PROCEEDINGS OF THE TENTH U.S.-JAPAN WORKSHOP ON GLOBAL CHANGE: CLIMATE AND WATER

Foreword

JOINT REPORT

The 10^th U.S.-Japan Workshop on Global Change was designed to bring Japanese and American scientists together to exchange views on research priorities and needs related to the effects of the water cycle on climate and the effects of climate variability and change on water resources. The workshop had two main sessions: one dealing with the key scientific issues related to the understanding and prediction of water cycle variability and the application of this knowledge to water management, and a second session to consider the capabilities of observational systems, models, and decision support resources for addressing these research priorities.

Keynote presentations summarizing the major issues were given in advance of each of the work sessions. The keynote speakers reviewing the scientific issues included Professor Dennis Lettenmaier, the GEWEX representative to the Global Water System Project, who reviewed long-term climate and water cycle variability and change; Professor Masahide Kimoto who gave an overview of the physical basis for making predictions of climate and water cycle variability on monthly to annual time scales, and Professor Soroosh Sorooshian, the Chair of the GEWEX Scientific Steering Group, who described the prospects and challenges for using the results of hydroclimatic research in water resources management. The second set of introductory lecturers included Professor Kenji Nakamura, senior NASDA scientist for the GPM, who reviewed the Global Precipitation Measurement (GPM) Mission, a major observational initiative of Japan and the USA; Professor Roni Avissar, chair of the U.S. Global Water Cycle Scientific Steering Group, who described the characteristics of a new model for predicting water cycle variability; and Professor Kuniyoshi Tackeuchi, President of IAHS, who reviewed international activities related to global water resource assessments and opportunities for U.S.-Japan cooperation in this area.

The first set of working group sessions addressed research needs related to climate variability and change, predictions of water cycle variability, and applications of climate information in water resources. The second set of working groups on Thursday afternoon and Friday morning addressed the roles of observations, modeling capabilities and assessments in relation to the priorities raised in the first working group sessions. Reports from the individual sessions are summarized in the following pages.

The last afternoon of the workshop was devoted to a plenary discussion to identify ways in which the priority issues could be addressed through U.S.-Japan collaboration. Based on these discussions the plenary session recommended that:

Participants: Toshiya Aramaki, University of Tokyo

Mike Bosilovich, Data Assimilation Office, NASA

Seita Emori, FRSGC

Efi Foufoula-Georgiou, University of Minnesota

Haruyasu Nagai, Japan Atomic Energy Research Institute

Tadahiro Hayasaka, Research Institute for Humanity and Nature

Akihiko Ito, FRSGC

Rick Lawford, GAPP Program Manager

Dennis Lettenmaier, University of Washington

Eric Wood, Princeton University

Using the reviews in this subject area given (with focus on meteorological variables) by the 2001 IPCC report [see http://www.grida.no/climate/ipcc_tar/wg1/338.htm] and (with greater focus on hydrological variables) by Professor Lettenmaier at this workshop as the basis for discussion, WG1 assessed what is known and what is unknown on this topic and identified the following gaps in understanding that merit the joint attention of U.S. and Japanese researchers.

Research Gaps

After detailed discussion of these research gaps, the following recommendations were made.

Co-chairs: Akio Kitoh and Jagadish Shukla

Participants: Roni Avissar, Richard Carbone, Jay Famiglietti, Masahide Kimoto, Akio Kitoh, Toshio Koike, Koichiro Kuraji, Norman Miller, Kenji Nakamura, John Roads, Takehiko Satomura, John Schaake, Zubin Zeng,

There are several ongoing international activities involving long-range (monthly to annual) predictions of the water cycle in which scientists from the United States and Japan play major roles. These include the Global Energy and Water-cycle Experiment (GEWEX), the new World Climate Research Project (WCRP) Coordinated Enhanced Observing Period (CEOP), and World Weather Research Program (WWRP), etc.

CEOP is a coordinated international experiment initiated by GEWEX and coordinated by

WCRP to develop a pilot global hydrometeorological data set, which can be used to enhance understanding of global water and energy budgets as well as monsoon characteristics, which should lead to increased seasonal predictions. The CEOP data set will include satellite measurements at 41 international reference sites, which also have detailed in situ measurements (almost all long-term global tower sites are included). These observations will be augmented by detailed Model Output Location Time Series (MOLTS) at each of the reference sites, as well as more extensive global gridded output. These global measurements will be augmented by regional output, pertinent to the GEWEX CSEs.

WMO Commission on Atmospheric Sciences. Some of its activities are conducted jointly with the WGNE (see below). Ten active or developing projects are in progress ranging from non-dynamical nowcasting of weather to extended range and intra-seasonal predictions of precipitation. Two projects are of especially high relevance to U.S.-Japan collaboration: THORPEX and Warm Season QPF/flooding Initiative.

THORPEX (joint with WGNE) is a global atmospheric research program, much in the spirit of GARP, that seeks to explore the limits of predictability in relation to initial and boundary conditions analyzed from an adaptive global observing system of the future; improved model representations of physical processes; and advanced data assimilation techniques. Essentially all of the major NWP centers, including JMA, NCEP and ECMWF are involved. Current agreements call for global high-resolution “nature runs -- on which to base subsequent predictability studies. Experimentally, Asia and the western N. Pacific will be a regional focus of THORPEX to explore the impact of adaptive satellite and in situ observing techniques.

The Warm Season QPF/Flood initiative is at a formative stage. Dynamical Climatology studies are in progress in North America (Carbone and Arkin); Australia (Keenan); China (Wang and Chen); Africa and Europe (Brozkova and Levizzani). Carbone is the lead U.S. scientist and Chair of the international working group. Japan's Baiu/Mei-Yu heavy rainfall is central to this study and Japanese-American collaboration is highly desired for the region in the lee of the Tibetan Plateau in the Mei-Yu/Baiu season. Convective System-Resolving Model (CSRM) simulations, regional in scale and seasonal in duration are needed over East Asia and North America.

The Working Group on Numerical Experimentation (WGNE) is an established group using numerical models on a wide variety of time scales. WGNE projects also include the

There are a number of research areas that need improvements for quantifying seasonal predictions. Determining land surface initial conditions and teleconnected global processes, linking atmospheric, oceanic, and land-surface processes with lagged responses, and how their interrelationships modify seasonal variations may be viewed as general needs for improvement.

Summertime continental precipitation needs improvements in observations and models. Precipitation in cloud-resolving models needs to be further evaluated against observations. Observations that include satellite-derived data (e.g. GPM, TRMM) and surface gage stations are required as part of a more comprehensive observing network that will provide a more comprehensive model validation. Researching regions affected by orographically forced precipitation (e.g. Tibetan Plateau, Rocky Mountains, Sierra Nevada Mountains) is an area of joint Japan-U.S. collaboration.

Land-surface soil moisture evolution and its impacts on atmospheric flux exchange are poorly predicted due to lack of sufficient observations. New satellite-derived data from the AMSR has approximately a 50-70km footprint, is unable to penetrate soil depths beyond a skin layer, and cannot resolve soil moisture under canopies. The GRACE measurement satellite is able to measure large area total deep-water fluctuations, but cannot provide any fine horizontal or vertical information. Field measurements are mostly limited to small-scale study areas and are difficult to upscale and apply to numerical models for predicting soil moisture evolution. Data assimilation of the above satellite-derived information combined with sparse observations and new spatio-temporal aggregation schemes will allow for improvements in land-surface initial conditions and the predictability of soil moisture, snow water equivalent, and surface fluxes. New algorithms that combine this information can be developed to generate time-evolving soil moisture profiles with depth.

There is a need to build upon existing intensive observing programs by designing a capability to understand the ocean and atmosphere low-to-high frequency teleconnected systems that influence precipitation, surface processes, and their feedbacks. Coordinate joint U.S.-Japan research, along with other international programs, will advance our understanding and provide a framework for the design of next generation seasonal predictions that capture feedback mechanisms and new observations.

In order to define initial conditions and land surface states for seasonal and longer time scales, a description of vegetation, soil wetness, and snow coverage is required. Further, assimilation of precipitation may enhance predictability. Current U. S. and Japanese sensors will have the capability to provide these data either for initialization or routine assimilation.

Visible and near-infrared sensors (MODIS, GLI, AVHRR) can provide LAI, snow areal extent and surface temperature. Infrared sounders (AIRS, TOVS) can provide water vapor and temperature profiles in the atmosphere over land and ocean under cloud-free conditions. Microwave sounders (AMSU-A, AMSU-B) can monitor these profiles only over the ocean, but for all weather conditions. Microwave radiometers (AMSR, AMSR-E, TMI, SSM/I) can provide surface moisture content, snow water equivalent, vegetation water content, and precipitation. Precipitation radar (PR) can provide the profile of precipitation. GRACE will provide estimates of water storage changes on land.

Limitations in the representativeness and the spatial-temporal coverage of the available products (for example, in soil moisture profiles, estimation of vegetation and snow characteristics) and physical consistency with numerical models, require continued collaboration between U. S and Japanese scientists on assimilation schemes, in addition to retrieval algorithms, in order to fully exploit the potential of remotely sensed data for improved predictability. Optimal combination of in situ data with satellite observations also an important component in these efforts.

In long-range prediction (monthly to seasonal), boundary conditions and model physics assume greater importance than in short range forecasts. Among the model physics the representation of clouds is of primary of importance. Convection parameterizations are known to be deficient in the representation of convection in forecast models at all ranges of prediction. Such parameterizations fail to account for the capacity of convection to dynamically organize up to 1000 times the scale of its original elements. The mesoscale dynamics is influenced strongly by physical processes such as: aerosol, cloud, and precipitation microphysics, heating and moistening of the free troposphere, diabatic cooling and the representation of downdrafts, land-surface initiation or triggering of free convection, moisture flux convergence driven by elevated heat sources and gravity-inertia oscillations, heterogeneous environments created by height and areas of clouds associated with convection, and vertical redistribution of horizontal momentum.

It is recognized that there is a stochastic component of convective lifecycles at the very short range, governed by highly non-linear dynamics, and not specifically predictable beyond two or three convective overturning times. The concept of equilibrium on larger scales of prediction applies to regional-seasonal periods such as those within or influenced by monsoonal circulations as observed in Asia and North America. It is also recognized that episodes of convection over tropical-oceanic and many continental regions exhibit sustained organization and coherency on nearly continental scales and may be amenable to skillful intra-seasonal and seasonal predictions.

Such skillful predictions rely on appropriate parameterizations of convection, representation of mesoscale convective systems and the coherent regeneration of these. Skillful predictions also rely heavily on the dependence on larger scale forcing and boundary forcing that must be represented adequately.

Improved fluxes of various physical variables are also needed. The representation of fluxes in stable PBLs must be improved. The moist PBL is also very difficult to represent along with other processes including, cloud topped PBL, cloud development and dissipation related to entrainment from the free troposphere. In sloping and complex terrain special considerations come into play with respect to mixing, and mechanical turbulence generation processes. Observations from GAME, CASES, and extensions of these should guide improved knowledge of model physics and contribute to the development of adequate parameterizations. Also, the same observational data may be used to compare performance of various models in the United States and Japan. This will help employ mesoscale and microscale models in climate research.

A hierarchal approach to the use of models is required, from CSRMs to GCMs. This includes the development of methodologies for super-parameterization, adaptive grids and one way and two way nesting. A definitive continental/seasonal-scale CSRM experiment should be conducted for conditions representative of around Tibetan Plateau in rainy season and for North America in mid-summer (July). This experiment would define the appropriate path toward super-parameterization in global climate models of the future. In this regard, validation observations in support of the simulations can make effective use of systematic observations over the United States, Tibetan Plateau, and GAME areas (Siberia to the tropics). Examples of systematic observations include TRMM-GPM, geostationary satellites, DOE ARM and US WSR-88D radar network. The only computational facility capable of CSRM-regional-seasonal simulations is Japan's Earth Simulator and thus could form the basis for a U.S.-Japan collaboration of great significance to prediction of precipitation within the warm season water cycle.

Hydrologic extremes can be defined on various time scales. At the seasonal time scales, extremes include highly abnormal wet regimes and highly abnormal dry regimes, which are manifested in characteristics of the land surface, especially in the case of drought. There is potential skill in developing these seasonal anomalies, especially if the land surface feedbacks are included in seasonal prediction models. At shorter time scales, abnormal wet episodes result in flood events, which cannot be predicted in detail (timing, duration, magnitude) at seasonal time scales. Changes in the seasonal mean anomaly may be related to changes in the distribution functions, which should allow at least indication of potential extremes. For example ENSO predictions have been used to alert policy makers for potential flood events.

The question of the influence of human activities such as land cover/land use change, irrigation, groundwater withdrawal and reservoirs affect on regional climate variability and seasonal predictability is of great social relevance.

However, this problem has not been addressed in the systematic way. In fact this is because of adequate data. Global land cover data are available from AVHRR and MODIS. Land use data are available over individual regions and may be available globally in the near future. Irrigation data may be available based on various field surveys but is not available to global modelers yet. Groundwater withdrawal has been estimated over some region of the world. Geographic locations of reservoirs are available but the additional information on each reservoir is still unavailable. At present, only land cover data and vegetation data (such as fractional vegetation cover and leaf area index) are used in global and regional models.

If irrigation and land use data are available, they can be directly used by current land surface models. However, land surface models need to be further improved in order to make use of groundwater withdrawal and reservoir data. Based on this analysis, for the near term, we should focus on developing the global data sets of irrigation and land use, and incorporate them into land surface models. For the long term, a comprehensive hydro-environmental model needs to be developed that can realistically treat groundwater withdrawal and reservoir as part of the overall water movements over lands.

Co-Chairs: Ruby Leung, Kaoru Takara

Participants: Allen Bradley, Shannon Cunniff, Konstantine Georgakakos, Harvey Hill, Hiroyuki Kawashima, Takao Masumoto, Dave Mathews, Ryohji Ohba, Taikan Oki, Michael Sale, Soroosh Sorooshian, Kuniyoshi Takeuchi,

During the first day, WG3 started the discussion with brief introductions by the co-chairs. The participants then introduced themselves and their research interests. Examples of work being done by the working group members include research on the effects of climate change on water resources, hydrologic and land surface modeling, application of remote sensing to hydrology, regional climate modeling, ecology and biogeography, economic development and water resources issues in northern China, flow modeling in low lying areas, study of water issues of the Mekong River, integrating climate information in resource decision making, assessing impacts of climate variability and change on multiuse allocation and reallocation in complex reservoir systems, and effects of social changes on balance in water supply and demand. It was noted that Japan's interests are more closely related to the effects of global warming on water resources while U.S. scientists are studying both the use of seasonal forecasts and climate change information in water management. Members then collectively reviewed the state of knowledge in the field by listing different types of climate information, water applications, and end users that are relevant for the discussion. Each member was then asked to develop a few water application priorities that are of mutual interests with high impacts and leverage on existing programs. Our goal was to develop a few recommendations for collaborative work that can be moved forward.

During the second day, each member of WG3 provided a short list of water application priorities for discussion. These include:

Based on the above list, a set of focused research was developed for potential U.S.-Japan collaboration. This includes:

Actions:

We propose a workshop to be attended by members of WG3 and potential collaborators to follow up on the two recommendations. For the first recommendation, basins or regions for the demonstration project(s) will be selected after reviewing recommendations from other working groups and on-going programs. Workshop participants will initiate planning of the demonstration project(s) and determine research focus and data requirement that are specific to the basins or regions chosen. For the second recommendation, workshop participants will aim at refining the objectives and develop outlines for projects that can address the issues and define achievable outcomes.

Co-Chairs: Eric Wood and Toshio Koike

Participants: Roni Avissar, Efi Foufoula-Georgiou, Konstantine Georgakakos, Tadahiro Hayasaka, Koichiro Kuraji, Rick Lawford, Kenji Nakamura, Tetsuo Ohata, Takehiko Satomura, John Schaake, Jagidish Shukla

1) For Japan and U.S. to provide international leadership in global water cycle observations in support of water cycle research through activities that include joint programs in water cycle observations (validation sites and analysis) and the establishment of data integration centers.

This recommendation builds upon recent international agreements (WSSD Para. 27: IGOS water cycle theme; submitted UNFCCC/SBSTA/COP agreements.)

Specifically, the working group recommends:

Action Item: White Paper on reference sites. (Schaake, Lettenmaier, Kuraji, Nakamura, Ohata)

2) Japan and the U.S. jointly organize a major scientific workshop organized around the

“Application of observing systems to document the global water cycle. --

Action Item: Write a short proposal for such a workshop and present it to NASA and NASDA (and other interested US and Japanese agencies) for their support and organization.

Possible date for the workshop: summer 2004.

(Japan: Nakamura, Koike/U.S.: Wood, Shuttleworth, Eric Smith)

The workshop would be organized around major science questions and programmatic issues centered on the water cycle. For example:

Format:

Participants: Mike Bosilovich, NASA (data assimilation)

Allen Bradley, University of Iowa- hydrology, quality of information going into models Akihito Jay Famiglietti, University of California at Irvine- soil moisture and soil water remote sensing

Ito, terrestrial

Masahide Kimoto, GCM development

Akio Kitoh, Meteorological Research Institute- Monsoon, and ENSO

Dennis Lettenmaier, University of Washington- VIC and land surface modeling analysis

Ruby Leung, PNNL- coupling land atmosphere models mesoscale models

Norman Miller, University of Arizona -land surface groundwater

Taikan Oki, Research institute for humanity

Soroosh Sorooshian, GEWEX- hydrologist

Xubin Zeng, University of Arizona- interface process and atmospheric boundary layer

Major modeling Groups

Several associated university groups are working with and somewhat independently of these major modeling centers

Working group 5 recognizes several ongoing international activities involving long-range (monthly to annual) predictions of the water cycle in which the United States and Japan play major roles.

a. Global Precipitation Climatology Project (GPCP), which includes TRMM, a joint Japan-U.S. effort to develop global precipitation products on daily to monthly time scales.

b. Global Hydrometeorology Project (GHP), which includes substantial participation by U.S. GAPP and GAME continental scale experiments. A major goal of GHP is to develop integrated models of the water cycle.

c. Global Soil Wetness Project (GWSP): A project to develop the best estimate for global soil moisture.

d. GEWEX Cloud Systems Study (GCSS): A project to develop cloud-resolving models that will become the basis for the next generation cloud parameterizations.

e. International Satellite Comparison Project (ISSCP)

f. International Satellite Land Surface Comparison Project (ISLSCP)

CEOP is a coordinated international experiment initiated by GEWEX and coordinated by WCRP to develop a pilot global hydrometeorological data set, which can be used to enhance understanding of global water and energy budgets as well as monsoon characteristics, which should lead to increased seasonal predictions. The CEOP data set will include satellite measurements at 41 international reference sites, which also have detailed in situ measurements (almost all long term global tower sites are included). These observations will be augmented by detailed Model Output Location Time Series (MOLTS) at each of the reference sites, as well as more extensive global-gridded output. These global measurements will be augmented by regional output, pertinent to the GEWEX CSEs.

WGNE (Working Group on Numerical Experimentation)

IAEA International Atomic Energy Association

WCRP CLIC/ACSYS cold season processes

IARC, International Arctic Research Center (in Alaska)

Background

An overarching issue raised in many of the working groups is that deficiencies leading to biases in model data can significantly impact the usefulness of the data in both scientific investigation of climate change and variability and in the applications of the data to societal problems. While this is not a new issue, many discussions have taken place during the workshop on how Japanese and United States scientists could lead or collaborate in improving the current situation. One possible setting to foster this collaboration may be the Coordinated Enhanced Observing Period (CEOP), where United States and Japanese scientists are already working together (along with the international GEWEX community).

CEOP is currently collecting model forecasts and analyses, as well as in-situ and remotely sensed observations. The observing period will be for more than three years. This convergence of many types of data should provide a critical period for global and regional model development and validation. With multiple models, analyses and observations, uncertainties for certain parameters can be quantitatively estimated. While this is only a three-year period (short for climate studies), GCMs could be evaluated by producing climate statistics given appropriate forcing.

Recommendation: Make CEOP a benchmark period for model (global and regional) validation and development and promote interdisciplinary studies to evaluate the impact of model biases on societal decision-making and sub-models (e.g. ground water, water quality etc.).

Action Item:

Encourage American and Japanese participation in CEOP program to not only provide data but also to utilize the data for various scientific studies. Develop independent model intercomparisons over the CEOP time period.

Management of surface water quality is often complicated by interactions between surface water and groundwater. Traditional Land-Surface Models (LSM) used for numerical weather prediction, climate projection, and as inputs to water management decision support systems, do not treat the lower boundary in a fully process-based fashion. LSMs have evolved from a leaky bucket to more sophisticated land surface water and energy budgets that typically have a so-called basement term to depict the bottom model layer exchange with deeper aquifers. Nevertheless, the LSM lower boundary is often assumed zero flux or the soil moisture content is set to a constant value; an approach that while mass conservative, ignores processes that can alter surface fluxes, runoff, and water quantity and quality. Conversely, models for saturated and unsaturated water flow, while addressing important features such as subsurface heterogeneity and three-dimensional flow, often have overly simplified upper boundary conditions that ignore soil heating, runoff, snow and root-zone uptake.

Recommendation: Joint US-Japan studies of the coupled atmosphere-land-groundwater need to be initiated. Research watersheds, one each in the United States and Japan, need to be identified. These watersheds should be part of a larger set of study basins (e.g. GEWEX, HELP). Regional models would be a useful starting point for global models.

Action Items:

A workshop to determine watersheds to be used as part of these studies and an exchange of scientists to develop new models and datasets

To quantify global water cycle processes, the implementation of natural isotopes as tracers in the atmosphere, land-surface, and ground water is viewed as an advancement in model development. Inclusion of hydrogen and oxygen isotopes as tracers will provide a means for measuring atmospheric flux sources and sinks, land-surface and sub-surface fluxes and the partitioning between old and new water.

Atmospheric water isotope cloud physics packages and land-surface/sub-surface schemes need to be developed in collaboration with the International Atomic Energy Agency and the Global Natural Isotopes in Precipitation Program. Joint research and development among Japan and U.S. modeling groups working through partnered projects, workshops, and scientific exchanges programs will help to bring this about.

Recommendations:

More coordination of sampling and measurement through IAEA.

Develop a U.S.-Japan working group to develop a major 10-year initiative.

An important area for U.S. -- Japan collaboration is in the area of incorporating water management practices (e.g. irrigation, reservoir storage) in land surface models. Human manipulation of surface and groundwater likely has significant impacts on regional, and perhaps global climate, and its representation in land surface models has been lacking.

Recommendations: A two-level model development strategy could be pursued. One level could be coarse representation of irrigated lands and reservoir storage in land surface and river transport models. Such a representation will enable important sensitivity studies with regional and global climate models in which Japanese and American scientists could collaborate.

A second level could model these and other aspects of water management (e.g. water transfers) in more detail in offline models. These offline models could be used as “reference -- models for the coarser representation in climate models, and could include detailed information such as irrigation scheduling and reservoir operating rules. The offline models would also be critical for downscaling output from global climate change scenarios. Assembling the relevant data sets will require considerable effort and should involve the Japan, the U. S. and other international collaborators. The GAPP applications program for incorporating water management practices into streamflow prediction could provide valuable guidance for how to proceed with this effort.

More generally, both countries could collaborate on the further development of the coupling between land surface and river transport models. The representation of lakes and wetlands along the channel network are a good example. These features impart important storage lags and are a function of both water table dynamics in the land surface model and the morphology of the landscape along the network. Realistic modeling of these storages, which are important to both the water cycle (through evapotranspiration and hydrograph impacts) and biogeochemical cycles (through trace gas fluxes), requires a holistic approach.

Begin exchange of relevant data sets for irrigation, reservoir storage, water management practices.

Action Item: Workshop between the United States and Japan on water management practices.

Under a warming climate, changes in permafrost will alter both the rate of fresh water input to high latitude oceans and the release of methane from peat land bogs. These two feedback mechanisms may contribute to a thermohaline circulation slowdown and an increase in green house gas absorption, respectively. Land surface models in current climate models include ice phase in soil moisture and represents permafrost processes in a simple fashion. This approach does dynamically handle the complexity of the land surface -- subsurface rates of change of multi-phase water movement. A one-dimensional permafrost model needs to be incorporated into land surface models as an advanced model development.

Recommendation: International collaborations with contributions from the United States and Japan are needed. Model development of permafrost processes and the associated biogeochemical processes will be need to be developed as part of advanced land-surface modeling studies and evaluation.

Action Item:

Support for scientific exchange through IARC.

On the US side, there are separate Carbon Cycle Initiative and Water Cycle Initiative. On the Japanese side, there is the All Japan Carbon Cycle Project, and various water cycle projects funded by various agencies. There are about 30 to 40 land surface parameterization schemes simulating water and energy cycles without considering carbon cycle. There are also many carbon cycle models without careful consideration of the energy and water cycles. The linkage of water and carbon cycles is concurrent photosynthesis and transpiration processes. Most models have in one way or another considered the photosynthesis -- conductance relationship, but many other linkages between the water and carbon cycles have not been considered or understood yet. For examples, most of the biophysical land surface models use specified vegetation parameters. In reality, most vegetation parameters, such as rooting depth, albedo, roughness, and water holding capacity, vary with age of the vegetation. For carbon cycle models, in contrast, the water cycle treatment is typically simple (e.g. Sim-CYCLE with only 2-layer).

Recommendation: Recognizing the importance of water and carbon cycles in climate and global change, we recommend having a joint Japan-U.S. workshop of water and carbon cycle modeling. The purpose of this workshop is to: (a) bring water and carbon modelers together; and (b) develop a priority list for joint research in linking water and carbon cycles. The current FLUXNET and to certain degree CEOP in which both Japan and the United States are playing leading roles could provide the necessary datasets for the model evaluation and validation.

Action Item: Ito and Zeng to develop focused workshop

The global oceanic thermohaline circulation is a main driver of climate on long time scales. The formation of deep water, which forces the THC, is critically sensitive to freshwater inputs from the continents, sea ice, and precipitation. A key area for collaboration is hence the impact of continental freshwater outflows on the THC.

Recommendations: Areas for exploration include the uncertainty in riverine outflow rates from routing models and the need for streamflow observations at high latitudes for validation; the representation of river freezing in river transport models; and the sensitivity of North Atlantic Deep Water formation to ocean model physics and dynamics.

Action Items: Perform simulations with U.S. and Japanese coupled models and interact with ocean measurement programs.

Co-chairs: Shannon Cunniff, Hiroyuki Kawashima

Participants: Toshiya Aramaki, Susanna Eden, Harvey Hill, Rick Lawford, Takao Masumoto, Dave Matthews, Haruyasu Nagai

IFPRI and IWMI have predicted the rapid rise in demand for the world's increasingly scarce water supply. Agriculture needs will continue to complete with industrial and municipal demands. Environmental uses will be stressed and water quality will decline. As water resources degrade, the cost of developing new sources and treating water is increasing.

Climate change will affect both water supply and water demand. As such, knowledge of both the potential and significance of climate change on global and regional shifts in supply and demand will be crucial for development of policies and programs and will likewise be critical for regional and local water resources planning. Understanding the factors affecting demand and patterns of use will enable policy makers to evaluate the most effective options for meeting demands with supplies. The impact of climate change must be integrated with other environmental, technological, social, and economic factors to effectively identify the challenges and evaluate opportunities for averting crisis.

Working Group 6 focused on issues related to enhancing the assessment of water supply and demand and integrating this information with climate change and other factors to enhance informed planning and decision-making regarding strategies to minimize vulnerability and increase flexibility.

Discussion guiding questions included:

Better predictions or forecasts will increase confidence and allow water managers to more effectively incorporate this new information. Data on water supply, yields, demand, and actual use are inconsistent, making meaningful comparisons and conclusions difficult.

Early collaboration is necessary to build the compatibility necessary for developing the capacity to create sophisticated models necessary for integrating global climate change with the numerous other factors effecting water supply and demand.

To a degree, we know where the potential crises are likely to be based on current conditions and trends.

There will be regional differences in the ability to deal with water stress. In Japan, there will be no water crisis where river systems are well controlled. Where there is a lack of control for irrigation withdrawals in other places in Asia, such as the Mekong, there is a higher likelihood for crisis from competing demands for water. In the United States, population shifts to water stressed regions indicate the need for increased attention to long-term planning and development of incentives to reduce demand and re-align uses. Differing political systems will require different strategies to address issues.

Urbanization and water for food production are the biggest challenges of the next century, especially for developing countries.

Changing life styles, including diet, will place additional demands on land and water well in excess of current demands.

Application of new technologies can help meet demand.

Occurrence of more extreme events such as floods and drought will financially stress countries and could cause temporary and long-term population shifts.

Reducing demand is critical for future water resources planning.

Water resources planning typically considers 25 to 50-year horizons in part to ensure that infrastructure development and institutional change can keep pace with demand. Using the emerging knowledge of climate patterns and climate change is crucial to sound planning.

There are experts in demand estimation as well as water use that need to be consulted to further develop these ideas.

There are compelling needs to:

Accurately forecasting demand and consumption will require more accurate estimates of cropping patterns (types and acreage), type of irrigation, if any, or the potential to use alternative irrigation methods. Similarly, trends in agricultural research need to be accounted for to consider the effects of changing yields, the advent of expanded use of drought and salt tolerant species, and improvements in irrigation technologies, to name just a few factors. Shifts in economic status, life styles and diet will also affect demand and consumption. Expansion of new water supply technologies, such as water recycling and desalination, could significantly alter water stresses. Ground water conditions, as well as the potential for aquifer storage and retrieval and strategies that more effectively and sustainably integrate groundwater and surface water management, need to be considered. These are only a few of the potentially significant factors whose evaluation and integration are needed to more accurately predict global, regional, and local water yields, water demands and actual consumptive use. Global climate change has the potential to affect many of these factors beyond simply the quantity and timing of water supply. For example, climate change impacts consumptive use by crops, evaporation rates of water storage and delivery infrastructure, and water quality. These factors need to be an integral part of analyses to begin to construct meaningful planning scenarios that can guide decision making at multiple scales. Scenario development can help to identify “no regret -- decisions on investments in technologies or actions, e.g., efficient irrigation system. Water supply and demand assessment would support scenario development.

In developed countries, such as the United States and Japan, where water control and supply infrastructure is well established, modeling emphasizes decision support tools. Demand management is, however, becoming of increased importance in the United States. In developing countries, where control of supply is not as advanced, demand management appears to be more important. Transitional countries in the process of developing water supply infrastructure, such as China, may benefit from tools assisting both supply and demand.

Scale (global, regional, local) will be important in data acquisition, tool development, and interpretation of results. Enhanced understanding of water supply and demand is needed at all scales, local to global.

There exist common interests among researchers in Japan and the United States due, in part, to the similar level of development of water resources infrastructure. Development and testing of hypotheses in a basin in Japan and in the United States will help to create a more robust model. Analysis of a basin in an emerging country would yield additional information and allow expansion of model applications.

APPENDICES

7:45 Shuttle to Beckman Center -- Meet at Hyatt Newporter Front Drive

8:00-9:00 Breakfast - Beckman Center Dining Room

9:00-9:30 Registration-Check in Tables Near Auditorium

9:45-10:15 Co-conveners' Addresses

10:15-10:45 Keynote Addresses

10:45-11:00 Break -- Refreshments Available in Atrium

11:00-12:30 Keynote Addresses -- Auditorium

Soroosh Sorooshian, University of Arizona

3:30-4:00 Keynote Address -- Auditorium