H31A-0116
Remote Sensing of Open Water in Northern High Latitudes for use in Hydrologic Modeling
In the northern high latitudes open water bodies are common landscape features, having a large influence on hydrologic processes as well as surface-atmosphere carbon exchange and associated impacts on global climate. It is therefore of great importance to assess their spatial extent and temporal character in order to improve hydrologic and ecosystem process modeling. Spaceborne synthetic aperture radar (SAR) is an effective tool for this purpose since it is particularly sensitive to surface water and it can monitor large inaccessible areas on a temporal basis regardless of atmospheric conditions or solar illumination. We employ multi-temporal L-band SAR data from the Japanese Earth Remote Sensing Satellite (JERS-1) and ALOS PALSAR to map open water bodies across Alaska and Eurasia. A supervised decision tree-based classification approach was used to generate open water maps. For Alaska, we assembled regional-scale monthly JERS-1 SAR mosaics from data acquired during 1998. Digital elevation model (DEM) terrain and slope information were also employed in the decision tree classifier. These supplementary data aided significantly in improving classification performance in topographically complex regions where radar shadowing was prevalent. For study regions in Eurasia, PALSAR data was used in conjunction with JERS-1 imagery to map spatial patterns and seasonal variability in open water characteristics over selected study basins. These results were examined in relation to regional topographic and land cover characteristics. Classification results were also evaluated relative to other open water and land cover classification maps derived from Landsat, AVHRR, MODIS and SRTM. This work was carried out at the Jet Propulsion Laboratory, California Institute of Technology; at the University of Montana; at the University of Washington; and at Purdue University under contract with the National Aeronautics and Space Administration.
H31A-0117
River Runoff Estimates on the Basis of Satellite-Derived Surface Currents and Water Levels
One promising technique for river runoff estimates from space is the retrieval of surface currents on the basis of synthetic aperture radar along-track interferometry (ATI). The German satellite TerraSAR-X, which was launched in June 2007, permits current measurements by ATI in an experimental mode of operation. Based on numerical simulations, we present first findings of a research project in which the potential of satellite measurements of various parameters with different temporal and spatial sampling characteristics is evaluated and a dedicated data synthesis system for river discharge estimates is developed. We address the achievable accuracy and limitations of such estimates for different local flow conditions at selected test sites. High-resolution three- dimensional current fields in the Elbe river (Germany) from a numerical model of the German Federal Waterways Engineering and Research Institute (BAW) are used as reference data set and input for simulations of a variety of possible measuring and data interpretation strategies to be evaluated. For example, runoff estimates on the basis of measured surface current fields and river widths from TerraSAR-X and water levels from radar altimetry are simulated. Despite the simplicity of some of the applied methods, the results provide quite comprehensive pictures of the Elbe river runoff dynamics. Although the satellite-based river runoff estimates exhibit a lower accuracy in comparison to traditional gauge measurements, the proposed measuring strategies are quite promising for the monitoring of river discharge dynamics in regions where only sparse in-situ measurements are available. We discuss the applicability to a number of major rivers around the world.
H31A-0118
RivWidth: A Software Tool for the Calculation of River Width from Remotely Sensed Imagery
RivWidth is an implementation in ITTVIS IDL of a new algorithm that automates the calculation of river width using raster-based classifications of inundation extent derived from remotely sensed imagery. The algorithm utilizes techniques of boundary definition to define a river centerline, derives a line segment orthogonal to this line at each centerline pixel, and then computes total river width along each orthogonal. The output of RivWidth is comparable in quality to measurements derived using manual techniques, yet generates thousands of width values continuously along an entire stream course, even in multi-channel river systems. Uncertainty in RivWidth depends principally on the quality of the water classification used as an input, though pixel resolution and the values of input parameters play lesser roles.
H31A-0119
Evaluation of 1D numerical models in the prediction of distributed flow series constrained by radar altimetry data
Recently, efforts have been done forward the development of methodologies to obtain flow estimates from time series of water height at virtual stations (VS - intersections between satellite tracks and water surfaces) from Topex/Poseidon (TP) and ENVISAT radar mission measurements. Relatively simple 1D propagation models such as Muskingum-Cunge (M-C) Routing Model have presented good results in the water flow propagation in Amazonian rivers. These regions suffer, in most cases, of lack of information witch constrains more accurate hydrologic and hydraulic studies. Nevertheless, these M-C models make a lot of simplification witch may cause the lost of important information that these quasi-ungaged basins can offer. This paper presents a comparison between two propagation models to the estimation of water flow series constrained by T/P and ENVISAT data. The models are: ProGUM, a Muskingum-Cunge flow routing model with diffusion-cum-dynamic wave propagation, and HEC-RAS, a well-known 1D hydrodynamic model. The verification was made in four reaches in the Negro River Basin. Each of these reaches are limited by two gauge stations, one upstream and the other one downstream. A total of five VS distributed over the reaches were analyzed. Previous studies have shown that ProGUM may yield errors less then 10% in validation phases. Here, it is demonstrated that, no much improvements can be achieved by using a more complete model capable to absorb the available data and simplifications of a M-C Model do not make significant modifications in the results of rating curve generation from satellite altimetry.
H31A-0120
Quantifying the spatial scaling relationship for dominant hydraulic routing parameters in the Amazon Basin
The current state of terrestrial water cycle models may be adequate, particularly as spatial resolution is increased, for capturing the role of terrestrial processes in the circulation dynamics of global climate. However, as more specific questions are asked (e.g., what is role of the terrestrial water cycle in global biogeochemical cycling; what are the impacts of climate change on global flooding; how can we improve understanding and prediction of the magnitude, trend, timing, and partitioning of terrestrial water stores and fluxes), a large scale hydrologic framework that provides a realistic representation of fluvial flow and transport processes is needed. The challenge in modeling these fine scale processes is that they require parameterization and are highly scale dependent. In this research, we investigate the effects of spatial scale on horizontal conductivity and channel roughness in the Amazon Basin using a hydrologic model that simulates both water-balance and hydraulic transport for uplands, channels, and floodplains. The model uses an irregular computational grid designed for efficient parallelization. The model is parameterized, driven and assessed using six NASA systems: GRACE, JERS1, LANDSAT7, MODIS Terra, SRTM and TRMM. Model results are compared to point/hillslope measurements of conductivity and measured streamflow data. Results are presented for the Purus Basin, a tributary to the Amazon (>350,000 sq km), over a five year period (2001-2005). Preliminary results indicate that horizontal conductivity (cm/hr) increases approximately 2 to 3 orders of magnitude for a one order of magnitude increase in length scale (10 to 200 km). http://spatialhydro.sdsu.edu
H31A-0121
Using LandSat and SRTM datasets to develop relationships for estimating bankfull channel widths in the Amazon Basin
Modeling fine scale spatial and temporal processes of the hydrologic cycle over continental to global extents is vital for assessing the potential impacts of climate and land use change on global water resources and related systems. Significant advancement in understanding and predicting the magnitude, trend, timing and partitioning of terrestrial water stores and fluxes requires the development of methodologies and knowledge for extracting representative hydraulic geometries from remote sensing data products and field data, suitable for estimating inundation characteristics and water storage changes which are limited for much of the globe. In this research, relationships between channel and floodplain widths and spatial drainage characteristics are developed for the Amazon Basin. Channel and floodplain widths were measured using SRTM data and LandSat TM/ETM imagery at 510 sites. The study sites were selected based on the Pfafstetter decomposition methodology which provides an irregular model grid based on repeatedly subdividing landscape units into nine subunits consisting of basins and interbasins. The selected sites encompass all possible combinations of Pfafstetter modeling units (ex., basins of interbasins, interbasins of basins, etc.). The 510 study sites are within the Amazon Basin with drainage areas ranging 10 to 5.4 million sq km and mean watershed ground slopes ranging from 0.4 and 30 percent. Preliminary results indicate that channel widths can be predicted using drainage area and mean watershed slope (R2 = 0.85). Floodplain widths can be predicted using channel width and the local slope (R2 = 0.70). Using the Purus watershed, a sub-basin to the Amazon (350,000 sq km), effects of channel and floodplain widths on simulated hydrographs are presented. http://spatialhydro.sdsu.edu
H31A-0122
GRACE Satellite Data and In-Situ Water Level Time-Variation Correlation Analysis in the Amazon Basin.
This work is based on a comparison between daily in situ water level time series measured at hydrometric stations of A.N.A. ground-based compared with vertically-integrated water height deduced from GRACE geoids computed by GRGS at 10-day interval, converted into equivalent water heights, over the Amazon basin for a approx. 4-year period (July-2002 to May-2006). A high correlation, 80 per cent in most cases, between water heights from GRACE-based and in situ water levels over the Amazon basin is detected. This correlation allows to define a local transfer function by adjusting a linear relationship between GRACE-based and in situ observations time-series. The slope of the transfer function decreases from upstream to downstream along the rivers and a high correlation of these coefficients of proportionality versus the distance from estuary is detected. Finally, an error budget is made taking satellite errors, leakage effect and errors from truncation of spectrum into account.
H31A-0123
GHYRAF (Gravity and HYdrology in AFrica): a New Experiment Combining Hydrology and Geodesy to Investigate Water Storage Changes from the Sahara to the Equatorial Monsoon Zone.
We present a new project using multi-disciplinary data (gravity, geodesy, hydrology, and meteorology) to determine seasonal changes in water storage in Africa. We foresee to concentrate on two areas: the desert zone in the Sahara (Tamanrasset, Algeria) where almost no hydrological change is present, and the equatorial monsoon band (Niger, Benin Republic) that provides on the contrary a large rainfall signal. This project will be a first attempt to do a ground validation of satellite-derived gravity observations such as GRACE or GOCE in Africa. We will pay attention to the various length scales involved in hydrological processes that are differently retrieved whether gravity is measured at the ground or by satellite. Our experience includes two types of ground-based gravity measurements. First we will perform a repeated survey with absolute gravimeters (AG) on a North-South profile during 2-3 years (2008-2010) to assess the large soil moisture changes as predicted by existing hydrological models such as GLDAS or LadWorld. Second we plan to establish a superconducting gravimeter (SG) to act as a continuously monitored base station in a region of large soil moisture changes. In addition, continuous geodetic GPS measurements will be made along the gravity profile to assess the vertical deformation which acts to alter ground gravity but not satellite gravity. In-situ measurements of hydrological parameters at each station will assist us in modelling local gravity effects. The goal is a better characterization of continental water storage, particularly in the critical areas of water management in the north and central parts of the African continent.
H31A-0124
Multi-objective calibration of a global hydrology model using GRACE water storage variations
The satellite gravity mission GRACE (Gravity Recovery And Climate Experiment) was launched in 2002. Since then, GRACE provides global maps of time variations of the Earth's gravity field. These gravity variations are directly linked to mass redistributions on the Earth's surface. Thus, various geophysical and climatologically- driven processes can be observed. In particular, GRACE observations of large-scale water storage changes provide a comprehensive data set to analyze the global water cycle and to validate and calibrate hydrological models. In this contribution, we present the first results on the calibration of the WaterGAP Global Hydrology Model (WGHM) based on surface mass variations derived from the current RL04 GRACE-only gravity field model time series generated at GFZ Potsdam (GFZ-RL04). WGHM models the continental water storage including the most important water storage components, i.e., soil, snow, groundwater and surface water. The model is forced by climate data at a 0.5 degree resolution and hitherto calibrated against observed river runoff at 1235 gauging stations world wide. The station-based calibration of WGHM results in locally fitted discharge data, but model accuracy may decrease with distance from the calibration stations and for other water flux components or storage compartments. Previous studies of measured (by GRACE) and simulated (by WGHM) seasonal variations of total water storage in large river basins partially show significant differences, especially for the seasonal amplitudes. Combining both the present station-based accuracy of the model in terms of river discharge and the integrative nature of the GRACE data with global coverage, more realistic and improved simulation results are expected from a multi-objective calibration approach. This denotes the evaluation of model parameters through their simulation performance against more than one model output objective. In this contribution, these objectives are river discharge and total water storage change from GRACE. The Dynamically Dimension Search (DDS) calibration method was extended for a multi-objective problem and used to improve WGHM parameterization. Results of the multi-objective calibration of WGHM against GRACE data are presented and improvements compared to a single- objective approach are highlighted. Furthermore, calibrations against a reduced GRACE signal containing only the dominant (annual, semi-annual, or longer) signal components were tested. The results reveal the innovative contribution of the GRACE satellite mission to the field of hydrological modeling.
H31A-0125
Comparison of Terrestrial Water Storage Variations from GRACE With In-Situ Soil Moisture and Groundwater Level Measurements in Semiarid Irrigated Systems: Case Study High Plains Aquifer, USA
Depletion of groundwater storage in semiarid regions as a result of intensive irrigation is a critical water resource issue. Many of these systems are poorly monitored, such as the North China Plain and western India. The objective of this study was to evaluate the ability of the Gravity Recovery and Climate Experiment (GRACE) to quantify changes in groundwater storage using detailed monitoring records available for the High Plains aquifer (450,000 km2 area). This study presents a comparison of terrestrial water storage changes derived from GRACE gravity measurements between 2003 and 2006 with in-situ soil moisture and groundwater level measurements covering the High Plains aquifer. Soil moisture measurements from 80 shallow (~1 m depth) mesonet stations from Texas, Oklahoma, and Nebraska, were combined with data from deeper (up to 7 m) monitoring sites to estimate temporal and spatial variations in soil moisture over the High Plains. Anomalies in soil moisture were compared with soil moisture changes simulated by the Noah Land surface model. Groundwater storage variations over the aquifer were estimated by assimilating groundwater level measurements from multiple state and federal agencies. Good correspondence between soil moisture storage from the ground based networks and the Noah land surface model increased confidence in the soil moisture storage variations. Terrestrial water storage (TWS) changes from GRACE compared favorably with TWS (approximated as changes in soil moisture + groundwater storage) from the monitoring networks. Results from this study demonstrate the potential for the GRACE satellites to monitor water storage variations in semiarid irrigated systems, where mining of groundwater resources is a critical issue.
H31A-0126
Evaluations of global land-to-oceans fresh water discharge and evapotranspiration
We estimate global fresh water discharge from land to oceans (R) and evapotranspiration (ET) on monthly time scales using a number of complimentary hydrologic data sets. This estimate is possible due to the new capability of measuring oceanic and land water mass changes (dSo/dt,dSl/dt) from GRACE as well as the space-based measurements of oceanic and land precipitation (Po,Pl) and oceanic evaporation (Eo). R can be estimated through the water balance equation over oceans, R=Eo-Po-dSo/dt, and then ET is estimated based on water mass balance over land, ET=Pl-R-dSl/dt. Monthly time series of R show peaks in July and December, and those of ET shows peaks in March and August. In general, our estimates of R and ET are correlated with Pl well indicating qualitatively that our estimates capture temporal patterns of R and ET reasonably. We compare our estimation for R and ET to 20th century simulations from the WCRP CMIP3 multi-model archive that are assessed in the IPCC 4th Assessment Report. R and ET from AOGCMs mainly show annual cycles, but those estimated in this study exhibit additional semi-annual changes. The estimates of R and ET examined in this study are potentially useful to constraint current AOGCMs models.
H31A-0127
Near Real Time Evapotranspiration Estimation Using Remote Sensing Data
Satellite remote sensing is a promising tool to estimate the spatial distribution of evapotranspiration (ET) at regional or global scales with minimal use of in situ observational data. A spatial mapping of the solar energy partition is the central challenge for ET estimation from satellite remote sensing data. We developed a near real time ET estimation system using MODIS data products. A "VI-Ts" (vegetation index and the surface radiant temperature) diagram is used to estimate soil temperature and air temperature. Downward short-wave radiation and cloud cover data are taken from GEWEX Continental Scale International Project (GCIP) and GEWEX Americas Prediction Project (GAPP) Surface Radiation Budget (SRB) Data. Other radiation components are calculated from remote sensing albedo, emissivity, and temperatures. The landscape is assumed to be a mixture of vegetation and bare soil. Therefore, a simple two-source model of ET is used. A canopy conductance model is used to describe vegetation physiology and to estimate vegetation ET. The temperatures are used to estimate bare soil evaporation. The near real time system is primarily driven by remote sensing data, yet is flexible enough to incorporate meteorological data when they are available. The near real time ET estimation system, which is housed at the University of Washington (UW) and runs in connection with the UW westwide seasonal streamflow prediction system, updates daily for a domain spanning the entire western United States. http://www.hydro.washington.edu/forecast/rset_ca/
H31A-0128
Satellite Based Mapping of Land Surface ET using MODIS and Alternate Surface Meteorological Inputs from AMSR-E, Reanalysis, and Surface Weather Stations
Regional evapotranspiration (ET), including water loss from plant transpiration and soil evaporation, is essential to understanding interactions between land-atmosphere surface energy and water balances. Vapor pressure deficit (VPD) and surface air temperature are key variables for stomatal conductance and ET estimation. We developed an algorithm to estimate ET using a modified Penman-Monteith approach driven by MODIS derived vegetation data and daily surface meteorological inputs including net incoming solar radiation, air temperature and VPD. The model was applied using alternate daily meteorological inputs, including: 1) site level weather station observations, 2) VPD and air temperature derived from the Advanced Microwave Scanning Radiometer (AMSR-E) on the EOS Aqua satellite, and 3) Global Modeling and Assimilation Office (GMAO) reanalysis based surface temperature, humidity and solar radiation data. Model performance was assessed across a North American boreal-Arctic transect (>50o N) of six eddy covariance flux towers representing boreal grassland, boreal forest and tundra biomes. Model results derived from the three meteorology data sets agree well with observed tower fluxes (r>0.6; P<0.00001; RMSE<30W/m2) and capture spatial patterns and seasonal variability in ET. The MODIS-AMSR-E derived ET results also show comparable accuracy to ET results derived from the reanalysis meteorology, while ET estimation error was generally more a function of algorithm parameterization than differences in meteorology drivers. Our results indicate significant potential for regional mapping and monitoring daily land surface evaporation using synergistic information from satellite optical-IR and microwave remote sensing.
H31A-0129
Satellite Estimation of Daily Land Surface Water Vapor Pressure Deficit from AMSR- E
Vapor pressure deficit (VPD) is a key variable for monitoring land surface water and energy exchanges, and estimating plant water stress. Multi-frequency day/night brightness temperatures from the Advanced Microwave Scanning Radiometer on EOS Aqua (AMSR-E) were used to estimate daily minimum and average near surface (2 m) air temperatures across a North American boreal-Arctic transect. A simple method for determining daily mean VPD (Pa) from AMSR-E air temperature retrievals was developed and validated against observations across a regional network of eight study sites ranging from boreal grassland and forest to arctic tundra. The method assumes that the dew point and minimum daily air temperatures tend to equilibrate in areas with low night time temperatures and relatively moist conditions. This assumption was tested by comparing the VPD algorithm results derived from site daily temperature observations against results derived from AMSR-E retrieved temperatures alone. An error analysis was conducted to determine the amount of error introduced in VPD estimates given known levels of error in satellite retrieved temperatures. Results indicate that the assumption generally holds for the high latitude study sites except for arid locations in mid-summer. VPD estimates using the method with AMSR-E retrieved temperatures compare favorably with site observations. The method can be applied to land surface temperature retrievals from any sensor with day and night surface or near-surface thermal measurements and shows potential for inferring near-surface wetness conditions where dense vegetation may hinder surface soil moisture retrievals from low-frequency microwave sensors. This work was carried out at The University of Montana, at San Diego State University, and at the Jet Propulsion Laboratory, California Institute of Technology, under contract to the National Aeronautics and Space Administration.
H31A-0130
Integrating Remotely Sensed Data for Water Budget Studies in the San Luis Basin, Colorado
Water budget analyses are frequently used to examine long-term trends in the freshwater cycle at the regional scale. We have investigated the water balance in the San Luis Basin, CO for the years 2001-2006 by integrating data from a combination of direct measurements, satellite observations, and land surface models. The water budget was calculated using a basic mass-balance approach. Average monthly precipitation values were derived from measurements taken by the Tropical Rainfall Measuring Mission (TRMM). The monthly flux of surface water runoff was calculated using USGS stream gages along the Rio Grande. Monthly changes in groundwater storage, determined from the Rio Grande Water Conservation District well network, and changes in soil moisture storage, derived from the Global Land Data Assimilation System (GLDAS) soil moisture model, were also considered. Monthly values of Evapotranspiration (ET) were calculated for the time period using latent heat flux data from the North American Data Assimilation System (NLDAS) 1997-2000 model record. A relationship between calculated ET (1997-2000) and direct measurements of average monthly temperatures for the same time period in Alamosa, CO was then derived and used to calculate representative ET values for the period 2000-2006. The final water budget for the San Luis Basin is balanced within 10% when examined over the entire study period. The main limitations of our approach include the coarse resolution of the satellite data used in this analysis, TRMM's limited sensitivity to snowfall, and uncertainties in the model used to calculate ET. Nonetheless, this approach of integrating freshwater observations from a variety of data sources and scales shows promise for improving future water budget analyses.
H31A-0131
Estimates of Water Use of Saltcedar (Tamarix ramosissima) on the Lower Colorado River: from Plant to Stand to River Reach
Saltcedar (Tamarix ramosissima) removal projects have been proposed to salvage water that would other wise support saltcedar evapotranspiration (ET), and to allow native vegetation to recolonize western U.S. riparian corridors. We measured stem-level sap flow at Cibola NWR on the Lower Colorado River to answer some of the scientific questions about the possible consequences of saltcedar removal. We then conducted wide-area studies using remote sensing technology by scaling from the three ground sites using TM- and MODIS-based ET estimates. The sites were different distances from the river channel and differed in depth to water table and salinity of the ground water. Results were then extrapolated to the river reach (from Davis Dam to the delta of the river in Mexico). Saltcedar stands at Cibola had moderate rates of ET, based on remote sensing estimates, averaging 1.1 m yr-1, similar to rates determined for other locations on the river and for other river systems. Leaf area index (LAI) values were also moderate, and stands were relatively open, with areas of bare soil interspersed within stands. Despite high ground water salinity (5,000-10,000 mg l-1), the sites away from the river did not have saline surface soils, supporting studies showing that saltcedar does not salinize riverbanks. Approximately 1 percent of the mean annual river flow is lost to saltcedar ET on the Lower Colorado River in the U.S. Based on these results, the opportunities for water salvage through saltcedar removal appear to be constrained by its modest ET rates.
H31A-0132
NASA Hydrology Data and Information Services Center (HDISC) Supports Global Land Data Assimilation System (GLDAS) Products
The Global Land Data Assimilation System (GLDAS) is generating a series of land surface state (e.g., soil moisture and surface temperature) and flux (e.g., evaporation and sensible heat flux) products simulated by four land surface models (CLM, Mosaic, Noah and VIC). These products are now accessible at the Hydrology Data and Information Services Center (HDISC), a component of NASA Goddard Earth Sciences Data and Information Services Center (GES DISC). In addition to the basic anonymous data downloading, HDISC provides several advanced data search and downloading services, such as Mirador, OPeNDAP, and Giovanni. Mirador is a Google-based search tool that provides discovery of and access to data based on keywords. Mirador also provides users the capability to perform on-the-fly spatial and parameter subsetting and temporal aggregation of selected data. OPeNDAP (Open-source Project for a Network Data Access Protocol) enables remote OPeNDAP clients to access OPeNDAP served data regardless of local storage format. Giovanni is an online visualization and analysis tool that provides a simple way to visualize, analyze, and access vast amounts of data without having to download the data. With Giovanni, users can perform spatial and temporal subsetting and explore spatial and temporal correlation of various parameters. Current GLDAS data hosted at HDISC include a set of 1.0 degree resolution data products, covering 1979 to the present, from the four models and a 0.25 degree data product, covering 2000 to the present, from the Noah model. The HDISC has the capability to support more hydrology data products and more advanced analysis tools. The goal is to develop HDISC as a data and services portal that supports weather and climate forecast, and water and energy cycle research.
H31A-0133
Factors Affecting Remotely Sensed MODIS Snow Cover Area Uncertainty
Accurate knowledge of snow pack properties is important for short-term weather forecasts, climate change prediction, and hydrologic forecasting. As both the model predictions and passive microwave SWE observations contain large errors due to land surface complexities and temporally frequent snowmelt processes in the western United States, the 500m daily MODIS snow cover area (SCA) product has been widely used as an important constraint on snowpack processes in land surface and hydrological models. Snow pack is an integrated response to both climate and land surface complexity. Understanding and quantifying MODIS snow cover retrieval errors are critical for successful utilization of the MODIS SCA product. This study has made a thorough uncertainty assessment of the remotely-sensed MODIS snow cover product. We consider the MODIS SCA retrieval errors associate with climate and different land surface characteristics. For the first time, we demonstrate a relationship between the MODIS SCA retrieval errors and temperature. We extend this work by seeking relationships between the parameters of the error model and known sources of SCA error, such as landcover (e.g. forest versus grassland), forest fraction, and topographic roughness.
H31A-0134
Estimates of total downwelling surface radiation using a high-resolution GOES-based cloud product along with MODIS and AIRS products
Total downwelling radiation, along with precipitation, is the primary forcing of land surface processes. Clouds are a first-order modulator in radiation processes as they attenuate solar insolation while simultaneously emitting longwave radiation and therefore are key to capturing the space-time variability in downwelling surface radiation. In this study, a cloud-coupled solar insolation and longwave radiation model is developed using readily available satellite-based measurements of land surface and atmospheric states. The downwelling radiation model uses inputs from the Moderate Resolution Spectroradiometer (MODIS) and Atmospheric Infrared Sounder (AIRS) sensors while cloud coupling is achieved via use of the NASA Visible Infrared Solar-Infrared Split Window Technique (VISST) cloud product. Use of the 4-km VISST product effectively downscales larger scale satellite- derived products (e.g. from AIRS) while at the same time provides finer-scale temporal estimates of atmospheric conditions (i.e., every 30 minutes instead of twice daily and four times daily for AIRS and MODIS measurements, respectively). Temporal gaps in AIRS and MODIS products are further refined by scaling a diurnal climatology look-up table derived from ground-based measurements collected by the Oklahoma Mesonet (OKMESONET) for both clear-sky and cloudy-sky conditions. Error characterization studies comparing satellite-derived estimates of surface states/fluxes against ground-based measurements from the Atmospheric Radiation Measurement Program (ARM) and OKMESONET have been conducted. Analysis of the results focuses on identifying key errors and uncertainties that can be accounted for using an ensemble approach for eventual use in a data assimilation framework. The resulting coupled downwelling radiation model provides estimates at high resolution (4-km spatial resolution at 30 minute intervals) and results from an application in the Southern Great Plains (SGP) region of the United States are presented.
H31A-0135
Exploring possibility of using the brightness temperature difference between water vapor and thermal channels to estimate monsoon precipitation
The infrared measurements from sensors onboard geostationary satellites are often used to quantify and estimate precipitation, which is one of the most important components of the water cycle. In recent years, emphasis is given on the synergetic use of different satellite sensors by exploiting their advantages and constraining their limitations to better quantify precipitation. We assess plausibility of using the brightness temperature difference between water vapor and thermal infrared channels (WVTD) of MVIRI sensor onboard METEOSAT-5 to estimate monsoon precipitation over the Indian continent. The data for the JJAS of 2003 are used in the present study. We compare our estimates with three different datasets namely, raingauge station data from India Meteorological Department, GPCP 1-Degree Daily (1DD), and TRMM 3B42. The comparisons will be discussed for eleven small study areas which are selected in such a way that they exhibit different geographical and rainfall conditions. We show that WVTD could be very useful in quantifying the convective precipitation and to monitor deep convection. A very good agreement is found among estimates based on WVTD and other three independant datasets. This difference could also serve as an additional constraint while preparing precipitation climatology by merging information obtained from different satellite sensors.