EXECUTIVE SUMMARY -- GCIP MAP for 1997, 1998 and Outlook for 1999

EXECUTIVE SUMMARY

S1. Background

The World Climate Research Program in its Global Energy and Water Cycle Experiment (GEWEX) has established Continental Scale Experiments to improve scientific understanding and to model on a continental scale the coupling between the atmosphere and the land surface hydrologic processes for climate prediction purposes. The GEWEX Continental-scale International Project (GCIP) was established in the Mississippi River basin in 1992 to take advantage of the extensive meteorological and hydrological networks including the new Doppler radars, wind profilers, and automatic weather stations. GCIP is contributing to the long-term goal of demonstrating skill in predicting changes in water resources on time scales up to seasonal, annual, and interannual as an integral part of the climate prediction system. The overall strategy framework for implementing GCIP is shown in Figure S-1.

Figure S-1 Strategy Framework for Implementing GCIP.

The understanding and modeling of a continental scale watershed requires, from the outset, consideration of nonlinear-scale interactions in the aggregation of smaller processes to the larger scale and vice versa. GCIP research involves a systematic multiscale approach to accommodate physical process studies, model development, data assimilation, diagnostics,validation and data acquisition topics. GCIP research activities occur in a phased timetable and emphasize a particular region with special characteristics for a period of about two years. Four Large Scale Areas (LSAs) have been identified which encompass major river sub-basins of the Mississippi River basin and which, in aggregate, cover most of the GCIP domain, as shown in Figure S-2. The time phasing of activities within each of these areas is also shown in the figure. The GCIP Enhanced Observing Period started on 1 October 1995 and will continue for five years. Although the developmental activities are being initiated in limited regions; a fundamental thrust of the GCIP implementation strategy is that they lead toward an integrated continental-scale capability.

Figure S-2 The Mississippi River basin with boundaries defining the Large Scale Areas (LSAs) for GCIP Focused Studies (top). Temporal emphasis for each LSA from 1994 through 2000 (bottom).

S2. Coupled Hydrologic/Atmospheric Modeling

GCIP OBJECTIVE: Develop and evaluate coupled hydrologic/atmospheric models at resolutions appropriate to large-scale continental basins.

Model development in GCIP has two paths as shown in Figure S-1. A key strategy adopted early in GCIP was to fully exploit the high resolution limited area models that were being applied to regional weather prediction through various nesting procedures in the global models. This strategy was implemented as part of the "operational" path to provide the model assimilated and forecast data products for GCIP research as well as serving as a "proof of concept" for components of a coupled hydro-climate model. The "research" path focuses on the longer term activities needed for a coupled hydro-climate model.

Coupled Modeling Research Objective: Identify and understand the coupled processes that influence predictability at temporal time scales ranging from diurnal to seasonal and spatial scales relevant to water resources applications, and to develop a coupled model or models which can be validated (at these scales) using data from the Mississippi River basin.

S2.1 Near-Term Priorities for Coupled Modeling Research

In accordance with the overall goals of GCIP, the coupled modeling activities will focus on regional mesoscale modeling activities, to include the imbedding of regional models in global climate models. as an element in developing a capability to produce experimental seasonal-to-interannual climate predictions for the North American continent and evaluate these predictions relative to GCIP data. While recognizing that initially such experimental forecasts are likely only to have limited skill, GCIP will initiate an exploratory investigation of the potential value of such predictions in the context of water resource applications. This initiative will also serve as a mechanism through which to understand and develop the required interface between climate and weather predictions and their hydrological interpretation.

The focus of interest within GCIP in the next two to three years will be on continued development of improved representations of processes in coupled models with an emphasis on:

development of methods to improve values of parameters in these process representations;

transition improved process representations and parameters into coupled models, and,

exploratory runs and evaluation of their descriptive and predictive ability at time scales from diurnal to annual.

S2.2 Coupled Modeling Research: Long-term Items to be Initiated in the Next Two Years

To achieve the GCIP coupled modeling objective given at the beginning of Section 2, some long term initiatives need to begin in the next two years.These include:

(1) Definition and implementation of a measure of success for hydrologic predictions such as a "hydrologically relevant skill score".

(2) Characterization of hydrological storage in large-scale basins using tracer techniques using hydrograph separation techniques and geologic methods.

(3) Evaluation of relative contributions of land and oceanic influences to precipitation amounts in different seasons and in different regions of the North American continent with initial focus on the Mississippi River basin.

(4) Initiation of a regional ground water element in GCIP, focused on deep aquifier recharge and extraction.

S2.3 Improvements to Operational Coupled Mesoscale Models

The "operational" path (Figure S-1) provides the model assimilated and forecast output products for GCIP research, especially for energy and water budget studies. The regional mesoscale models also serve to test components of an imbedded regional climate model and can provide output for the evaluation of a coupled hydrologic/atmospheric model during the assimilation and early prediction time periods as a precursor to developing and testing a coupled hydrologic/atmospheric climate model. The output from the Eta, Mesoscale Analysis and Prediction System (MAPS), and Global Environmental Multiscale (GEM) regional mesoscale models is routinely compiled as part of the GCIP data set.

S2.3.1 Near-Term Priorities for Operational Coupled Mesoscale Models

(1) Use the GCIP special data sets to validate and evaluate the regional model output. Concentrate on validation of surface energy fluxes, surface skin temperature, soil moisture, cloud cover, precipitation, and diurnal planetary boundary layer profiles of temperature and humidity.

(2) Produce plots and graphs of the monthly Mississippi River Basin water budget components from the Eta, MAPS, and GEM model output. Compare with similar but independently computed budget components from observations.

S2.3.2 Operational Coupled Mesoscale Models: Long-term Items to start in the Next Two Years

(1) Validate and evaluate the 4DDA and forecast runoff of the Eta, MAPS, and GEM models (and later their companion land data assimilation systems), by applying streamflow/river routing algorithms to the gridded runoff archives from these systems.

(2) Investigate and develop algorithms for parameterizing sub-grid scale fractional precipitation distribution for use in the surface infiltration algorithms of coupled mesoscale models. Study the spatial and temporal distribution characteristics of the precipitation fields from the Eta, MAPS, and GEM model assimilation and forecast systems. Also, study the convective stability index products from these three systems.

(3) Investigate and develop strategies for a priori continental-scale estimation of key hydrological parameters, such as saturation hydraulic conductivity, soil moisture capacity ("bucket depth"), rooting depth, soil porosity, active soil column depth, and slope.

(4) Imbed coupled mesoscale models into global ocean/atmosphere models and investigate the advantages of imbedding (if any) on the skill and utility of seasonal and annual forecasts.

S3. Hydrological And Water Resources Modeling

GCIP Objective: Improve the utility of hydrologic predictions for water resources management up to seasonal and interannual time scales.

In the context of GCIP, one of the eventual aims of the modeling effort is to generate inputs for operational hydrological and water resources management models over a range of time scales up to interannual. The approach will be to link the hydrological and water resources research activities with the coupled modeling and data collection activities to produce more accurate streamflow forecasts, and in turn, to develop methods for utilizing those forecasts in water management decisions. The lead times to be emphasized will be longer than the currently accepted upper limit of weather forecasts (which is currently about one week), up to interannual.The near-term priorities are:

1) To develop procedures to allow GCIP hydrologic models to produce ensemble streamflow forecasts, using ensemble climate forecast model surface fields as forcing values. This will require, in particular, development of schemes to remove bias in both the climate model surface fields, and hydrologic model output; and,

2) to evaluate the worth of climate model ensemble forecasts for operation of one or more water resources systems.

In the longer term (e.g., beyond 2000) it is expected that the research activities will focus on water resources in the western U.S. The hydrologic processes of concern in the West (such as, e.g., snow accumulation and ablation in mountainous regions) are, in some respects, more amenable to improved hydrologic forecasting than are the water resource systems of the Mississippi River basin. Also, linkages between seasonal-to-interannual climate variations and tropical ocean processes (which currently appear to offer the best hope for accurate seasonal to annual forecasts) are generally stronger in the West than in the current GCIP region, so the West arguably offers a better water resources testbed for GCIP models than does its current region. In any event, GCIP will place a higher priority on the development of a demonstration application of seasonal forecast tools in at least one of the major water resources systems.

S4. Data Assimilation

GCIP objective: Develop and evaluate atmospheric, land, and coupled data assimilation schemes that incorporate both remote and in-situ observations.

The priority areas for research activities in data assimilation are:

Assessment of consistency of water and energy budgets via comparative evaluations of data among the three regional mesoscale models (see section S2.3.2 above) and with other data products and observations. This includes soil moisture and temperature observations, cloud and radiation products, and precipitation products. This understanding of consistency and errors in models (and observations) is necessary not only to improve the models themselves, but also to provide necessary information for data assimilation of new GCIP data sets as they become available.
Development and implementation of assimilation techniques. For the atmospheric models this needs to include the 3-D and 4-D Variational techniques. The emphasis in this development should be on remotely sensed data, cloud and moisture fields, and on attaining consistency with vertical motion fields so that analyzed cloud/moisture features are retained in subsequent model forecasts. For the macroscale land-surface/hydrology models, this should include use of both in situ and remote measurements of soil temperature and moisture, and of snow cover and depth. In order to do this data assimilation work, it will be necessary to have an improved understanding of the error characteristics of both these observations and the models themselves. This improved understanding can come from the assessment efforts advocated in the paragraph above.
Use surface models from regional models and perhaps from other organizations for the development of land data assimilation systems (LDASs). The LDASs should use observed data sets where possible, e.g., of precipitation and radiation, but also may use combined observation/model techniques to account for inadequacy of observations in certain areas. Again, knowledge of data errors is needed.

An additional future priority is the re-analysis of assimilated data sets using future improvement in data assimilation. A plan for such a regional reanalysis should be started in the immediate future.

S5. Diagnostic Studies

GCIP OBJECTIVE: Determine and explain the annual, interannual and spatial variability of the water and energy cycles within the Mississippi River basin.

The ultimate aim of the Diagnostic Studies research is to contribute to further improvements of seasonal to interannual climate predictions in support of water resource management. Diagnostics Studies also provide a basis for evaluation of the atmospheric, land, and coupled model data assimilation schemes as well as the forecasts produced from the prediction models. The near term priority is to describe the water budgets over the Mississippi River Basin and major GCIP-defined sub-basins through the use of observations in conjunction with model analyses. Specific activities over the period covered in this Major Activities Plan include investigation of the full four-dimensional water budgets based on observations and model assimilated data with particular emphasis on the output from the regional scale models producing the output for GCIP. Water budget components will be examined over the Continental Scale Area as well as the Large, Intermediate, and Small Scale Areas identified as focus study areas for GCIP. The effects of spatial and temporal sampling on the evaluation of the water budgets will be examined as well as the multi-year behavior of water balance components including storage.

Energy budgets pose a more complex problem since there are fewer direct measurements of the individual components of the energy budgets available for comparison and evaluation. The analyses are more dependent on model estimates of the energy budget terms in conjunction with observations from GCIP-related projects such at the International Satellite Cloud Climatology Project (ISCCP) and the International Satellite Land-Surface Climatology Project (ISLSCP).

One of the primary goals of Diagnostics Studies is to provide a fuller understanding of long-lasting hydrologic regimes associated with floods and droughts over the Mississippi Basin. Diagnostics Studies aimed at improved understanding of the initiation and maintenance of floods and droughts as well as conditions associated with their demise will be initiated.

S6. Critical Variables

A number of meteorologicalm hydrological and land surface variables are critical to the success of GCIP and were designated as Research Areas for special emphasis in the early stages of GCIP. The priority research activities for each are summarized in this section.

S6.1 Precipitation

Precipitation Objective: Achieve a better understanding and estimation of the space-time structure of precipitation over the Mississippi River basin, including improvements in atmospheric model representation of precipitation to support improved coupled modeling.

S6.1.1 Precipitation Research Activities

The near-term priority areas for research in precipitation include:

Understand the basic physical reasons/processes behind anomalous precipitation at all scales of interest (daily, seasonal, interannual). Of special interest are the large anomalies which cause significant societal impacts. Topics of particular importance include: (a) understanding of land surface influences on precipitation (e.g., orography, soil moisture anomalies, spatial distribution of snow cover etc.), and (b) interaction of dynamics and cloud microphysics.

Develop innovative approaches and methods for validating rainfall predictions from coupled models at a range of space-time scales, including methods for validating ensemble predictions.. Of special interest are: (a) quantification of how well models capture important physical and statistical features that we have evidence for from observations. This sort of validation can provide guidance for improved cloud process parameterization in coupled modeling; and (b) recognizing the inherent limits of predictability of precipitation.

Study the sensitivity of predictions to initial and boundary conditions in a nested modeling environment and determine the scales at which it is ``better'' to use a nested approach.

Improve precipitation measurements with special emphasis on: (a) spatial and temporal distribution of solid precipitation including measurement corrections, (b) improved use of WSR-88D data in precipitation measurement including snowfall water equivalent, and (c) development and testing of methods for combining remote measurements (airborne gamma, satellite and radar) with ground measurements to provide enhanced gridded snow water equivalent fields.

The longer term precipitation research activities which should be initiated in the next two years include:

Develop active and direct collaboration between the Coupled Modeling and the Precipitation research activities to address precipitation validation issues.
Initiate research on developing a fully distributed energy balance snow model. Such a model is required to assimilate observed and modelled data sets in order to produce gridded snow water equivalent and snow cover fields. It is also needed to establish initial and boundary conditions for seasonal and interannual hydrologic forecasts.
Initiate and promote efforts to secure spatial and temporal homogeneity of in-situ precipitation (especially solid precipitation), wind and cloud cover measurements that are used for model validation, data assimilation and/or water and energy budget studies.

S6.1.2 Precipitation Measurement and Analysis

GCIP requires the best available precipitation products and recognizes the potential value of the WSR-88D radars in meeting this requirement. It is a goal of GCIP to contribute to the development of a derived product which combines WSR-88D, gauge, and satellite estimates of precipitation resulting in a product with a 4-km spatial and hourly temporal resolution. Such a goal is not expected to be achieved for a routine product until much later in the five-year Enhanced Observing Period since it is dependent upon some of the modernization improvements yet to be implemented by the NWS. GCIP has an ongoing effort to provide precipitation data products for GCIP investigators. A precipitation analysis is being produced routinely by the NOAA/NCEP and archived at UCAR. A composite of precipitation observations from all available observing networks is produced and archived as part of the GCIP data set in the in-situ data source module.

Associated with the measurement of precipitation caught by the gauge is the question of representative exposure of the gauge and the effect of not having wind shields or the characteristics of different shields on gauge catch, evaporation, etc. The systematic adjustment of gauge errors is a necessary requirement for the development of good-quality precipitation fields. GCIP is also supporting research activities to determine systematic errors in precipitation measurements and to derive adjusted values for in-situ solid precipitation measurements starting in the Upper Mississippi River basin. The results from the snowfall measurement corrections applied to the Upper Mississippi River basin will be used in other regions of the Mississippi River basin to compile corrected snowfall measurements, and thus compile reasonably accurate in-situ precipitation measurements over the full annual hydrologic cycle.

S6.2 Soil Moisture

Soil Moisture Objective: Improve understanding and estimation of the space-time structure of soil moisture, the relationship between model estimates of soil moisture and observations of soil moisture, and to produce soil moisture fields for the GCIP area to be used as diagnostic and input data for modeling.

The near-term priority areas for soil moisture research activities include:

Develop a daily soil moisture analysis at a scale of about 40 km for four depths for the Arkansas-Red River basin because this is the area where the most in-situ measurements are available and the region where current remote sensing can provide the best information because of the relatively dense vegetation cover.

Develop some initial cold season analysis products for soil moisture and soil temperature fields, including frozen and unfrozen soils.

Develop an understanding of the relationship between point measurements and areal representations of soil moisture as well as the relationship between surface measurements and the vertical profile of soil moisture.

Long term activities that should be started within the next two or three years include:

Develop approach and methods to produce a daily soil moisture analysis product for the Mississippi River basin at a scale of 40 km for four depths. Such an assimilated product must be produced from a variety of data sources, including output from hydrologic models driven by measured meteorological data, in situ soil moisture observations, aircraft and satellite remotely sensed data.
Increase the number of in situ monitoring stations to include additional soil/climate regimes. The North-South transect along 96W longitude represents a good start.
Develop research plan for using advanced satellite remote sensing capabilities and anticipate the infusion of Advanced Microwave Scanning Radiometer (AMSR) around 1999, and hopefully, L-Band satellite sensor data after the year 2000.

S6.3 Land Surface Characteristics

Land Surface Objective: Improve the quantitative understanding of the relationships between model parameterizations of land surface processes and land surface characteristics, while also facilitating the development, availability, evaluation, and validation of multiresolution land surface data sets required for land surface process research in GCIP.

S6.3.1 Near-Term Priorities for Land Surface Characteristics

Facilitate the test and evaluation of newly developed land surface characteristics data sets in GCIP's land-atmosphere coupled modeling research at the large river basin to continental scales.

Develop multiresolution, aggregated land cover characteristics data sets that are internally consistent and collectively "harmonized" within model spatial domain grid cells. Products will include, at 10 km and 30 km grid cell sizes, the three predominant land cover/land use classes in each grid cell and a climatology of the fractional presence of green vegetation associated with the three predominant land use/land cover classes. The product for the conterminous United States will be based on a five-year climatology, while the prototype global product will be derived from an 18-month time series.

Conduct land cover characterization research involving multiresolution aggregation, scaling, and validation studies based on 30 m land cover data sets developed from Landsat TM imagery as part of a USGS- and EPA-led activity. Preliminary 30 m land cover data are scheduled for completion in the Eastern U.S. by late 1997 with the completion of the remainder of the conterminous U.S. planned by the end of 1999.

S6.3.2 Land Surface Characteristics Long-Term Items to be Started in the Next Two Years

Conduct GIS-based land surface characterization research studies within the Mississippi River Basin to determine multiresolution interrelationships among land cover, biophysical parameters, soils, and topographic data sets including derived parameter fields. In addition to verification of physically-consistent associations, these studies should also focus on the role of landscape heterogeneity in parameterizing land surface processes.
Conduct advanced research on satellite data processing and physically-based remote sensing algorithms. Advanced satellite-derived land surface temperature/emissivity algorithms are needed to study land surface fluxes.
Develop plans to test and evaluate remote sensing data sets that will become available following the launches of the NASA-led Earth Observing System (EOS) AM1 Platform and Landsat 7 during the mid-1998 time-frames. The advanced remote sensing science algorithms under development by the MODIS Land Science Team are of particular relevance to GCIP. In addition, current NASA plans call for near-synchronous orbits of the EOS AM1 and Landsat 7 satellite systems, thereby creating a substantial potential for the complementary operation of coarse-and high-resolution satellite data of interest to some GCIP researchers.

S6.4 Clouds and Radiation

Clouds and Radiation Objective: Improve the description and understanding of the radiative fluxes that drive land-atmosphere interactions and their parameterization in predictive models, while also facilitating the development, availability, evaluation, and validation of multiresolution clouds and radiation data sets required for process studies and coupled modeling research in GCIP.

As new and improved satellite products for GCIP are developed and brought into production, it is necessary to validate and tune the algorithms to provide the most consistently accurate quantities. This requires operating a parallel system that produces the satellite products off line using the same data and the same algorithms, so that the algorithms can be modified and tuned, and the results compared with ground truth. There are current problems with the retrieval of cloud cover and insolation over a snow covered surface that must be addressed through tuning with a parallel system.

Radiation budget components, cloud amounts and heights, and surface temperatures from the regional scale Numerical Weather Prediction models must be compared with satellite observations of the same quantities. Radiation and cloud output from the Eta model will be collected from selected forecast times and remapped into the resolution and map projection of the GOES satellite products and provided for comparison studies. The degree of agreement, conditions under which the model output and the observations are quite different (season, snow cover, bare soil, etc.), and the degree to which the diurnal cycle in observed variables are replicated by model output are both needing evaluation.

The cloud and radiation components in the Eta and other regional models need improvement and the research to upgrade them needs to be started in the next two years if GCIP is to benefit from the research results. Such topics as the interaction of cloud and radiation fields and surface variability within a grid box, use of better cloud parameterization, and cloud resolving models are all appropriate for research. The specific area of research may be dictated by the results of the comparison of model output with observations.

S6.5 Streamflow/Runoff

Streamflow data and runoff estimates are required both for the development and for the testing and verification of coupled atmospheric/hydrological models. Streamflow is determined from measurements of stream stage at a stream-gauging station. It is essential that the gauge data used for testing and verification of models be essentially unaffected by upstream regulation or diversion. Runoff is the spatially distributed supply of water to the stream network which cannot be measured directly. Both surface and sub-surface components are part of runoff.

Near-Term Priorities for Streamflow/Runoff:

Extend the available historical data base for unregulated basins at the intermediate and small scales (10 to 1000 km²) in the Arkansas-Red River basin by updating from 1988 the active streamflow to develop and demonstrate regionalization methods for the estimation of hydrologic model parameters. In addition to allowing the estimation of the land-surface model parameters these data are needed for the development of runoff routing parameters and gridding runoff.

Develop naturalized streamflow records at key locations in the Arkansas-Red River basin up to the current time to enable the validation of the atmospheric model predictions. Key locations would include the Red River at Shreveport and the Arkansas River at Little Rock, being the largest basins which can be feasibly considered.

Test a method for estimating gridded runoff data for the Arkansas-Red River basin to enable the direct validation of atmospheric model runoff predictions.

S7. Data Collection and Management

GCIP OBJECTIVE: Provide access to comprehensive in-situ, remote sensing and model output data sets for use in GCIP research and as a benchmark for future studies.

As noted in Figure S-2, the GCIP Enhanced Observing Period started on 1 October 1995 and will continue for five years. The data collected during each year will be compiled into a number of standard and custom data sets. The data collection periods for the GCIP standard data sets are shown in Figure S-3. These data sets will be published on CD-ROMs for distribution, especially to international scientists interested in GCIP. Increasingly, the national GCIP investigators are making use of the on-line GCIP data services available through the World Wide Web at the URL address: http://www.ogp.noaa.gov/gcip/

Figure S-3 Compiled and Planned Standard Data Sets for GCIP Research.

S7.1 Data Sets for Warm Periods

The initial focus of GCIP on the warm season processes in the annual hydrological cycle has produced data sets for three different periods in the LSA-SW(see Figure S-4). The data collected during the Enhanced Seasonal Observing Period in 1996 (ESOP-96) is scheduled to be compiled into a standard data set by December 1997. The types of data which comprise the ESOP-96 are described in the Tactical Data Collection and Management Plan for the 1996 Enhanced Seasonal Observing Period (ESOP-96).

Figure S-4 The LSA-SW Encompasses the Arkansas-Red river basin. GCIP Focus Study Areas in the LSA-SW Include the CART/ARM Site Operated by the Department of Energy and the Little Washita Watershed Operated by the USDA/Agriculture Research Service.

S7.2 Data Sets for Cold Periods

The data collection activities for Water Years(WY) 1997 and 1998 include the cold season in the Upper Mississippi River basin identified as the LSA-NC in Figure S-2. The details of the data to be collected during this period are given in the Tactical Data Collection and Management Plan for the 1997 Enhanced Seasonal Observing Period (ESOP-97).

S7.3 Data Sets for the Annual Hydrologic Cycle

The data collection for the next two years covering the full annual cycle will concentrate on the data needed for energy and water budget studies with some increasing emphasis on coupled modeling validation and evaluation. In this regard a Near Surface Observation (NESOB) Data set for at least one 12-month period beginning 1 April 1997 is being compiled. This special dataset is intended to fulfill the data requirements for:

Land surface process studies

Validation and verification of land surface processing schemes

Detailed validation and verification of model output from regional land-atmosphere coupled models.

Derivation of surface energy and water budgets.

This integrated dataset is being compiled for the LSA-SW which includes the ARM/CART site, the Little Washita Watershed and the Oklahoma Mesonet (if available) in the Arkansas-Red River basin. The vertical dimension includes from 3000m above the surface to 2m below the surface. The preparation of the archive data by the U.S. Geological Survey is done on a Water Year basis. The streamflow data for the Water Year are archived the following April and May. This will necessitate the compilation of the one-year Near Surface Observation Dataset in two parts. The period from 1 April through 30 September 1997 can be completed by June 1998 and the last six months of the one year dataset will be completed by June 1999.

The data sets for the whole of the Mississippi River basin, as shown in Figure S-3, are planned to be compiled beginning in 1999.