A53A-0847 1340h
Introducing Aerosol Observations in a Regional Atmosphere Model
The impact of aerosols on weather and climate is becoming increasingly important, while different satellite platforms continue to provide reliable information about global aerosol distribution in terms of vertically integrated depth and profiles. However, description of aerosols in numerical weather prediction and regional climate models has been neglected up to now. This neglect of aerosol parameterization reflects an important deficiency in the derivation of three important surface exchanges; namely, incoming shortwave radiation, incoming longwave radiation and precipitation. To confront this challenge, aerosol optical depth retrieved from the Model of Atmospheric Transport and Chemistry (MATCH) is ingested into the MM5 model to interact with the model's radiation scheme. The ingested aerosol data does not affect the cloud microphysics. The experiment is done to see how the coupled aerosol data will affect the local radiation and heat balance within the model. Preliminary results will be presented. Later, size distributions and total particle numbers within the cloud microphysics of a mesoscale model will be based on the MATCH output of aerosol mixing ratio for the five different aerosol types resolved by MATCH. This will allow us to see the impact of aerosol nudging in model-simulated clouds and precipitation by making comparison with the Tropical Rainfall Measuring Mission (TRMM), Moderate Resolution Imaging Spectroradiometer (MODIS), and ground site observations. It is anticipated that the more reliable components of surface energy and water cycle will be simulated from regional to continental scale with a mesoscale model.
A53A-0848 1340h
Development of a Regional Dust Transport Model: Preliminary Sensitivity Studies
The atmospheric transport of mineral dust is difficult to reproduce using global circulation model because their coarse spatial and temporal resolutions limit their ability to represent the small-scale processes that control the generation and the deposition of these particles. In the framework of a model/satellite intercomparison project, we are developing a new mineral dust transport model named CHIMERE-DUST. This model relies on the off-line chemistry-transport model CHIMERE and enables to generate concentration fields of mineral dust at high temporal (1h) and spatial (a few kilometers) resolutions for long time periods (several years). The modeled area include the dust sources of western Africa and the zone of long-range dust transport in the North Atlantic: from 1 degree to a few kilometers resolution, model domains are designed to perform nested simulations. Model simulations are compared to optical thicknesses measured by the PHOTONS-AERONET Network, to in-situ surface concentrations measurements and to satellite aerosol products (TOMS AI and METEOSAT optical Thickness). The accuracy of the results is discussed in term of sensitivity to the parameterizations implemented in the model for surface emission fluxes (depending on the parameterization employed, the sandblasting process used), for the number of dust bins (as well as the range values of these bins) used to account for the particle size distribution, and finally for the dry deposition.
A53A-0849 1340h
Aerosols deposition in a GCM Eulerian model with variable spatial resolution
Transport and deposition of cosmogenic 7Be is simulated by the GCM model, LMDZ, developed at the Laboratoire de Meteorologie Dynamique in Paris. The model calculates eulerian large-scale advection based upon finite-volume methods and parameterization of turbulent mixing and convection. Simulations with a 2° horizontal resolution and 19 levels vertically, are carried out in a "nudged" mode, where horizontal velocities and temperature are relaxed towards wind analyses ECMWF, available every 6 hours. Transport is based on mass conservation of the aerosol in the atmosphere, taking into account radioactive decay, dry and wet deposition. Concentrations of 7Be are initialized in the atmosphere as a function of latitude and pressure, based on data of. Lal and Peters (1967).We show that our model reproduces with reasonable accuracy the time series of 7Be recorded over 6 months and by 20 stations distributed over the globe, which essentially validates the precipitation model. Numerical results show the respective influence of convective and stratiform precipitation. The LMDZ model allows the use of a variable mesh define by coordinate stretching or "zoom", to increase resolution over any region of interest. Sensitivity tests are carried out by varying the spatial resolution of model, with resolution as fine as 0.5° by 0.5° over Europe.
A53A-0850 1340h
Downscaling Ability Of NCEP Regional Spectral Model, Version 97: The Big Brother Experiment
Regional climate modeling has become an important tool for the prediction of climatic changes in the past few decades. Current General Circulation Models (GCMs) still have resolutions too low to be able to resolve small-scale atmospheric circulation features, such as may be produced by orography or regional land surface details. Instead, limited-area models have been used to produce climatic simulations of certain regions with boundary conditions derived from observational data or a global General Circulation Models. The Big Brother Experiment (BBE) consists of first providing the reference climate by performing the large domain high resolution regional climate model simulation (The Big Brother). Then, this reference simulation is degraded by filtering out short scales that still remain unresolved in the GCMs. This new, filtered reference climate is then used to drive the same regional climate model (RCM) integrated at the same resolution as the Big Brother, but over a smaller domain embedded in the larger domain. This smaller domain simulation is called the Little Brother. We then finally compare the climate statistics of the Little Brother with those of the Big Brother over the Little Brother Domain. Since the same model is used for both the Big Brother and Little Brother experiments, any difference in the solutions can only be attributed to the nesting strategy. A good comparison between the two solutions is a necessary, though not sufficient prerequisite for the use of the standard one-way nesting strategy for regional climate modeling. Our study extends that of Denis et al. (2001) by using a different model, and mainly focusing on tropical rather than extratropical experiment. Our primary domains are located in the tropics of the South America and the Atlantic Ocean, with spatial resolution of 60 km, 180 x 181 grid points for Big Brother, 90 x 91 grid points for the Little Brother, and vertically, 19 model levels (18 layers). The simulation was completed for April 1994. The Big Brother simulation was driven directly by the ECHAM4.5 global model data set, archived every 6 hours. The Big Brother output was saved every 6 hours as well, for the nesting of the Little Brother. The initial experiments yielded that the small-scale features that were absent from the initial conditions were almost fully recovered within the first 24 hrs in most variables, except for the precipitation. The time variability of the daily domain-averaged precipitation patterns was well reproduced. The results obtained in our primary experiment resulted in further experimentation over a midlatitudes of North American continent. These simulations have yielded a significant improvement in recovery of the small-scale features, especially in the precipitation field, which will be presented in more detail. The paper is expected to be in press this Fall.
A53A-0851 1340h
Regional Climate Study over the Kyushu, Japan based on the annual simulation of a Mesoscale Meteorological Model
High-resolution mesoscale meteorological model with two-way nesting technique (minimum grids: 2km) was applied to study the seasonal variation and the geographical distribution of local circulation, prevailing wind and regional climate over the Kyushu, Japan. As a mesoscale meteorological model, RAMS (Regional Atmospheric Modeling System) was used and all of cloud microphysical process for precipitation, cumulus convection, boundary-layer turbulence and land surface process options were used in this study. The model results were validated using the Japan Meteorological Agency's observation network data within Kyushu area. We found the following, (1) the simulated temperature and wind speed explained the daily and seasonal variations of observed data very well. Correlation coefficient between the model results and the observation data for wind and temperature took the very high score, (2) the annual variation of the daily average wind speed on the Pacific Ocean side took the smaller value than that of the Japan and East China Sea side, (3) model results retrieved the precipitation events very well, however the model underestimated the total precipitation amount. We also found that the simulated temperature tends to overpredict during the overcast and/or small precipitation conditions, especially in summer time, (4) analysis of the frequency distribution of wind direction and wind speed shows that the model correctly caught the prevailing wind field for each season and we can confirm the importance of the local circulation (such as the land and sea breezes) for the formation of regional climate in Kyushu, (5) composite analysis indicated that the sea breeze from the Japan Sea and East China Sea side invaded further inland region than that from the Pacific Ocean side in summer.
A53A-0852 1340h
A Comparison of Statistical and Dynamical Downscaling for Surface Temperature in North America
A multiple linear regression statistical downscaling model is applied to estimate summer monthly mean surface temperatures over eastern North America. The model is calibrated with time series of $0.5\deg$ x $0.5\deg$ gridded observations from the U.S. Historical Climate Network (the predictand field) and a transient climate simulation with the NASA Goddard Institute for Space Sciences coupled ocean/atmosphere General Circulation Model (GCM), using forcings from the SRES A2 scenario (the predictor field). Three scales of predictor domain are assessed, ranging from regional to continental, in addition to predictor domains selected pointwise by objective analysis. The downscaling model is found to be much more sensitive to predictor domain than to the inclusion of mean sea level pressure as an additional predictor variable. Performance is compared to a dynamical regional climate simulation with the PSU/NCAR Mesoscale Model v.5 (MM5) for both current conditions (1997-1999) and future projections (2020s, 2050s, 2080s) with a standard suite of diagnostic statistics. The statistical and dynamical models exhibit comparable skill, and all consistently project greater warming than the GCM. Predictions from statistical and dynamical models approach convergence when adjusted for model bias. This experiment illustrates the research potential for employing GCM-derived surface temperature as a predictor field and highlights the importance of predictor domain selection in downscaling climate change scenarios.
A53A-0853 1340h
A Regional NWP System for Southern California: Applications of GPS-Derived PWV Retrievals
Complex local orographic and meteorological characteristics make natural hazards, such as flash-flooding and wildfires during the wet and dry seasons, respectively, an important concern in the Southern California region. Precipitation in Southern California is characterized by an extreme seasonal cycle with a high frequency of heavy downpours during the wet season. As many of the densely-populated areas are located in valleys and on surrounding hillsides, heavy precipitation events that last only a few hours can cause substantial amounts of damage. In addition, hot and dry conditions at the end of summer combined with high fuel loads due to dry vegetation, make wildfires another leading source of natural disasters in the region. Hence, improving quantitative precipitation and fire-weather forecasting is a crucial concern for reducing the damages from weather-related natural hazards. The dense Southern California Integrated GPS Network (SCIGN) provides precipitable water vapor (PWV) data at high temporal and spatial resolution from over 200 receivers located throughout the greater Los Angeles area. These constitute important local data for improving the accuracy of initial water vapor fields for quantitative forecasting and process modeling that are crucial for short-term NWP. Here, we review recent collaborative efforts towards application of SCIGN data to studies of water vapor variability in the Southern California region and applications of the GPS-derived water vapor data to fine-scale regional forecasting. The focus of this presentation includes retrieval of PWV from GPS receivers over complex terrain, the effects of the additional GPS-PWV data on regional simulations, and their implications for useful NWP lead times.
A53A-0854 1340h
A strong boundary layer mixing process used in the Canadian Regional Climate Model
In this study, a strong boundary layer mixing process from the Canadian Atmospheric General Circulation Model (AGCM) is implemented into the third generation of Canadian Regional Climate Model (CRCM) as an effort to improve the simulation of vertical distribution of temperature and water vapor in the boundary layer, further to improve the model performance in simulating summer land precipitation. In stead of exchanging latent and sensible heat fluxes only with lowest boundary layer, the new strong mixing process evenly imposes the increments of water vapor and potential temperature on the whole boundary layers so as to directly mimic the non-local mixing effects during the strong surface heating events. The new mixing scheme (along with some adjustment in the soil water holding capacity field) significantly improves the vertical distributions of water vapor and temperature in the boundary layer; the spatial distribution patterns of total cloudiness are also improved while compare with the observation from the International Satellite Cloud Climatology Project (ISCCP). In addition, as a consequence of improvement in the boundary layer simulations, the deep convection is triggered more realistically than before, so the intensity of convective precipitation is reduced drastically, this results in the simulation of the total summer precipitation becomes more realistic while compare with the observation.
A53A-0855 1340h
Evaluation of a Regional Climate Hindcast for East Asia
As a preliminary step in a climate change and impact assessment study for East Asia, which is among the most vulnerable regions to climate change, we have analyzed a long-term simulation by a regional model in hindcast mode. Socio-economical developments and accompanying urbanization in East Asia are placing ever increasing demands on natural resources that are already far stretched in many parts of the region, such as water and food. Shifts in the water cycle due to the climate change induced by anthropogenic emissions of greenhouse gases will inevitably affect human sectors in the region. Hence, assessing the regional climate change and its impacts on the water cycle is a crucial step for planning long-term sustainable development. Climate change projections for impact assessment studies are usually generated by dynamical models. Uncertainties originating from model errors remain an important concern in the interpretation of the results. As it is impossible to avoid model errors, the close examination of model results and errors is an important task in projections of future climates. In this study we use the Mesoscale Atmospheric Simulation (MAS) model driven by the NCEP R2 for the 22-yr period 1979-2000. The model simulation was successful in reproducing, at least qualitatively several important features of the regional climate, such as the spatial distributions of precipitation and temperature. The spatial anomaly correlations between the upper-air wind fields from the simulation and the R2 remained well above 0.95 throughout the 22-yr period, suggesting that the simulated structures are consistent with the large-scale forcing that drove the simulation. The simulation also reproduced extreme hydrologic events as measured by their recurrence periods. The model results showed significant local biases, however. An examination of the simulated standard deviations and coefficient of variations suggested that the simulated variables scaled by the model climatology can compare more closely with the similarly scaled data constructed from observations than the raw model data. For example, the precipitation data scaled with its own climatology agreed more closely with observations than the raw model data.
A53A-0856 1340h
Regional Climate Model Downscaling of the U.S. Present Climate and Projected Change
The MM5-based regional climate model (CMM5) simulations driven by the NCEP-DOE AMIP-II reanalysis and the integration of the NCAR Parallel Climate Model (PCM), an atmosphere-ocean coupled general circulation model (CGCM), for the present climate are inter-compared with observations to study the CMM5 downscaling skill and uncertainty. The result indicates that the CMM5, with its finer resolution and more detailed physics, simulates the present U.S. climate that is more accurate than the driving reanalysis and PCM output, especially for precipitation, including annual and diurnal cycles and daily variability. Hence, the CMM5 downscaling provides a credible tool to improve CGCM climate simulations. A parallel CMM5 run driven by the PCM future (2046-2050) projection is then compared to determine the downscaling impact on local-regional climate change. It is shown that the PCM-driven CMM5 generates very different patterns of U.S. climate change projections, with more spatially variable precipitation changes while smaller temperature increases than the PCM itself. This study suggests that the CMM5 downscaling can significantly reduce the uncertainty due to inadequate spatial resolution and incomplete physical representation of local and regional processes and hopefully the overall uncertainties among CGCMs.
A53A-0857 1340h
A Regional Climate Model Simulation of U.S. Soil Temperature and Moisture during 1982-2002
The MM5-based regional climate model (CMM5) simulation of the United States soil temperature and soil moisture annual cycle and interannual variability during 1982-2002 driven by the NCEP-DOE AMIP II reanalysis (R-2) is compared with observations and outputs from the R-2 and North American Land Data Assimilation System (NLDAS). The comparison demonstrates that the CMM5 has a pronounced downscaling skill. The CMM5 realistically simulates a warming trend of annual mean soil temperature during the past 20yr with a slightly smaller rate than observations. The seasonal mean trends are also realistic except that a decreasing trend in fall is opposite to observation. The CMM5 simulated annual cycle and interannual variations of soil moisture are in good agreement with measurements in Illinois (whole year) and Iowa (April-November). The largest CMM5 differences of soil temperature from observations and of soil moisture from NLDAS outputs are identified over the Great Plains. Although such soil temperature differences may be associated with the CMM5 biases in surface air temperature, the soil moisture differences do not correspond to precipitation differences. The result emphasizes the need for more comprehensive study on model evaluation and bias understanding of soil temperature and soil moisture.
A53A-0858 1340h
Comparison of Radix Similarity and Level 2.5 Turbulent Parameterization Scheme in the Prediction of Wind and Temperature Profiles in the Convective Boundary Layer
Santoso and Stull (1998, 2001) propose radix similarity equation (RSE) to describe the mean wind speed and potential temperature profiles in the radix layer (RxL): roughly the bottom fifth of the convective boundary layer. Their results suggest that the observations within the RxL collapse into similarity curves. The good ability of RSE in capturing the convective boundary structure motivates us this work to gain some insight on the appropriate treatment of boundary layer physical processes and the parameterization of turbulent transport processes. This work aims to compare the wind and temperature profiles predicted with radix similarity equation (RSE) and a level 2.5 turbulent parameterization scheme that currently in used in MM5 1-D model against First ISLSCP (International Satellite Land Surface Climatology Project) Field Experiment (FIFE) data. The preliminary results suggest that RSE realistically represent the evolvement of the wind and temperature in the convective boundary layer. MM5 1-D model simulation underestimates wind and overestimates potential temperature near the surface. Future work will attempt to carry out more comparisons among the selected observation cases using RSE to further investigate the deficiencies of the current turbulent scheme. Then the good aspect of RSE will be applied to the actual turbulent parameterization scheme to help improve model performance.
A53A-0859 1340h
Regional Climate Prediction and Water Resources Management: Hydropower Applications
The great concern due to a recent energy crisis in Brazil (in which hydropower accounts for the major part of power generated) brought to authorities' attention the need for better monitoring and predicting both precipitation and reservoir inflows. The possibility of streamflow prediction several months ahead can play a major role in regional economies highly dependent on hydropower generation and irrigation for crop yield. The streamflow variability is very connected to climatic signals which occurs in the tropical zone of the globe. In this context, the use of atmospheric numerical models for climate prediction is becoming widely used, as a tool for many applications, including management of water resources. As mesoscale models are able to better represent atmospheric circulations induced by local forcing (such as topography, land-water and land use contrasts, etc.), they are used nowadays for climate prediction at the regional scales, usually with the input of data from general circulation models (GCMs). In this work, a mesoscale model (the Regional Atmospheric Modeling System, RAMS), forced by data from the ECHAM model, is used to simulate the interannual variability over a major river basin in Brazil (Sao Francisco basin). Emphasis is given in the analysis of the precipitation field, to allow predictions of river discharge and reservoir levels, especially critical in hydropower reservoirs at the Sao Francisco basin. An 80x80 horizontal grid is used, with a 40 km grid-spacing in both directions, along with a vertically-stretched, 41-level vertical grid. A 30-year climatology for one of the ECHAM members was generated, and the mesoscale model sensitivity to the large-scale model forcing was explored, in order to achieve the best representation of the interannual variability. Along with seasonal prediction, the use of regional atmospheric model for short term streamflow forecast and simulation of scenarios for long term hydropower planning are also future goals in this project.
A53A-0860 1340h
Modeling and Evaluating the Impact of Deforestation on Precipitation over Central America
Deforestation in the countries comprising Central America can have a profound effect on climate, especially precipitation. Deforestation can induce these important local and regional effects through changes in the surface albedo and the evapotranspiration of water back into the atmosphere. In particular, the increased surface albedo and decline in evapotranspiration that follow the removal of forests both act to reduce precipitation, especially convective thunderstorms, by stabilizing the atmosphere and by reducing the amount of water available to contribute to rainfall. Surface roughness changes are also thought to be important in changing moisture transport. Deforestation can also act to modify or change the way in which Central America is influenced by interannual climate variability, especially the changes due to ENSO. Because of the importance of small-scale orography and land use issues, large-scale models and reanalyses by themselves are insufficient to fully understand and model precipitation variability over C.A. Regional modeling capabilities are critical as a means to examine how large-scale forcing gets downscaled by orographic effects, and to explore the impact of land surface state. We have made runs with the regional climate version of MM5 with the OSU (NOAH) land surface scheme, and with a simpler bucket hydrology scheme. When forests are replaced with grassland (pastureland), temperatures warm over almost all of Central America (and southwest Mexico) by up to 6 deg. C. Precipitation also shows a sharp decrease across much of the region. Changes are largest where the land mass is largest (i.e., smallest changes are found over the narrow countries of Panama and Costa Rica). Evaluating these results includes estimating how much deforestation has actually occurred, and at what rate, as well as using all available observational datasets (especially TRMM) to evaluate model capabilities at simulating precipitation variability over Central America.
A53A-0861 1340h
The Impacts of Vegetation Growth on the Development of the North American Monsoon System (NAMS)
The present study is to explore the impacts of vegetation growth on the developments of the North American Monsoon System (NAMS) by using the newly-released Weather Research and Forecasting (WRF) model version 2.02 coupled with different land schemes. To represent vegetation growth, we incorporate an interactive vegetation canopy (IVC) scheme, which calculates the vegetation carbon budgets (carbon assimilation, carbon allocation to leaf, stem, root, and wood, and respiration) and leaf area index (LAI), into the NOAH land scheme in WRF. LAI is exponentially transformed into fractional vegetation cover (Fveg) to couple with NOAH. A series of experiments are conducted from June 1 to August 31 of 2002 by using WRF with different land schemes, 1) SLAB, which predicts soil temperature but with constant soil moisture and no vegetation, 2) NOAH, which predicts soil temperature and soil moisture but with constant Fveg, 3) NOAH/SWF, which is based on NOAH but the root water uptake factor is modified to be close to a step function instead of a linear function of soil moisture, 4) NOAH/IVC, which is the same as NOAH/SWF but linked to IVC, and 5) NOAH/IVCR, NOAH/IVC with deeper roots. The model results are as follows. 1), the rainfall is strongly sensitive to land surface processes. SLAB fails to simulate the monsoon rainfall in all months, while NOAH produces comparable rainfall to observations except for August. 2) NOAH/SWF increases evapotranspiration (ET) and rainfall in southern NAMS region in August. 3) NOAH/IVC significantly improves the rainfall simulation in the southern NAMS and Southern Great Plain (SGP), especially in August. 4) Increasing rooting depth does not increase ET and rainfall in this study because the deeper soil layer is even drier than its upper layers when the model is initialised with the NCEP/NCAR Reanalysis. This indicates a more accurate initialization of soil moisture or a longer peroid of integration may be required.
http://www.geo.utexas.edu/climate/Research/publications.htm
A53A-0862 1340h
Effects of the Sky View Factor on the Nocturnal Cooling Rate of Urban Air Temperature: Numerical Experiments by an Urban Canopy Model
The diurnal variation of air temperature is different between in urban and rural areas. In urban area, the nocturnal cooling rate is moderate and constant until sunrise, while in rural area, the rate is large from sunset to around 20 Local Time (LT) and is changed to be small from 20 LT to sunrise. Using a simple urban canopy model, this work examines the physical mechanisms how such differences appear between in urban and rural areas. In particular, this work focuses on the small sky view factor and the large thermal inertia in urban area. The numerical experiments identify that the sky view factor is responsible for the differences appearing in the nocturnal cooling rate, while the thermal inertia controls the diurnal range of temperature rather than the cooling rate. The roles and contribution of anthropogenic heat in urban area would further be discussed at the meeting.
A53A-0863 1340h
Regional Climate Modeling for Global Warming Prediction in MRI/JMA
Regional Climate Models have been powerful tools to investigate and predict regional climate change over Japan due to Global Warming. MRI/JMA is developing new RCMs for Global Warming prediction: 1) Cloud-Resolving Nonhydrostatic Regional Climate Model (NHRCM, hereafter) with 5km resolution and 2) Atmosphere-Ocean Coupled Regional Climate Model (CRCM). The Cloud-Resolving Nonhydrostatic Regional Climate Model was developed in the project _gResearch Revolution 2002_h using the Earth Simulater, the biggest computer in the world. The NHRCM has 5km resolution (grid number: 800x600) and contains sophisticated cloud microphysics as precipitation process. The Spectral Boundary Coupling (SBC), which was developed and was modified for nonhydrostatic model in MRI, was adopted to introduce large-scale information from an outer model. The model is to predict climate change over Japan including small-scale intense rainstorm, which is generated by well-organized convective system of several tens of km. Long-term integration with objective analysis data for lateral boundary condition showed that the model is capable of reproducing present climate with detailed structures of precipitation. Time slice experiments for regional climate change due to Global Warming was conducted targeting the rainy season of Japan in June and July. Lateral boundary condition was supplied from a Global Climate Model of 20km resolution, which was also developed in MRI/JMA. Results by the NHRCM show that the Baiu front does not move northward and stayed along the southern coastal line of the main island of Japan in the future. As a result, in July, precipitation in western Japan increases on the southern side due to Global Warming. At the same time, number of intense precipitation also increase during the season. On the other hand, in northern part of Japan, precipitation amount decreases. Atmosphere-Ocean Coupled Model was developed to predict regional climate with atmosphere-ocean interactions in more sophisticated manner. Using the CRCM, regional climate changes of the ocean can be predicted as well as atmospheric changes around Japan. The CRCM consists of two part, a High-Resolution Ocean Model (Lon:1/4, Lat:1/6),which was developed in Oceanographic Research Department, MRI, and a 20km-mesh RCM. The 20km-mesh RCM is a hydrostatic model with parameterized precipitation processes. The SBC for hydrostatic model is used as the nesting scheme. They are coupled with each other by a coupler every hour. A present climate experiment and a warm climate experiment for Global Warming are being conducted. They are not completed at this stage, but preliminary results show that SST of the Japan Sea is improved by the CRCM. Details of the models and results of prediction will be shown in the meeting.
A53A-0864 1340h
The Role of Land Models in the FSU Regional Climate Model and its Implication to Crop Model Forecasting
The National Center for Atmospheric Research Community Land Model (NCAR CLM2) is coupled to the Florida State University (FSU) regional climate model to improve the seasonal land surface climate and to apply its output to the CERES-maize (Crop Environment Resource Synthesis) crop model. The regional model is placed over the Southeast United States and run at 20 km resolution, roughly resolving the county level. Outputs from the models (max/min surface temperature, precipitation, and shortwave radiation at the surface) are used as inputs into the crop model to determine the crop yields. Simulations with the atmospheric model coupled with the CLM2 (hereafter, FSUCLM) are compared to the control (FSUc, The FSU simplified land surface scheme includes a three layer soil temperature model based on the force-restore method). Results show that the FSUCLM experiment improves the seasonal simulation relative to the control in the regional model. Noticeable improvements were found in the simulation of the surface temperature, evaporation, and latent heat fluxes. The CLM2 reduces much of the surface temperature cold bias noted in the FSUc run. The inclusion of the CLM2 helps produce better crop yield forecasts.
http://www.coaps.fsu.edu
A53A-0865 1340h
Regional Simulations of the Water Cycle and its Relation to Soil Moisture Anomalies
The impact of soil moisture on the precipitation associated with the onset of the monsoon of South America is investigated with ensemble simulations of the National Centers for Environmental Prediction (NCEP) regional Mesoscale Eta model. The model has been evaluated in short and long term integration modes. Long term simulations show no evidence of a climate drift, while short term integrations reveal that it is a valuable tool for water cycle studies. The area average precipitation over La Plata basin over a recent two year period is similar to observations. Moreover, the ultimate verification of the reliability of the model estimates is by comparison of moisture flux convergence with river discharge. The long term river discharge of La Plata is equivalent to 0.61 mm day-1, which is very close to the MFC estimated from the Eta model (0.60 mm day-1). On the other hand, the model appears to overestimates evaporation by about 0.2-0.3 mm day-1 in the annual average. Multiple month long simulations are employed in ensemble mode to discuss the sensitivity of these features to changes in soil moisture. In particular, it is a well known fact that El Niño has a strong influence in south eastern South America, and particularly over La Plata basin. However, a less understood feature is that precipitation (and floods) can have additional peaks beyond the El Niño phase of the ENSO cycle. This less understood feature is the subject of the current presentation. We will discuss the hypothesis that abundant El Niño-induced rain during the southern winter favors anomalously moist soils during spring, hence providing an additional source of moisture and instability for precipitation even after El Niño has finished.
A53A-0866 1340h
High-Resolution Regional Climate Model for the Pacific Northwest
We have developed a high-resolution climate model of the Pacific Northwest based upon the MM5 mesoscale weather model and performed downscaling simulations forced by output from the National Center for Atmospheric Research-Department of Energy Parallel Climate Model (PCM). Decade-long simulations will be presented for present-day climate and for a future climate-change scenario. Results from these simulations are used in assessing the impacts of climate change on hydrology and air quality. To match the climate model grid spacing of approximately 300 km, we use 135, 45, and 15 km grids in MM5. The outer grid covers much of the Northeast Pacific and North America to encompass large-scale processes critical to Pacific Northwest climate. The intermediate grid encompasses the conterminous United States. The inner grid covers the states of Washington, Oregon, and Idaho. Initial and boundary conditions as well as interior nudging for MM5 runs are taken from the global climate model simulations. To perform long runs while maintaining stability and mass conservation of the simulation, we employ nudging (Newtonian relaxation) of the outer nest toward the global climate model simulation, thus the 135-km nest is constrained tightly to the global model simulation and yields a smooth transition from the global model to MM5. With this technique, we can run MM5 continuously for long periods, without periodic restarts. Nudging also preserves the large-scale state provided by the global model. Thus, the downscaling provides the regional meteorological details consistent with that large-scale state, which is assumed to be well resolved by the global model. The interaction of land-surface processes and mesoscale meteorological processes is a critical issue in regional climate modeling. We consider several issues and methods for initializing the land surface properties and applying a land-surface model. We discuss the advantages of performing a single continuous multi-year run versus several one-year runs in the context of the Pacific Northwest climate.
http://www.atmos.washington.edu/~salathe/reg_climate_mod/
A53A-0867 1340h
RSM transferability studies during CEOP
The main goal of this study is to evaluate the RSM simulated energy- and water budget by transferability studies. The predominance of either dynamical or physical processes varies in different regions of the globe. Therefore, regional models have been developed for different regions. As the regional simulation of the water- and energy cycle is very sensitive to how physical processes are represented within a regional model, transferability studies are a suitable approach to validate the performance of a regional model under different meteorological conditions. The RSM will be transferred to seven different domains all over the globe taken from the Continental Scale Experiments (CSE) of GEWEX. The data used for the validation are taken from the Coordinated Enhanced Observation Period (CEOP). Also, comparison with ISCCP data is carried out. The model runs are performed during CEOP, from 1 July 2001 until 31 December 2004. The RSM is run at 50 km horizontal resolution using NCEP reanalyses as initialization and boundary conditions. As equilibration of the land surface is necessary the runs begin July 1, 1999. Comparisons of the RSM simulations over the seven CSE domains with ISCCP data in July 1986 illustrated already the benefit of transferability studies: Evaluating the model performance under the different meteorological prerequisites of the various model domains showed that the two diagnostic cloud schemes used within RSM have different strengths connected with different dynamical and physical processes. This indicates that transferring the model to different domains leads to a more sophisticated validation than doing comparisons only over one domain.
A53A-0868 1340h
Impact of Precipitation Observations on Regional Climate Simulations
We seek the improvement of regional downscaling of large-scale analyses, aiming particularly at hydrological and energy budget consistency for improving regional climate analyses and forecasts. Currently available reanalyses (NCEP/NCAR Reanalysis, NCEP/DOE Reanalysis, ERA-15, ERA-40 and others) provide reasonably accurate analysis of atmospheric states, but the hydrological component of such analyses is weak, and the energy budget still suffers from a systematic tendency error that makes it difficult to close the budget equation. The weakest component of those reanalyses is the model-produced precipitation, which has very large errors compared to observations. For this reason, we intend to make the downscaled analysis suitable for regional forecast initial conditions and for consistent energy budget research by assimilating observed precipitation. In this study, we use a regional climate model to assimilate different precipitation data sets: (a) the .25 deg. National Oceanic and Atmospheric Administration's Climate Prediction Center (NOAA/CPC) daily precipitation analyses, disaggregated to hourly time scales from a coarser 2-deg. data set; (b) the new 1/2 hourly .25 deg NOAA/CPC MORPHed precipitation (CMORPH). To develop our study, we chose a large domain, which includes North and Central America. The sensitivity of the precipitation assimilation method to these precipitation data sets is being investigated and the results will be reported during the meeting.
A53A-0869 1340h
Role of soil water retention relations in a climate-crop-soil coupled regional model
Soil moisture prediction is of significance for agriculture, transportation, and other activities. By coupling models of regional climate, surface hydrology, and crop development, we have established a two-way interactive agroecosystem model that projects soil moisture and other hydrological variable variations in months in advance. One of the problems with this model, however, is its persistent over-drying soil that even sometimes prevents seeds from germination in a typical year. To solve this problem, we first ran a series sensitivity experiments diagnosing the error sources, determining whether the error comes from water input to soil (i.e., rainfall), soil property (e.g., bulk density, percentage triangle) or soil hydrology formulations. Of particular concern is the soil water retention curve used in the model. It is found that the soil water retention relation typically used in the atmospheric models tends to retain less water at a give water pressure than that typically used in soil science community. We will report the model improvement after adopting a new retention relation in the model by validating the model against in-situ and remote sensing soil moisture data.
A53A-0870 1340h
Nesting strategies in regional climate modeling
This paper will discuss the use of nesting strategies within the FSU regional spectral model. The FSU regional spectral model is embedded within the FSU global spectral model. Both models share the same physics and vertical structure making this modeling system ideal for boundary condition studies. The experiments consists of running the nested model for several months over northern South America. By varying the domain size and horizontal resolution we are able to examine the impact on the precipitation field. In addition an 'Acid Test' was conducted. Results will be presented.
A53A-0871 1340h
Dynamical Downscaling: Assessment of Value Retained and Added Using the Regional Atmospheric Modeling System (RAMS)
The value restored and added by dynamical downscaling is quantitatively evaluated by considering the spectral behavior of the Regional Atmospheric Modeling System (RAMS) in relation to its domain size and grid spacing. A regional climate model (RCM) simulation is compared with NCEP Reanalysis data regridded to the RAMS grid at each model analysis time for a set of six basic experiments. At large scales, RAMS underestimates atmospheric variability as determined by the column integrated kinetic energy and integrated moisture flux convergence. As the grid spacing increases or domain size increases, the underestimation of atmospheric variability at large scales worsens. The model simulated evolution of the kinetic energy relative to the reanalysis regridded kinetic energy exhibits a logarithmic decrease with time, which is more pronounced with larger grid spacing. Additional follow-on experiments confirm that the surface boundary forcing is the dominant factor in generating atmospheric variability for small-scale features and that it exerts greater control on the RCM solution as the influence of lateral boundary conditions diminish. The sensitivity to surface forcing is also influenced by the model parameterizations, as demonstrated by using a different convection scheme. For the particular case considered, dynamical downscaling with RAMS in RCM mode does not retain value of the large scale of that which exists in the larger global reanalysis. The utility of the RCM, or value added, is to resolve the smaller-scale features which have a greater dependence on the surface boundary. Additional evidence is shown which suggests this conclusion regarding RAMS is true for other RCMs as well.
A53A-0872 1340h
Versatility and Sensitivity Study of the PRECIS Regional Climate Modelling System - Simulations over North America using PRECIS
PRECIS, Providing REgional Climates for Impacts Studies, is a regional climate modelling system developed by the Hadley Centre. The PRECIS system is based on the latest RCM version developed by the Centre, it has been designed to be user-friendly and easily implemented on any fast PC with Linux system. Due to the versatility and capability of the system in providing high-resolution climate simulations over any area of the globe, users of PRECIS are gaining confidence quickly. More than a pedagogical and portable tool for teaching regional climate modelling, PRECIS have been tested over North America domains. The simulations of PRECIS were generated over North America with different domain sizes by applying lateral boundary conditions (dynamical atmospheric information) from ECMWF ReAnalysis data provided from 12/1978 to 05/1982. Monthly and seasonal means fields have been used for the different realizations. The author presents results of these experiments and provides comparisons between PRECIS, the driving GCM (ERA) and observational datasets (e.g. CRU).