evaporation), modelers can hope tohave more confidence in the coupling strength theysimulate, even if this coupling strength cannot bemeasured in nature. Hopefully, the broad disparityshown in the figure on Page 1 will diminish asmodels improve.

Further details regarding GLACE may be foundat http://glace.gsfc.nasa.gov/.

Acknowledgments

For the generation of the figure on Page 1, weacknowledge invaluable contributions from the fol-lowing participants: Tony Gordon and SergeyMalyshev (Geophysical Fluid Dynamics Laboratory);Yongkang Xue and Ratko Vasic (University ofCalifornia, Los Angeles); David Lawrence, PeterCox, and Chris Taylor (Hadley Centre); BryantMcAvaney (BMRC); Sarah Lu and Ken Mitchell(NCEP); Diana Verseghy and Edmond Chan (Ca-nadian Centre for Climate Modelling and Analysis);Ping Liu (NSIPP); and Eva Kowalczyk and HarveyDavies (CSIRO).

Reference

Koster, R. D., P. A. Dirmeyer, A. N. Hahmann, R. Ijpelaar, L.Tyahla, P. Cox, and M. J. Suarez, 2002. Comparing theDegree of Land-atmosphere Interaction in Four AtmosphericGeneral Circulation Models. J. Hydromet., 3, 363-375.

November 20038

AFRICAN MONSOONMULTIDISCIPLINARY ANALYSIS

(AMMA) PROJECT

Thierry Lebel1, Jean-Luc Redelsperger2, andChris Thorncroft3

1LTHE, Grenoble, France, 2CNRM, Toulouse,France, 3SUNY at Albany, USA

Recognizing the societal need to develop strate-gies that reduce the socioeconomic impacts of thevariability of the West African Monsoon (WAM),the African Monsoon Multidisciplinary Analysis(AMMA) Project will facilitate the multidisciplinaryresearch required to provide improved predictions ofthe WAM and its impacts on daily-to-decadal timescales.It will promote international coordination of relevantongoing activities, necessary basic research and amulti-year field campaign over West Africa (seefigure below) and the tropical Atlantic to supportthis research and achieve the following objectives:

1) Improve our understanding of the WAM andits influence on the physical, chemical andbiological environment regionally and globally.

2) Provide the underpinning science that relatesclimate variability to issues of health, waterresources and food security for West Africannations and defining relevant monitoringstrategies.

3) Ensure that the multidisciplinary research iseffectively integrated with prediction anddecision-making activity.

AMMA will extend the work done in theGEWEX Continental-Scale Affiliate, Couplage del’Atmosphère Tropicale et du Cycle Hydrologique(CATCH), and promote research on four inter-acting spatial scales: (i) global scale, whereconsideration is given to how the WAM interacts

with the globe; (ii) regional scale, where monsoonprocesses are emphasized including scale interac-tions and the coupled land-ocean-atmosphere system;(iii) mesoscale, where convective systems and pro-cesses operating at the catchment scale are emphasized;and (iv) local scale where hydrology and links withapplications including agriculture are emphasized.

The interannual and interdecadal variability ofthe WAM is well documented and has motivatedconsiderable research in this area. The dramaticchange from wet conditions in the 1950s and '60sto much drier conditions in the 1970s, '80s and '90sover the whole region represents one of the stron-gest interdecadal signals on the planet in the 20th

century. Superimposed on this, marked interannualvariations have resulted in extremely dry years withdevastating environmental and socio-economic impacts.

Further motivation for research concerned withthe WAM and its variability comes from recognizingthe role of Africa on the rest of the world. Latentheat release in deep cumulonimbus clouds in theIntertropical Convergence Zone (ITCZ) over Africarepresents one of the major heat sources on theplanet. Its annual migration and associated regionalcirculations impact other tropical regions, as exem-plified by the known correlation between Sahelianrainfall and Atlantic hurricane frequency.

The West African region is the world’s largestsource of aerosols. The emissions are modulated bythe activity of the WAM but, in contrast to othersurface impacts, they feed back directly on climate.The biogeneic and anthropogenic chemical emissionsand transport over the African region are also con-trolled by the WAM and they impact the globalchemical cycles.

AMMA is planned to be a multi-year project andwill involve three observing periods. It should beunderlined here that the enhancement of observa-

The AMMA observing region over West Af-rica with the location of radiosounding sta-tions (including examples of key stations thatneed reactivating for AMMA), operationalradar and possible airports for aircraft op-erations. Also indicated in the figure is theregion where surface and atmospheric ob-servations will be enhanced along a climatictransect (right) that includes three mesoscalesites that form part of the CATCH hydro-logical project (http://www.lthe.hmg.inpg.fr/Web_Catch/accueil_Catch_fr.html).

9November 2003

tions during these different periods will provide aunique opportunity to determine future operationalmonitoring necessary to improve weather and climateforecasts over the West African region.

The Long-term Observing Period (LOP) isconcerned with observations of two types: (i)unarchived historical observations to study interannual-to-decadal variability of the WAM; and (ii) additionallong-term observations (2002–2010) to documentand analyze the interannual variability of the WAM.

The Enhanced Observing Period (EOP) (2004–2006) is designed to serve as a link between theLOP and the SOP (below). Its main objective is todocument over a climatic transect the annual cycleof the surface conditions and atmosphere and tostudy the surface memory effects at the seasonal scale.

The Special Observing Period (SOP) will fo-cus on detailed observations of specific processesand weather systems at various key stages of therainy season during three periods in the summer of2006: (i) monsoon onset (15 May–30 June); (ii)Peak monsoon (1 July–14 August); and (iii) latemonsoon (15 August–15 September).

Satellite observations will strongly contributeto the objectives of the project by providing keyvariables of the surface/atmosphere system (e.g.,Meteosat/MSG, ENVISAT, TRMM, AURA, AQUA-Train, TERRA, SMOS). It is a major challenge toexploit this large volume of data (20 years forMeteosat, for example) by optimizing the retrievalsand data analysis for monitoring, as well as valida-tion of models and assimilation.

Motivated by the strong societal and scienceissues national and international efforts are under-way to help mobilize the extra funding needed toachieve all the AMMA aims. The AMMA web pagesbelow will provide more information about the AMMAproject and how it is progressing nationally andinternationally. Contact e-mails in Europe, the USAand Africa are also included below.

AMMA website: http://medias.obs-mip.fr/amma/AMMA-US website: http://www.joss.ucar.edu/amma

Contacts in Europe:Jean-Luc Redelsperger <jean-luc.redelsperger@meteo.fr>Doug Parker <doug@env.leeds.ac.uk>Thierry Lebel <Thierry.lebel@hmg.inpg.fr>Andreas Fink <fink@meteo.uni-koeln.de>Contacts in the USA:Peter Lamb <plamb@ou.edu>Chris Thorncroft <chris@atmos.albany.edu>Contacts in Africa:Cherif Diop <Cherifdiop@hotmail.com>Abou Amani <amani@sahel.agrhymet.ne>Leykan Oyebande <lekanoye@hotmail.com>AMMA African Network <ammanet@medias.cnes.fr>

WORKSHOP/MEETING SUMMARIES

The GEWEX Global Land-Atmosphere SystemStudy (GLASS) Science Panel meeting was held inconjunction with the Project for Intercomparison ofLand-surface Parameterization Schemes (PILPS) SanPedro/Sevilleta kick-off workshop. The meetingwas hosted by the Sustainability of semi-Arid Hy-drology and Riparian Areas (SAHRA) Project, theNational Science Foundation Science, and the Tech-nology Center at the University of Arizona (UA).

GLASS is the element of the GEWEX Model-ing and Prediction Panel (GMPP) concerned withsimulation of the terrestrial state variables and fluxesusing land surface schemes (LSS) and the interac-tions between land and atmosphere. GLASS isostensibly divided into four actions that constitute atwo-by-two matrix; one axis being coupled (land-atmosphere) versus offline (land-only) modeling, andthe other being local (point, plot and catchmentscale) versus large-scale (continental to global) modeling.

PILPS is the local uncoupled component ofGLASS. Much of the workshop was given to theplanning and assessment for the PILPS San Pedro/Sevilleta experiment, set to begin in October. PILPSSan Pedro/Sevilleta is a multi-model comparison inthe PILPS vein, led by Luis Bastidas (Utah StateUniversity), Hoshin Gupta and Bart Nijssen (UA),and Eric Small (University of Colorado). PILPSSan Pedro/Sevilleta is a unique experiment in threeways. It is the first PILPS study in a semi-aridregion, with study sites in Arizona and New Mexico.Second, code for performing a multi-criteria opti-mization procedure developed by Gupta and Bastidaswill be made available to all model participants forcalibration of their schemes. Third, the experimentwill be a direct test of spatial transferability ofsurface parameters, as is commonly practiced butnot validated in weather, climate and hydrologicmodels. Terri Hogue (University of California, LosAngeles) also presented some test-bed results forthe experiment using the NOAH LSS. The websitefor the experiment is http://www.sahra.arizona.edu/pilpssanpedro/.

GLASS SCIENCE PANEL MEETING(Recent Results and New Initiatives)

25–27 August 2003Tucson, Arizona USA

Paul DirmeyerCenter for Ocean-Land-Atmosphere Studies

November 200310

observed from remote sensing platforms. Formore information about GSWP-2, see http://www.iges.org/gswp/.

Randy Koster (NASA/GSFC/GMAO) andZhichang Guo (COLA) compiled an update of theGlobal Land-Atmosphere Coupling Experiment(GLACE), a joint project with the CLIVAR Work-ing Group on Seasonal-to-Interannual Prediction. Theexperiment aims to quantify the strength of land-atmosphere coupling and its variability among mostof the major climate models in use around theworld. Integrations from five GCMs had beencompleted, with a total of 10–12 expected to beavailable by the end of September. Preliminaryresults showed the expected spread among modelsof the sensitivity of simulated precipitation to theland surface thermal and moisture state. (SeeGLACE article on Page 6.)

Proposals from Paul Houser and Christa Pe-ters-Lidard (NASA/GSFC/HSB) were heard regardingthe local coupled action. A workshop was held inDe Bild, Netherlands last year to initiate progressin this area, which is responsible for exploring thelocal interplay between the land surface and plan-etary boundary layer (PBL) with an eye towardimproved assimilation of both surface and atmo-spheric data. Houser argued the need for a fieldstudy to isolate conditions when land-atmosphereinteractions are most important, to study whetherthe absence of coupling in PILPS-type experimentscompromises their general applicability, and to dis-cern the important differences between land dataassimilation in coupled versus offline configurations.Peters-Lidard proposed a project for testing coupledLSS-PBL models. The panel encouraged collabo-ration with the GEWEX Atmospheric Boundary LayerStudy (GABLS). Peters-Lidard and Bart van derHurk will write a white paper to motivate an imple-mentation workshop.

GLASS is also addressing crosscutting issues inthe climate modeling community. One involves thecontinuing problem of initialization of soil wetness inclimate models, and the lack of transferability ofsoil moisture data sets from one model to another.GLASS is preparing a summary paper on the issueto educate the broader modeling community as tothe pitfalls of treating soil moisture as a uniformlydefined quantity across models, with some proof-of-concept simulations to illustrate the point.

Another issue of concern to the panel is theapplication of LSSs in regional models. The re-

Updates on other PILPS activities, past andfuture, were also presented. Nicolas Viovy(Laboratoire des Sciences du Climat et del’Environnement) presented a summary of re-sults from the PILPS C-1 experiment, whichwas the first to compare the abilities of LSSswith explicit carbon cycles. A well monitoredand documented wooded EuroFlux site in the Neth-erlands was used as the setting for the modelingexperiment. Participants were asked to simulatethe growth of the forest stand over the last 100years (from the time it was planted at a previouslycleared site), and then to simulate the fluxes ofwater, energy and carbon over the last few years,when intensive monitoring has been in place. Someof the more outstanding results include thedivergence among the dynamic vegetation mod-els in their simulation of biomass growth andcarbon fixation over the last century while simu-lating similar net ecosystem exchange, and thehappy fact that incorporation of carbon compo-nents into LSSs appeared to have no deleteriouseffect on their simulation of the surface energybalance. Han Dolman (Vrije Universiteit Amsterdam)gave an overview of FluxNet sites that could beused for follow-on studies that could examine thelink between carbon and water.

Ann Henderson-Sellers (Australian Nuclear Sci-ence and Technology Organization) presented asummary of the PILPS program to date, and putforward a proposal for an isotopic PILPS study inthe future, where the ability of LSSs to simulatethe fluxes of traceable chemical constituents couldbe tested. The panel encouraged a proof-of-con-cept exercise to be conducted, with an eye towardsmarrying carbon and water isotopic modeling, aswell as nutrients and other passive tracers.

There were several reports on projects that arecurrently underway. Paul Dirmeyer and Xiang Gao[Center for Ocean-Land-Atmosphere Studies(COLA)] reported on progress with the second GlobalSoil Wetness Project (GSWP-2), which is a multi-model large-scale offline simulation of the landsurface state variables and fluxes over the period1986-1995. Over a dozen modeling groups arecompleting baseline simulations, and will be sub-mitting their results to the project’s Inter-ComparisonCenter (ICC) at the University of Tokyo by theend of October. One of the new elements inGSWP-2 is a focus on remote sensing applica-tions. The project will test the ability of multipleLSSs to simulate infrared (thermal) and micro-wave (soil wetness) brightness temperatures as

11November 2003

gional modeling community is not well connectedto the global modeling community who seem to domuch of the land model development work. LSSsin regional models are often not well initialized, andthe panel perceives that the regional modeling com-munity underestimates the severity of this problem.This is particularly worrisome because of the ques-tion of scale – the same parameterizations are usedbut land surface errors are a more acute problemfor regional models as they cannot perform a self-contained spin-up. A workshop on the topic isbeing considered to bring the land surface and re-gional modeling communities together. A potentialstage for such a workshop may be the AfricanMonsoon Multidisciplinary Analysis (AMMA) Project.(See Page 8 for a description of AMMA.)

Other topics discussed included a briefing onurban modeling from Martin Best (UKMO), a brief-ing on the Integrated Land Ecosystem – AtmosphereProcesses Study and discussion of the status of theInternational Satellite Land Surface ClimatologyProject with Pavel Kabat (Alterra). There wasalso discussion of a proposed new initiative withinGMPP focusing on modeling of the diurnal cycle,brought up by Jan Polcher (CNRS/LMD). Theinitiative is aimed to improve communication andcollaboration across the three modeling studies inGMPP (land surface, cloud, and boundary layer).

NEW GLASS CHAIR

Dr. Paul Dirmeyer, Re-search Scientist at the Centerfor Ocean-Land-AtmosphereStudies (COLA), is the newchair of the Global Land-At-mosphere System Study(GLASS). Paul chaired andoversaw the successful comple-tion of the first Global SoilWetness Project (GSWP) andis now chairing the follow-on,

GSWP-2. He replaces Dr. Jan Polcher, Laboratoirede Météorologie Dynamique du CNRS, the founderand first chair of GLASS. Jan is chairman ofthe GEWEX Modeling and Prediction Panel(GMPP) and remains involved in GLASS activi-ties, including the data standardization effort andopen-source software bazaar he co-founded asan instrument to aid GLASS research, which hasbeen adopted and used by many land modelingprojects outside of GEWEX.

GABLS WORKSHOP ON STABLEBOUNDARY LAYERS

22–26 September 2003University of the Balearic Islands,

Mallorca, SpainAlbert A.M. Holtslag1, Robert J. Beare2, and

Joan Cuxart Rodamilans3

1Meteorology and Air Quality Section,Wageningen University, The Netherlands

2Met Office (UK),3University of the Balearic Islands, Mallorca (SP)

The GEWEX Atmospheric Boundary Layer Study(GABLS) aims to improve the understanding andthe representation of the atmospheric boundarylayer in regional and large-scale climate models.Recall that much of the warming predicted byclimate models is during stable conditions overland and ice (either in winter or at night).Consequently, the first focus of GABLS is onthe representation of the atmospheric StableBoundary Layer (SBL) (see Holtslag, 2003 formore background information).

On the basis of previous discussions and meet-ings, a benchmark case was selected to discussthe state of the art and to compare the skills ofsingle column (1D) models and Large-Eddy Simu-lation (LES) models. The case is based on theresults presented in a study by Kosovic and Curry(2000) for a shear-driven, stable case over ice. Assuch the boundary layer is driven by an imposed,uniform geostrophic wind, with a specified sur-face-cooling rate, which attains a quasi-steady stateSBL (after about 9 hours). The specifications ofthe case and the forcing conditions have beendistributed to the community in the previous year.Consequently, about 10 groups participated in thecomparison for the LES models and more than 10groups for the 1D models.

The findings were presented at a workshop.which was attended by about 30 scientists. Discus-sion sessions were held on the results for thedifferent models, and presentations were given onthe specifications of the different models, as wellas presentations on how to analyze the modelresults. Also, presentations were given on existingobservations, proposals for new data (includinglaboratory experiments) and on recent develop-ments in theory.

The purpose of the single-column intercomparisonwas to check the performance of any turbulence or

November 200312

vertical diffusion scheme for the selected case. Asan example, the figure above shows the outcomefor the potential temperature profile as representedby (most of) the 1D models after 9 hours ofcooling. Similar differences appear in the meanwind profiles, as well as the turbulent fluxes andother model outputs. Thus, with the same initialconditions and forcing conditions, the models indi-cate a large range of results. It appears that this isvery strongly related to the choice of the turbulentlength scale in the turbulence schemes, and not somuch to the vertical resolution. As expected, typi-cally the operational models allow for enhancedmixing resulting in deep boundary layers, whileresearch models, some of which have been ad-justed for the present case, show less mixing (morein agreement with data and LES). This issue will beexplored further.

The results for the different LES models aregiven in the figure in the next column. These re-sults are much closer to each other than shown inthe figure above for the 1D models. Note that mostof the LES models are given at grid length of 6.25

A comparison of the results for potential tempera-ture of the 1D models for the selected GABLS case.

meters. Sensitivity tests indicated that most of theremaining spread is attributable to differences informulation and configuration of the sub-grid model.However, even at very high resolution of 2 meters,some LES models did not converge, although thosethat were performed at a 1-meter resolution indi-cated a good degree of convergence. It seems thatif resolution increases, the SBL decreases. Thisindicates that more work on the representation ofthe subgrid scales in LES is still needed (in par-ticular, near the surface). Note that LES of SBLsonly began in the early 1990s and since then sig-nificant progress is made as indicated by the currentfindings.

At the workshop several options were exploredfor future activities. It was suggested that the stud-ies with LES and 1D should follow different routesfor SBLs. As such it is felt that 1D proposalsshould be compared to real data to see if theyfulfill the requirements of the climate models. Mean-while, LES must advance more carefully towardsstronger stratification, taking special considerationon the developments of the subgrid scale modeling.

A comparison of the results for potential tempera-ture of the LES models for the selected GABLS case.

Participants attending the GABLS Workshop.

13November 2003

The 1D models are ready now to comparewith more elaborate data, such as for cases withstronger cooling over different types of surfacesand increasing complexity. As such, it was dis-cussed to set up a new case over land dealingwith the full diurnal cycle. It was also suggestedto run the models with a simple surface energybudget (rather than the prescribed cooling) toallow for feedbacks between the land surfaceand the boundary layer. It should not be difficultto find suitable cases in the existing data sets(e.g., ARM, CASES-99, Cabauw, Lindenberg,SABLES98, AMERIFLUX, EUROFLUX). In ad-dition, suggestions were made to exploreobservations over the Baltic Sea and at Antarc-tica (notably Halley).

It is foreseen that the outcome of the MallorcaWorkshop will be presented in a number ofjournal papers, as well as a meeting connected tothe upcoming "Boundary Layers and TurbulenceConference," in Portland, Maine, USA, August2004. If you would like to participate in thecurrent 1D model intercomparison (open untilJanuary 1, 2004) please contact Joan CuxartRodamilans (joan.cuxart@uib.es). If you havesuggestions or are willing to prepare a bench-mark case on your data, please contact BertHoltslag (Bert.Holtslag@wur.nl). Please also consulthttp://www.gewex.org/ or http://www.met.wau.nl/for updates on GABLS activities in the nearfuture.

Acknowledgments

We wish to thank all participants at theMallorca Workshop and the involved modelinggroups for their inputs and comments. In par-ticular, we would like to acknowledge MalcolmMacVean, Anne McCabe, Maria Jimenez, andLaura Conangla for their help in the preparationof the selected case study and the involved han-dling of data and model results.

References

Holtslag, A. A. M., 2003. GABLS Initiates Intercomparisonfor Stable Boundary Layers. GEWEX News, 13 (May 2003),p.7-8.

Kosovic, B. and J. A. Curry, 2000. A Large Eddy Simula-tion Study of a Quasi-steady, Stable Stratified AtmosphericBoundary Layer. J. Atmos. Sci., 57, 1052-1068.

9TH ANNUAL MEETING OF GHP ANDFIVE FOCUSED WORKSHOPS

Lüneburg, GermanySeptember 22–26, 2003

John Roads, GHP ChairScripps ECPC, UCSD, La Jolla, California

The ninth annual meeting of the GEWEX Hy-drometeorology Panel (GHP) was hosted byHans-Jörg Isemer, GKSS Research Centre,Geesthacht, and Hartmut Grassl, Max-Planck-Insti-tute (MPI) for Meteorology, Hamburg. The meetingbegan with focused workshops on five GHP initia-tives: 1) Water and Energy Budget Studies (WEBS),chaired by J. Roads; 2) Water Resources Applica-tion Project (WRAP), chaired by L. Martz; 3)Sources and Cycling of Water (SCW), chaired byM. Bosilovich; 4) Extremes, chaired by R. Stewart;and 5) Predictability, chaired by J. Marengo.

A workshop on the Coordinated Enhanced Ob-serving Period (CEOP), chaired by S. Benedictwas also held during the meeting. From its earlybeginnings as a GHP working group, CEOP even-tually became a project of the World ClimateResearch Programme, with major contributions fromGEWEX. Of particular interest during the work-shop was the developing model data archive, amajor contribution from the Baltic Sea Experiment(BALTEX), which was described by H. Luther andM. Lautenschlager. S. Williams described theextensive in situ data archive, which is a majorcontribution of the GEWEX Americas PredictionProject (GAPP). J. Matsumoto described the Uni-versity of Tokyo remote sensing archive, a majorcontribution of the GEWEX Asian MonsoonExperiment (GAME). It should be noted that all ofthe Continental-Scale Experiments are contributorsto the CEOP in situ measurements, model output,and remote sensing products. A number of presen-tations were given describing the developing scienceinvestigations that will be using the extensive inter-national multi-sensor model CEOP data sets.

The official GHP meeting began after the fo-cused workshops. Summary presentations of progressand plans for the coming year were given by theGEWEX Continental-Scale Experiments (CSE) andaffiliated representatives, GEWEX and affiliatedglobal projects, and the GHP working groups.

CSE reports were presented for: the MackenzieGEWEX Experiment (MAGS) by K. Szeto; GAPPby R. Lawford; the Large Scale Biosphere-Atmo-sphere Experiment (LBA) by J. Marengo; BALTEX

November 200314

isotopes and working with GHP/SCW to under-stand water sources and sinks. The GHP wasespecially encouraged that IAEA will be sponsor-ing a workshop on isotope modeling in the nearfuture in order to accelerate progress.

Representatives of a few international regionalmodel intercomparison projects were also invited tomake presentations to the GHP. G. Takle describedthe US based Project to Intercompare RegionalClimate Situations (PIRCS) Project. J. Marengodescribed a beginning study over Brazil. B. Rockeldescribed a potential project focused on the GHPCSEs. H. Berbery described regional modeling ef-forts for LPB. Currently, these regional projectswork independently, but given that each of themhas substantial international involvement and is mak-ing substantial contributions to GEWEX/GHPtransferability goals, it may be of interest to de-velop a GHP Transferability Working Group. Thiswill be explored further during the coming year.

On the last day, summary presentations weremade by the chairs of the various GHP workinggroups on their plans for the coming year. GHP/WEBS and WRAP are now actively developingproject plans that will ultimately include GHP-widepapers; objectives and plans for these projects willbe reported on at the GEWEX Scientific SteeringGroup Meeting in January in Morocco. The SCW,Extremes, Predictability, and Transferability work-ing groups are just beginning and may have focusedworkshops next year along with the Data Manage-ment Working Group. The 10th annual GHP meeting(August or September 2004) will be hosted byLPB in Montevideo, Uruguay.

by H. J. Isemer; GAME by T. Yasunari; and theMurray-Darling Basin (MDB) Project by A. Seed.The GHP was especially interested in learning howthese CSEs are planning to evolve in the near- andlong-term. Some of the original CSEs now havesunset dates for their projects and new regionalinitiatives will have to be undertaken for them toremain viable in the face of the many new chal-lenges facing the scientific community.

C. Mechoso made a persuasive presentation onwhy the La Plata River Basin (LPB) experimentshould be considered for official status as a GEWEXCSE. The CSE representatives unanimously agreedthat the LPB would be able to eventually fully meetall of the established GHP criteria. The GHP will,therefore, be forwarding its recommendation to theGEWEX SSG that LPB be recognized as an offi-cial GEWEX CSE.

Presentations were also made from affiliatedGEWEX global projects, including: 1) Thomas Maurerdescribed the Global Runoff Data Center (GRDC),which will be working with GHP/WEBS; 2) B.Rudolf described the Global Precipitation Climatol-ogy Center (GPCC), which will also be workingwith GHP/WEBS. A. Hall described the Interna-tional Association of Hydrologic Sciences (IAHS)and its interest in working with GHP/WRAP on theIAHS Prediction of Ungauged Basins (PUB) initia-tive and in defining additional hydrologic referencesites for CEOP. D. Lettenmaier described aspectsof the Global Water Systems Project (GWSP),which may also be useful for links to GHP/WRAP.P. Aggarwal described the International AtomicEnergy Agency (IAEA) and its interest in using

Attendees at the 9th annual GHP meeting included: (front row, left to right) K. Masuda, B. Rudolf, A. Hall, J.Roads, G. Sommeria, S. Köppe, S. Sorooshian, A. Sugimoto, T. Yasunari, J. Matsumoto, P. Aggarwal, K. Szeto;(back row, left to right) L. Horta, L. Martz, G. Takle, J. Tomasella, B. Rockel, D. Lettenmaier, S. Williams, J.Marengo, R. Stewart, S. Benedict, A. Seed, D. Noone, R. Lawford, J. Bradd, H.-J. Isemer, T. Maurer, H. Berbery.

15November 2003

Pinker, R. T., I. Laszlo, C. H. Whitlock and T. P. Charlock, 1995.Radiative Flux Opens New Window on Climate Research. EOS,76, No. 15, April 11.

Pinker, R. T., J. D. Tarpley, I. Laszlo, K. E. Mitchell, et al., 2003.Surface Radiation Budgets in Support of the GEWEX ContinentalScale International Project (GCIP) and the GEWEX AmericasPrediction Project (GAPP), including the North American LandData Assimilation System (NLDAS) Project. J. Geophys. Res., inpress.

Platnick, S., M. D. King, S. A. Ackerman, W. P. Menzel, B. A.Baum, J. C. Riédi, and R. A. Frey, 2003. The MODIS CloudProducts: Algorithms and Examples from Terra. IEEE Trans.Geosci. Remote Sens., 41, 459-473.

Rossow, W. B. and R. A. Schiffer, 1991. ISCCP Cloud DataProducts. Bull. Amer. Meteor. Soc. 72, 2-20.

Rossow W. B. and R. A. Schiffer, 1999. Advances in Under-standing Clouds from ISCCP. Bull. Amer. Meteor. Soc., 80,2261-2287.

Schaaf, C. B., F. Gao, A. H. Strahler, W. Lucht, et al., 2002. FirstOperational BRDF, Albedo and Nadir Reflectance Products fromMODIS. Remote Sens. Environ., 83, 135-148.

Wang, H., Takashi Y. Nakajima, Akiko Higurashi, Teruyuki Nakajima,et al., 2002. A Multi-Satellite Approach to Estimate RadiativeFluxes over the Oceans. Spring AGU, Washington, DC, May 28–June 1.

Yu, H., R. E. Dickinson, M. Chin, Y. J. Kaufman, et al., 2003.Annual Cycle of Global Distributions of Aerosol Optical Depthfrom Integration of MODIS Retrievals and GOCART Model Simu-lations, 2003. J. Geophys. Res., 108(D3), 4128, doi: 10.1029/2002JD002717.

GEWEX/WCRP MEETINGS CALENDAR

For calendar updates, see the GEWEX web site:http://www.gewex.org

10–11 November 2003—8th GAME-ISP MEETING, KhonKaen, Thailand.

10–13 November 2003—14TH SESSION OF THEGEWEX RADIATION PANEL, Victoria, British Colum-bia, Canada.

10–14 November 2003—19TH SESSION OF THE CAS/JSC WORKING GROUP ON NUMERICAL EXPERI-MENTATION (WGNE)/7TH SESSION OF THE GEWEXMODELLING AND PREDICTION PANEL (GMPP), Sal-vador, Brazil.

11–14 November 2003—ACSYS FINAL SCIENCE CON-FERENCE AARI, St. Petersburg, Russia.

12–14 November 2003—MAGS ANNUAL MEETING#9, Montreal, Canada.

13–14 November 2003—WORKSHOP ON PROBLEMSWITH CLOUDS AND 3-D RADIATIVE TRANSFER,Victoria, British Columbia, Canada.

10–15 January 2004—84th ANNUAL AMERICAN ME-TEOROLOGICAL SOCIETY Meeting, Seattle, Washington.

26–30 January 2004—16TH SESSION OF THE GEWEXSSG, Marrakesh, Morocco.

2–4 Febuary 2004—WORKSHOP ON SEMI-ARID RE-GIONS, Marrakesh, Morocco.

1–6 March 2004—25TH SESSION OF THE WCRP JOINTSCIENTIFIC COMMITTEE, Moscow, Russian Federa-tion.

10–12 March 2004—THIRD CEOP IMPLEMENTA-TION PLANNING MEETING, Irvine, California, USA.

24–28 May 2004—4TH STUDY CONFERENCE ONBALTEX, Island of Bornholm, Denmark.

GEWEX NEWSPublished by the International GEWEX Project Office

Richard G. Lawford, Director

Mail: International GEWEX Project Office1010 Wayne Avenue, Suite 450Silver Spring, MD 20910, USA

Tel: (301) 565-8345Fax: (301) 565-8279

E-mail: gewex@gewex.orgWWW Site: http://www.gewex.org

Editor: Dawn P. ErlichLayout: Joseph L. Trowell

Piers Sellers, NASA astronaut and former GEWEX scientist,carried the GEWEX logo aboard the space shuttle Atlantisin October 2002. In the photograph above, Piers (left) ispresenting a GEWEX certificate recognizing the event toDavid Carson, Director of the World Climate ResearchProgramme, at a short ceremony that was held at theUniversity of Maryland, Baltimore Campus (UMBC) onAugust 4, 2003. Photograph courtesy of UMBC.

GEWEX FLIES ABOARD THE SHUTTLE

FIRST USE OF MODIS DATA...(Continued from Page 5)

November 200316

Over land, especially over Africa, estimates of fluxes from MODIS are higher than those from ISCCP D1,whereas over oceans, fluxes from ISCCP D1 tend to be larger than those from MODIS. This may beattributed to the cloud amount differences between these two data sets, as well as differences in cloudoptical depth, as illustrated above. See article on Page 4.

NEW 24-YEAR REGIONAL REANALYSIS NOW AVAILABLE(http://wwwt.emc.ncep.noaa.gov/mmb/rreanl/index.html)

FIRST MODIS AND GEWEX/SRB COMPARISONSHOWS REGIONAL DIFFERENCES

Top panels show fits to RAOBSof the NCEP North America Re-gional Reanalysis (NARR) winds,24-year average (1979–1992) forJuly, analyses (left) and firstguess fields (right) [dashed lines];and the NCEP/DOE global re-analysis (GR) [solid lines]. Thebottom left panel shows the NARR“observed” precipitation that isassimilated by the NARR systemover land and over southern por-tions of the oceans, monthly av-erage, July 1993 minus 1988.Bottom right panel: Same, butNARR generated precipitation. Seerelated article on Page 3.