H31D-0628
A Flux Tower Instrument Intercomparison in Support of the DOE CLASIC Field Intensive Campaign
The DOE Cloud and Atmosphere Land Surface Interaction Campaign or CLASIC experiment was designed to further better understanding of the effects of the land surface on cumulus convection. The intensive observation period (IOP) of this project was conducted in the summer of 2007 and was distributed throughout much of the ARM-Climate Research Facility, Southern Great Plains region in Oklahoma. The experiment utilized a range of measurement platforms. From satellites and high-altitude aircraft, to surface measurements. At the surface level, ten eddy covariance flux towers were distributed at three geographically separated sites. These towers measured fluxes of carbon, water, and energy, as well as components of the radiation budget. The data set generated by these towers will not only be used in local process-based investigations, but will also be used as calibration points for some of the aircraft measurements, and will server as the basis of regional up- scaling efforts. To insure consistency of synthesis products derived from this flux tower data set, an instrument intercomparison was undertaken prior to field deployment. We report here, the results of this comparison for 14 different parameters among the participating flux tower groups.
H31D-0629
Measurements of Boundary Layer Structure at Fort Cobb During CLASIC, June 2007
A tethersonde system was deployed at Fort Cobb, Oklahoma during the Cloud and Land Surface Interaction Campaign (CLASIC) June 8-24 2007 with the objective of characterizing the diurnal cycle of lower boundary layer structure up to 500 m including wind, pressure, temperature, humidity as well as CO2 profiles over harvested wheat. One unique feature of this data set is that includes fair weather, pre-storm and post-storm conditions for a record monthly rainfall in Oklahoma, in excess of 300 mm at the site. Here, we discuss specifically the diurnal cycle of (potential temperature) and q (specific humidity) and overall boundary layer structure during the duration of the field campaign with an emphasis on conditions before and after one major rain event. Preliminary regional estimates of surface roughness and friction velocity, and sensible heat flux and latent heat flux are also presented.
H31D-0630
Vegetation Water Content Retrievals for NAFE06 and CLASIC
Vegetation water content (VWC) is a valuable input to many microwave based soil moisture retrieval algorithms. Previous research, both theoretical and experimental, has established that VWC can be estimated using multispectral remote sensing. There are limits on the reliability of these methods that are related to canopy characteristics. For some crops index based techniques typically saturate before peak VWC occurs. Therefore, a careful evaluation using validation data is useful, especially in an intensive field campaign. As part of NAFE06 and CLASIC, ground based VWC sampling was conducted in conjunction with surface reflectance and Leaf Area Index (LAI) measurements. These data were used to establish crop/cover condition dependent relationships between the Normalized Difference Water Index (NDWI) and VWC. In order to apply these relationships over larger regions it was also necessary to develop land cover classifications using satellite data that incorporated the temporal changes that occurred during the study (i.e. flooding of rice fields). This was accomplished using multiple Landsat Thematic Mapper or ASTER images and a decision tree classifier. The NDWI relationships were applied to the available vegetation index images to generate VWC.
H31D-0631
Soil Moisture Retrieval During a Corn Growth Cycle Using L\-Band \1.6 GHz\) Radar Observations
This paper reports on the retrieval of soil moisture from dual-polarized L-band (1.6 GHz) radar observations acquired at view angles of 15, 35 and 55 degrees collected during a field campaign covering a corn growth cycle of 2002. The applied soil moisture retrieval algorithm includes a surface roughness and vegetation correction, and could potentially be implemented as an operational global soil moisture retrieval algorithm. The surface roughness parameterization is obtained through inversion of the Integral Equation Method (IEM) from dual- polarized (HH and VV) radar observations acquired under nearly bare soil conditions. The vegetation correction is based on the relationship found between the ratio of model bare soil scattering contribution and observed backscatter coefficient (ósoil/óobs) and vegetation water content (W). Validation of the retrieval algorithm against ground measurements shows that the top-5cm soil moisture can be estimated with an accuracy of up to 0.033 cm3cm-3.
H31D-0632
Design of soil moisture observatories for remote sensing calibration and validation
Knowledge of the soil moisture state is critical to our understanding of the global water and energy cycles yet routine observation is hindered by the cost of establishing appropriate sampling arrays and the high spatial variability of the soil moisture state. Passive microwave sensors aboard existing and proposed satellite platforms offer the best solution for estimating the surface soil moisture content, which can then be assimilated into land surface parameterization schemes for estimates of root zone soil moisture. One promising satellite mission planned for launch in early 2008 is the Soil Moisture and Ocean Salinity Mission (SMOS). To date, however, very few ground-based networks have been established for ongoing validation and calibration of the satellite observations. To establish appropriate soil moisture networks it is important, from a financial standpoint, that we can minimize the costs associated with validation/calibration by determining the minimum number of sites required to characterize the mean soil moisture of a satellite pixel. Using data from the SMEX 2002, SMEX 2003, and NAFE 2005 field experiments we have estimated the minimum sampling density necessary at the satellite pixel scale to characterize the mean and have used this information to establish two in situ soil moisture monitoring networks in Ontario and Saskatchewan Canada. Details of these new soil moisture observatories will also be presented.
H31D-0633
Assessment of Potential AMSR-E Soil Moisture Disaggregation Using Scatterometer Observations
Advanced Microwave Scanning Radiometer (AMSR-E) on the NASA's Aqua platform has been providing land surface variables such as soil moisture in near real-time since 2002. A fundamental ongoing issue with the satellite estimates, as with any satellite passive microwave sensor, is their coarse spatial scale and how to downscale them to spatial resolutions compatible with a wider range of applications. Disaggregation techniques based on a synergism between passive and active microwave observations have shown promising results. However the available radar systems have limited temporal and spatial coverage in the mid-latitudes where soil moisture is important. Most of the disaggregation methodologies are based on temporal change detection in soil moisture. Another alternative is the QuikSCAT scatterometer, which offers daily observations with a 2.225km ground pixel size for the enhanced backscattering coefficient product. This may be a desirable option that offers a long-term data set with high temporal resolution for developing downscaling technique for disaggregation of radiometer derived soil moisture estimates (i.e. AMSR-E). QuikSCAT backscatter sensitivity to soil moisture was studied over the National Airborne Field Experiment 2006 (NAFE"06) area located in the south-eastern part of Australia. The domain encompasses a wide range of ground conditions including flood irrigation. Point comparisons between QuikSCAT backscattering coefficients and AMSR-E soil moisture revealed the greatest sensitivity of QuikSCAT backscatter to soil moisture over the Kyeamba study area, which was mostly grazing land. North-south and east-west oriented transect lines for a variety soil moisture conditions were also examined. Overall, the QuikSCAT backscatter and AMSR-E soil moisture show similar trends. However the larger variability of the backscatter values was evident in more of the transect lines. As a result, further analysis of the impact of NDVI and vegetation type is needed. The proposed assimilation of radiometer and scatterometer obtained observations will result into high temporal and fine spatial resolution soil moisture product. Aquarius, due for launch in 2009, carries on board both instruments. On that way exploring the possibility of combining AMSR-E and QuikSCAT observations can be beneficial for Aquarius by building more knowledge on the soil moisture temporal and spatial variability and by improving the available soil moisture and disaggregation algorithms.
H31D-0634
Leaf wetness distributions in a heterogeneous agricultural landscape
Spatial variability of leaf wetness quantity is a rising concern for remote sensing and hydrology. The presence of liquid water on the plant surface may impact the ability of new and future remote sensing technologies to measure surface soil moisture. Furthermore, the potential recharge of surface soil moisture from leaf wetness is small but critical element of the water balance, especially in dry environments. Measuring the variability and spatial extent of leaf wetness events will provide an upper limit for modeling and remote sensing in determine the effect of such events on hydrologic studies. In coordination with the SMEX05 experiment, leaf wetness sensors were deployed and measurements collected during June of 2005 in and around the Walnut Creek Watershed near Ames, Iowa. Column density estimates of leaf wetness were calculated hourly for each day of record for the study region at 20 different fields. These data were combined with a vegetation leaf area index map to produce a spatial leaf wetness product daily during the experiment.
H31D-0635
Effect of land cover classification map resolution in land surface modeling studies
Evapotranspiration, soil moisture and surface temperature are the key variables controlling the partitioning of energy and water fluxes at the land surface. Over the last two decades, satellite data have been used to constrain these variables in land surface models using data assimilation and parameter calibration methods. For example, the Noah land surface model uses satellite-based land cover information to estimate vegetation parameters and improve its predictions. For simplicity, coarse resolution land cover maps are often used across a range of model resolutions. This study focused on the effect of using coarse spatial resolution land cover classification (LCC) remote- sensing images in land surface models. It was hypothesized that a scaling-threshold exists where high resolution variability is averaged out that can result in significant model prediction errors. To test this hypothesis, a 1 km resolution University of Maryland LCC map was rescaled into several coarser resolution images. Each LCC map was used in a separate Noah model run within the Land Information System modeling framework to estimate evapotranspiration, bare soil evaporation, soil moisture and upper layer soil temperature. Model results were compared using three methods: (1) observing the direct differences; (2) decomposition of the differences into long term, cyclic and random components; and (3) quantifying the overall root mean square error. The direct difference method revealed that increasing biases were introduced as LCC maps were rescaled to coarser resolutions. The decomposition method showed that the magnitude and the sign of the seasonal and annual variations of time series differences are linked to climatological variables like temperature and precipitation. Both methods also showed that land cover sub-pixel variations have strong effects on the direction of the bias introduced by increased heterogeneity. Root mean square error method showed rescaling to coarser resolution increased the overall bias. The hypothesized scaling-threshold was not found. However all methods showed the resulting biases increases as the LCC map rescaling factors increased.
H31D-0636
Parameterization of the spatial varaibility of cold season processes in the Midwestern United States
The spatial distribution of frozen soil and snow cover at the start of the spring melt season plays an important role in the generation of spring runoff and in the exchange of energy between the land surface and the atmosphere. Infiltration into frozen soils is underestimated by models that simulate uniformly frozen soil. While variability in snow cover, whether from drifting and differences in melt caused by topography and local shading, can lead to a mixture of snow covered and snow free areas during melt. Snow free areas with their lower albedos warm faster in the spring and can contribute to faster melt through the advection of warmer air to neighboring snow covered areas. Algorithms for the representation of spatial variability in snow and soil frost exist in the Variable Infiltration Capacity (VIC) model, but parameterization has been limited to a single set of observations from central Minnesota. Using field observations, remote sensing and NSIDC snow cover products spatially distributed parameterizations have been developed for the north central United States based on three statistical distributions: uniform, three parameter gamma and three parameter log-normal distribution. These parameterizations are used within the VIC model to study the effect of the selection of distribution type on the performance of the snow and soil frost algorithms.
H31D-0637
Indirect Measurement of Evapotranspiration from Soil Moisture Depletion
Direct and in situ measurement of evapotranspiration (ET), such as the eddy covariance (EC) method, is often expensive and complicated, especially over tall canopy. In view of soil water balance, depletion of soil moisture can be attributed to canopy ET when horizontal soil moisture movement is negligible and percolation ceases. This study computed the daily soil moisture depletion at the Lien-Hua-Chih (LHC) station (23o55'52"N, 120o53'39"E, 773 m elevation) from July, 2004 to June, 2007 to estimate daily ET. The station is inside an experimental watershed of a natural evergreen forest and the canopy height is about 17 m. Rainfall days are assumed to be no ET. For those days with high soil moisture content, normally 2 to 3 days after significant rainfall input, ET is estimated by potential ET. Soil moistures were measured by capacitance probes at -10 cm, - 30 cm, -50 cm, -70 cm, and -90 cm. A soil heat flux plate was placed at -5 cm. In the summer of 2006, a 22 m tall observation tower was constructed. Temperature and relative humidity sensors were placed every 5 m from ground surface to 20 m for inner and above canopy measurements. Net radiation and wind speed/directions were also installed. A drainage gauge was installed at -50 cm to collect infiltrated water. Continuous measurements of low response instruments were recorded every 30-minute averaged from 10-minute samplings. A nearby weather station provides daily pan evaporation and precipitation data. Since the response of soil water variations is relatively slow to the fluctuations of atmospheric forcing, only daily ET is estimated from daily soil moisture depletion. The annual average precipitation is 2902 mm and the annual average ET is 700 mm. The seasonal ET patterns of the first two water years are similar. The third year has a higher ET because soil moisture was recharged frequently by rainfall In order to examine the applicability of this approach, an EC system, including a 3-D sonic anemometer (Young 81000) and a krypton hygrometer (Campbell KH20), were practiced for LH and SH measurements at the top of the tower. During wet seasons (summer and autumn), fogs and afternoon thunderstorms often caused failures of the EC system. The ETs directly measured by the EC are slightly higher than those estimated from soil moisture depletions. Spatial soil moisture heterogeneity is suspected to be the cause of such discrepancies.
H31D-0638
Development of a Remotely-sensed Soil Heat Flux Parameterization for Natural Landscapes in Semi-arid Regions
Due to the difficulties in directly measuring soil heat flux (G), research on empirical estimation has moved toward use of a strong association between soil heat flux and net radiation (Rnet). The majority of these studies are concentrated on the estimation of soil heat flux from mature agricultural areas in semi-arid regions due to the high demand for irrigation water. However, natural land surfaces, the largest fraction of semi-arid and arid regions, have not been well studied with regards to soil heat flux estimation. Therefore, application of the previously developed empirical equations to natural land surfaces results in large uncertainty in soil heat flux estimates. This study explores development of an empirical relationship that is well-suited for natural landscapes within semi-arid areas in order to provide a more thorough assessment of regional evaporation (i.e. water consumption) in water-limited regions. Particularly, we seek to develop an empirical relationship between soil heat flux and net radiation when observations from the mid-day polar orbiting satellites (i.e. Terra/Aqua) are available. MODIS-derived components such as vegetation indices, albedo and surface temperature are being used to characterize this relationship over a set of flux tower sites in southern Arizona. Evaluation of existing soil heat flux schemes as well as results from validation of a new formulation suitable for use in natural landscapes within semi-arid regions will be presented.
H31D-0639
Intercomparison of remote sensing-based evapotranspiration models using SGP and SMEX data
Accurate characterization of evapotranspiration (ET) over a range of spatial and temporal scales is critical for many applications in hydrology, ecohydrology, meteorology, climatology, and agriculture. Over the past several years, there has been a major effort devoted to the development and refinement of remote sensing-based energy balance models that provide spatially-distributed ET maps operationally using satellite data. Validation of the product (ET maps) is typically performed using a handful of tower-based flux observations, and hence little is known about the reliability of the ET maps for the majority of the scene. Very few studies have attempted to inter- compare ET models over the same experimental site in order to quantify and gain greater insight as to the possible uncertainty in ET estimation using different modeling approaches over the same landscape/region. In this study, we compare several remote sensing-based energy balance/ET modeling schemes, which have operational capabilities using remote sensing, with imagery and ground-truth data from the 1997 Southern Great Plains (SGP) experiment and the 2002 Soil Moisture/ Atmosphere Coupling EXperiment (SMEX02/SMACEX). The models differ in the complexity of the algorithms used in computing energy flux exchange, estimating model parameters/variables, and ancillary data requirements. However, all modeling approaches require surface temperature, vegetation cover and meteorological inputs. In this initial inter-comparison we will investigate if model differences are significant and can be associated with land cover or other landscape features, procedures used in defining model inputs or other factors. We will also compare model output with flux tower observations and contrast difference statistics produced between the various models and the measurements and between the different models. This type of investigation may ultimately lead to improvements in the algorithms used by the various models and/or provide an opportunity for incorporating the strengths of the different approaches in the development of a hybrid remote sensing ET model with significantly greater utility.
H31D-0640
Estimation of Global Ground Heat Flux
This study investigates the feasibility of a previously published algorithm for estimating global ground heat flux (GHF). The proposed method is based on an analytical solution of the diffusion equation for heat transfer in a soil layer that has been shown to be effective at local scales. The algorithm has several advantageous properties: (1) single-level input of surface (skin) temperature, (2) time-mean GHF derived directly from time-mean temperature input, (3) reduced sensitivity to the variability in soil thermal properties and moisture, (4) insensitivity to snow depth for GHF over snow cover, and (5) computationally effective. NECP Reanalysis data is used to obtain the thermal inertia parameter needed as a function of soil type globally. These parameter estimates are comparable to values obtained from in- situ observations. GHF is then estimated globally. Results are generally consistent with reanalysis products that use two-layer soil hydrology models to obtain GHF. Where they differ, we argue that the new algorithm is more robust and trustworthy. The procedure offers potential benefits for direct assimilation of surface temperature into the reanalysis models at various time scales.
H31D-0641
Reanalysis of Data from River Discharge and Gauge Height from the Amazon Basin
The Amazon is the world's largest, discharging more water to the ocean than any other river. Study of the world's fresh water sources becomes more significant with increasing awareness of global climate change and its potential affect on those resources. In this paper, we present a data reanalysis of the daily discharge and the respective gauge height for 87 active gauge stations throughout the Amazon River Basin. The data set was originally obtained from the ANEEL (Brazilian Electricity Regulatory Agency). The reanalysis consists of normalizing the decimal notation, filtering inconsistencies, and filling in missing data by an averaging procedure. These three problems proved to be nontrivial ones and prevented full benefit from the data. Having the reanalyzed data available will help improve understanding of the spatio-temporal variations of the water budget component of the Amazon basin, corresponding to the fundamental and difficult problem of modeling basin and continental scale hydrologic routing models.
H31D-0642
Mountain Meadows and their contribution to Sierra Nevada Water Resources
Human alterations of California's waterscape have exploited rivers, wetlands and meadows of the Sierra Nevada. A century of intensive logging, mining, railroad building, development, fire suppression, and grazing by sheep and cattle has left only 25 percent "intact" natural habitat in the Sierra Nevada (SNEP 1995). Much of this intact habitat occurs at higher elevations, often in non-forested alpine or in less productive forests and woodlands where mountain meadows exist. Mountain meadows serve many ecological functions including habitat for threatened and endangered terrestrial and aquatic species, and are considered to be essential physical components to watershed function and hydrology with significant water storage, filtration and flood attenuation properties. This study evaluates the physical characteristics and hydrologic function of Clarks Meadow located in northern Sierra Nevada, Plumas County, California. In 2001, Clarks Meadow received significant restoration work in the upstream half of the meadow which diverted the stream from an incised channel to a shallow remnant channel, creating a stable channel and reconnecting the groundwater table to the stream. No restoration work was done in the lower half of Clarks Meadow where the stream still flows through an incised channel. Clarks Meadow offers a unique opportunity to study both a restored, hydrologically functional meadow and an incised, hydrologically disconnected stretch of the same stream and meadow. The physical characteristics of Clarks Meadows that were measured include surface area, subsurface thickness, porosity and permeability of subsurface materials, potential water storage volume, and surface infiltration rates. The goal of this study is to refine hydrologic characterization methods, quantify water storage potential of a healthy, non-incised meadow and assess its role in attenuating flood flows during high discharge times. Initial results suggest that significant subsurface storage volume is available in the meadow. Incising conditions in the unstable lower channel tends to dewater the lower portion of the meadow which encourages bank erosion through piping and corrasion. This study addresses questions that have broad implications for water management throughout the state because much of California gets water from Sierra high elevation watersheds in which meadows are thought to play a critical role in sustained long-term hydrologic function. The results of this study will be used to inform Integrated Regional Water Management Plans throughout Northern California.
H31D-0643
Modeling Stream/Aquifer Interactions on a Reach of the Upper San Pedro River Basin, AZ, by Integrating KINEROS and MODFLOW
In a groundwater model, the exchange with the surface water system is commonly simulated with boundary conditions. By the same token, in physically based rainfall-runoff models, the interaction with the groundwater system is represented in a simplified manner. Both approaches are defendable in many practical applications. However, in semiarid regions, stream/aquifer interactions play a critical role and a more integrated approach is advisable. Integrating two models commonly used in scientific and engineering applications is an approach successfully pursued by several researchers. In turn, that approach is the main concern of this short communication. A semi- automatic methodology that allows simulating stream/aquifer interactions combining two public domain codes, MODFLOW for groundwater flow, and KINEROS, for surface flow, is then presented. The feasibility of the approach was first tested against analytical solutions, were feedback from both models were exchanged by means of auxiliary computational codes. A study reach along the Upper San Pedro River, AZ, was selected to test the methodology in a real case situation. The site was selected due to its perennial character and well data availability for model calibration. Previous modeling efforts in the area were also a factor in the selection. No rainfall-generated runoff was simulated at this point, only a flood wave propagating through the reach interacting with the alluvial aquifer, including evapotranspiration from riparian vegetation. Exchange flows between subsurface and surface systems were computed as the flood wave traversed the study reach. Groundwater flow patterns and mass balance terms were acceptable for the study, establishing how feasible is to couple MODFLOW with KINEROS, though a more fine-tuning calibration is needed.
H31D-0644
Watershed Airborne Telemetry Experimental Research (WATER): An Remote Sensing Experiment in a Typical Arid Region Inland River Basin of China
Among the many land surface experiments have been carried out so far, arid and cold regions were paid little attentions. The land surface observations in arid and cold regions, both remotely sensed and in situ, need to be strengthened for a better understanding of hydrological and ecological processes at different scales. The Watershed Airborne Telemetry Experimental Research (WATER) is a simultaneous air-borne, satellite- borne, and ground-based remote sensing experiment conducted in the Heihe Basin, the second largest inland river basin in the northwest arid regions of China. The WATER is aiming at the research on water cycles, eco- hydrological and other land surface processes in catchment-scale. Data sets with high-resolution and spatiotemporal consistency will be generated based on this experiment. An integrated watershed model and a catchment-scale land/hydrological data assimilation system is proposed to be developed. The mission of WATER is to improve the observability, understanding, and predictability of hydrological and related ecological processes at catchmental scale, accumulate basic data for the development of watershed science and promote the applicability of quantitative remote sensing in watershed science studies. The objectives of the experiment will be (1) Observing major components of water cycle in three experiment areas, i.e., cold region, forest, and arid region hydrology experiment areas, by carrying out a simultaneous air-borne, satellite-borne, and ground-based experiment. (2) Developing the scaling method using airborne high-resolution remote sensing data and intensive in situ observations, and improving remote sensing retrieval models and algorithms of water cycle variables and corresponding ecological and other land variables/parameters. (3) Developing a catchment-scale land data assimilation system, which is capable of merging multi-source and multi-scale remote sensing data to generate high resolution and spatiotemporal consistent data sets in order to improve the predictability of water resources and environmental changes. (4) Using all the available data in the validation, possible improvement and development of catchment-scale hydrological and ecological models as well as decision support tools for water resource management.
H31D-0645
Impervious Surface Mapping of Jungnang-cheon Basin of Korea Using Remote Sensing Images
Impervious surface is the important index for the estimation of urbanization and environmental change. In addition, impervious surface affects on various hydrological process such as the short-term rainfall runoff modeling, water balance analysis, and groundwater estimation in urban area. Therefore, the estimation of impervious surface is an important factor to analyze urban flood. The main objective of this study is the impervious surface mapping of case study area using remote sensing images. Case study area is Jungnang- cheon basin in South Korea. Remote sensing images for the impervious surface mapping are landsat-7 ETM+ and high resolution satellite image of Jungnang-cheon basin. Moreover, a tasseled cap transformation and NDVI transformation apply to landsat-7 ETM+ for considering various predicted parameters. Impervious surface is estimated by using regression tree algorithm which is a binary recursive partitioning process and a rule-based model for the prediction of continuous variables based on training data. Regression tree algorithm is applied to training data sets which are collected by overlaying between landsat-7 ETM+ and high resolution satellite image with different spatial resolution. Then, the predicted variables such as band 3(red), band 4(nearIR), band 5(midIR), and band 7(nearIR) of landsat-7 ETM+ and TC2(greenness) and TC3(wetness) of a tasseled cap transformed image and NDVI transformed image are selected for the efficient and fast prediction modeling. The independent variable of model is a continuous impervious index represented by percentage. The accuracy of variables combination is compared by the average error(AE), the relative error(RE), and correlation coefficient. As the results, the selected test composes with band 3, 4, 5 and 7 of landsat-7 ETM+, the greenness of a tasseled cap transformed image and NDVI. It shows the highest correlation coefficient(0.92) and the smallest the total average error(9.2). In addition, 10-folds cross-validation is used to evaluate the performance of regression tree for all tests. The suggested test has the highest correlation coefficient and the smallest error. Finally, the impervious surface mapping is performed by using the predicted variables from the selected prediction model.
H31D-0646
LIS-Noah land surface model validation in the Southern Great Plains (SGP)
Land surface models are used in various studies to investigate the effects of environmental and climate changes. However, running these models at increasingly high resolutions or for many point measurement locations is computationally and observationally challenging. In this study, we test and validate the Noah land surface model within the framework of the Land Information System. The forcing input data such as radiation, precipitation, wind and temperature records are gathered from 12 SGP sites located in the Midwestern USA. These forcing data are input to the LIS-Noah model on a 30-min interval. Other required data for the LIS-Noah model, such as the vegetation, soil, albedo, greenness fraction parameters are taken from various data sources using the 1-km resolution grid settings of the model assuming that those 1-km blocks fairly represent the site specific vegetation, soil, albedo and greenness fraction. Simulated fluxes from the model are compared with the observed fluxes at the same SGP sites. The results of the study show that the simulated and observed fluxes such as the sensible, latent and ground heat fluxes match fairly well at those SGP sites during warm weather period. The timing and the rate of change in the fluxes are in good agreement. However, the model overestimates the fluxes during the winter season despite maintaining the seasonal patterns of the fluxes. The results indicate that the model has the capability to capture the main features of land surface radiative fluxes except in the winter. The problem with the wintertime radiative fluxes is related to the modelfs higher sensitivity to snow cover on the ground triggering positive temperature-albedo feedback mechanisms and degraded quality of the forcing data in winter.
H31D-0647
Evaluation of South American LDAS atmospheric forcing datasets for use in regional land surface modeling over the LBA region
Significant advances have been made in the past few years by the LBA project on towards understanding how the water, energy and carbon cycles function in the Amazon. However, most of these studies have been limited to results from point measurements from strategically located sites in the tropical forest and other LBA-related areas. As the LBA project progresses into its synthesis phase, there is increased interest in using the acquired knowledge to better understand how Amazonia works as a regional entity. The South American Land Data Assimilation System (SALDAS) initiative, which involves NASA/GSFC, CPTEC/INPE and University of Arizona, provides the capability to integrate results within the robust land surface modeling and data assimilation infrastructure that has already been developed at NASA/GSFC and used for regional studies over the LBA region. This study investigates the feasibility of using the SALDAS atmospheric forcing datasets (a 5 years combination of CPTEC reanalysis and surface observations) for land surface modeling over the Amazonia by comparing these forcing data with seven LBA flux towers observations. The discussion of the results focuses on whether the ranges shown in the evaluation (e.g. standard deviation, bias) are within acceptable ranges for land surface modeling over the region. The results of applying this forcing datasets to force the Noah and SiB3 land surface models over the LBA region are also discussed, with emphasis on the integrated water, energy and carbon budgets.
H31D-0648
Estimation of Land Surface Parameters by LDAS-UT: Model Development and Validation on Tanashi Field Experiment
The estimation of soil moisture and surface energy fluxes at various temporal and spatial scales remains to be an outstanding problem in hydrologic and meteorological researches. Remote sensed data retrieval algorithms, land surface models and data assimilation systems are highly expected to provide a solution to this problem. But the parameters required by those algorithms and systems, such as the soil texture, porosity, roughness parameters and so on, are highly variable or unavailable. In this study, a land data assimilation system (LDAS- UT) is employed to inversely estimate the optimal values of those land surface parameters with meteorological forcing data and remote sensed data. And a field experiment is designed to provide a well-controlled data set for the system validation. The Tanashi experiment has been in operation since November, 2006 in the farm of the University of Tokyo. Continuous ground measurements of meteorological variables, soil moisture and temperature profiles and vegetation status have been taken over a plot, in which winter wheat was planted. At the same time, the ground based microwave radiometers (GBMR) are employed to provide accurate field measurements of brightness temperature up-welling from the plot, at the frequencies of 6.925, 10.65, 18.7, 23.8, 36.5 and 89 GHz. The LDAS_UT is then run with using data obtained from this experiment to retrieval parameters for two periods. One is the period from December 2006 to February 2007, the germination period of winter wheat, and during which the vegetation effects are small. The second period is from April to May 2007, during which the winter wheat was developing rapidly. The optimize parameters were compared with the in situ observed ”®real' ones. It found that, for the first period, the retrieved parameters are close to the ”®real' values, while for the second period, the gap between the retrieved parameters and the ”®real' values are much bigger. The difference between the optimized parameters and the observed ”®real' ones reveals the potential source of uncertainty in the system due to the limitation of current RTM and LSS.