Page Banner

United States Department of Agriculture

Agricultural Research Service

Research Project: USING REMOTE SENSING & MODELING FOR EVALUATING HYDROLOGIC FLUXES, STATES, & CONSTITUENT TRANSPORT PROCESSES WITHIN AGRICULTURAL LANDSCAPES
2007 Annual Report


1a.Objectives (from AD-416)
Develop new remote sensing, modeling and data assimilation techniques to improve the monitoring of hydrologic fluxes and agricultural pollutant pathways at the field, watershed and regional scale.


1b.Approach (from AD-416)
Effective management of these watersheds requires detailed process-level understanding concerning the complex hydrologic and constitutive flux pathways that govern: the availability of root-zone soil moisture, the delivery of agricultural pollutants to surface water bodies and feedbacks operating along the soil-plant-atmosphere continuum. The inability to measure the relative magnitude of these pathways hampers the development of effective water management strategies at watershed- and regional-scales. This project attempts to develop novel remote sensing and modeling tools to better characterize key hydrologic and constitutive flux pathways operating within agricultural watersheds. An overarching theme of the project is that the integration of remote sensing products into models can enhance the utility of models for critical agricultural applications.


3.Progress Report
This project attempts to develop novel remote sensing and modeling tools to better characterize key hydrologic and constitutive flux pathways operating within agricultural watersheds. An overarching theme of the project is that the integration of remote sensing products into models can enhance the utility of models for critical agricultural applications. Year 1 (2007) milestones for the project are generally dominated by data collection, early model development and refinement of existing techniques. All have been either “fully met” of “substantially met.” The research team is on track to reach future project goals without substantial modification to the approved research plan.


4.Accomplishments
1. Monitoring drought at local scales with polar orbiting satellites: Effectively managing agricultural drought requires soil water availability information at finer spatial scales that existing drought monitoring techniques can provide. To improve the spatial resolution of agricultural drought information, data from the Landsat and Terra polar orbiting satellites operated by NASA were used in a model to map evapotranspiration and moisture stress at 60m-1km spatial resolution. These high-detail maps can be nested within a coarser resolution product covering the continental U.S. at 10km resolution, generated with data from the Geostationary Operational Environmental Satellites (GOES). At the finer resolutions, individual fields can be resolved facilitating assessment of vegetation stress by crop type at the sub-county level. This type of information may be extremely useful for improving yield projections and for administering crop insurance and drought assistance programs. Paired with drought assessments on the continental-scale coarse grid, these local evaluations can be placed within a regional context, which reflects larger scale climatic conditions. Research addresses the Irrigation Water Management and Watershed Management, Water Availability, and Ecosystem Restoration NP 201 problem areas.

2. Implementation of a new approach for country-level yield monitoring using remote sensing: Current ground-based methods for crop yield assessments yield only state-only information. To address this limitation, a 5-year database (2002-06) of Moderate Resolution Imaging Spectroradiometer (MODIS) from the NASA Terra imagery, surface climatic data from NOAA stations and soil profile physical characteristics was completed. A new algorithm for disaggregating National Agricultural Statistics Service (NASS) state yields down to a county level using only MODIS imagery was tested for Iowa and Illinois and county-level corn and soybean yields were predicted at the end of the 2006 crop season. The results were provided to the NASS Research and Development Division for further evaluation and potential use in operational applications. Research addresses the NP 201 Watershed Management, Water Availability, and Ecosystem Restoration problem area.

3. Completed testing of an innovative soil moisture remote sensing technology: Although microwave radiometry at L-band (~1.4 GHz) has been known as an optimal solution for remote-sensing of soil moisture, the constraint on large antennas size required for the low-frequency radiometry has limited the development of L-band satellite instrument for global-scale observation. This limitation can be overcome by using an interferometric technology, called aperture synthesis. Experiments were conducted to demonstrate the soil moisture mapping capability of the first aircraft L-band radiometer that employed two-dimensional aperture synthesis. During the Soil Moisture Experiment in 2003 (SMEX03), the two-dimensional synthetic aperture radiometer (2D-STAR) was flown onboard an aircraft over Alabama, Georgia and Oklahoma. There was fairly good instrument sensitivity to soil moisture variations in both pasture and forest areas. The 2D-STAR’s sensitivity to the soil moisture under dense vegetation canopy was remarkably superior to a conventional high-frequency radiometer that was used in the same experiment. The results of this study are very important to the future of spaceborne remote sensing and provide justification for the further development of this technology. The improved remote sensing capabilities of L-band interferometric radiometry will be of benefit to hydrologic applications in agriculture and climate. Research addresses the NP 201 Watershed Management, Water Availability, and Ecosystem Restoration problem area.

4. A new remote sensing technique to describe nutrient uptake by cover crops on Maryland’s Eastern Shore: Throughout the Chesapeake Bay area, a strong emphasis has been placed on the use of winter cover crops to reduce the loss of excess agricultural nutrients following the growing season, but current understanding of cover crop effectiveness relies upon small-plot experimental data and implementation acreage sums, rather than upon landscape analysis. We have developed and employed an innovative methodology that combines farm program records, satellite remote sensing, and on-farm sampling to calculate the amount of nitrogen sequestered in cover crop biomass on farms enrolled in state cover crop cost share programs within the Choptank River watershed. Results indicate that planting date, planting method, and cover crop species significantly influence nutrient uptake efficiency, with early-planted, drilled, rye cover crops exhibiting the highest fall nitrogen uptake (up to 60 lb/acre). This efficient monitoring technology for cover crops will allow program and watershed managers to optimize implementation of this important best management practice at watershed and regional scales, accounting for the effects of spatial variability (watershed position, soil type, hydrology) as well as agronomic factors. Research addresses the NP 201 Conservation Practice Assessment problem area.

5. Development of a new soil moisture remote sensing technique: Microwave instruments can be designed to measure many features of the Earth that are important in agricultural monitoring. Polarimetric measurements focus on variations related to the orientation of the waves. These measurements may provide new information that can be used to estimate or describe land surface features (e.g. crop row direction). Data from a polarimetric satellite instrument called WindSat were used in this study to determine if this was possible. Four large and homogenous regions were identified that represented unique land surface features (forest, deserts, agriculture). The results of this study clearly show that when aligned land surface features (topographic or agricultural row structure) are present there were unique polarimetric responses. This is the first time this has been observed at satellite footprint scales. These results contribute to the accuracy and reliability of operational soil moisture retrievals, which should lead to increased acceptance of the soil moisture products in applications involving hydrology and agriculture. Research addresses the NP 201 Watershed Management, Water Availability, and Ecosystem Restoration problem area.

6. Sharpening of thermal imagery for monitoring field-scale crop water use and water stress: High spatial resolution (< or = 100 m) thermal infrared (TIR) band imagery has utility in a variety of agricultural applications, which include detecting crop water stress conditions and monitoring water use or evapotranspiration (ET) down to scales of individual agricultural fields. This type of information is needed to reliably assess water availability and crop yield in the Midwest and Western United States as well as in other agricultural regions around the world. The continuation of thermal band imaging on Landsat Data Continuity Mission platforms is currently under debate, and such high resolution thermal data (~100 m) may soon be unavailable. A technique to derive higher resolution land surface temperature (LST) from other available data is therefore highly desirable, which has led to an approach for sharpening LST using vegetation index (VI)-LST relationships. This method has been refined and evaluated over an extensive corn/soybean production area in central Iowa during a period of rapid crop growth. The results prove a great potential in significantly enhancing the thermal information available over this agricultural area, and indicating the potential for routine monitoring of ET and stress conditions for the two important crops (corn and soybean) produced in the Midwest region. Research addresses the Irrigation Water Management and Watershed Management, Water Availability, and Ecosystem Restoration NP 201 problem areas.

7. Development of new techniques for evaluating remotely-sensed precipitation and soil moisture retrievals: A persistent problem for the development of accuracy remote sensing algorithms for precipitation and surface soil moisture is a lack of adequate ground-based observations to compare satellite retrievals to. Consequently, it is often difficult to evaluate and inter-compare competing retrieval approaches. To address this issue, a novel, data-assimilation based approach was developed that evaluates remotely sensed precipitation and soil moisture products based on their mutual dynamic consistency. In this way the accuracy of remotely sensed soil moisture products can be indirectly inferred based on how well in responds to known variations in rainfall (and vise versa). Testing of the method over data rich areas of the United States has clearly demonstrated the value of the approach for accurately inferring soil moisture and precipitation retrieval accuracies over data-poor areas lacking sufficient ground-based observations for traditional validation approaches. Continental- and global-scale application will provide a valuable performance feedback tool for efforts to map soil water availability from space. Research addresses the NP 201 Watershed Management, Water Availability, and Ecosystem Restoration problem area.

8. Collection of datasets required for conservation effects assessment: Optimal conservation practices (CP) for improved water quality are not known quantitatively at the watershed scale. The Conservation Effects Assessment Project (CEAP) was created as a joint NRCS/ARS program to assess and quantify the effects and benefits of USDA conservation programs. Water quality models can be used for such assessments; however their effective application requires extensive ground-based data sets for model calibration and validation. To address this need in the Choptank watershed, datasets describing hydrologic and nutrient transport variables have been gathered for a sample of sub-watersheds (no-tidal portion) within the watershed. After a thorough calibration and validation exercise, these datasets will allow for the effective application of models to provide conservation practice assessments for critical water quality issues within the Choptank watershed and the larger Chesapeake drainage region (e.g. poultry litter application). Research addresses the NP 201 Conservation Practice Assessment problem area.

9. A new algorithm for the remote sensing of vegetation water content: Routine monitoring of vegetation water content in current and future operational meteorological satellites using this data product algorithm will help in determining drought stress, potential for wildfires, and soil moisture content. ARS scientists in Beltsville, MD in collaboration with University of California Davis and NASA Ames Research Center, have developed algorithms for remote sensing of vegetation water content as a standard data product for one of NASA’s premier instruments, the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard the Terra and Aqua satellites. The algorithm incorporated research from the Soil Moisture Experiments 2004 and 2005 conducted in ARS experimental watersheds in Arizona and Iowa. The research is presented in a comprehensive Algorithm Technical Basis Document. Research addresses the NP 201 Watershed Management, Water Availability, and Ecosystem Restoration problem area.

10. Improvements in riparian area modeling for water quality assessments: Accurate modeling of water movement through riparian buffer areas in critical for watershed-scale water quality assessments. In a collaborative work with ARS scientists at Tifton, GA, the Riparian Ecosystem management Model (REMM) was qualitatively tested, using GA riparian database, for sensitivities in parameterizations associated with buffer characteristics. Specifically, the sensitivity of REMM nutrient and sediment output was quantified with respect to the changes in key riparian buffer parameters including, vegetation and soil characteristics. Findings revealed that parameters associated with vegetation characteristics (such as rooting depth and plant height) only moderately affect nutrients and sediment yields from buffer zones, while outputs are highly sensitive to changes in physical parameters such as slope and Manning's surface roughness coefficient. HRSL scientists are developing a similar database for the Choptank (MD) watershed for REMM model evaluation. It is anticipated that this new modeling framework will assist farmers and policy makers on the mitigation effectiveness of specific riparian buffers relative to pollutant loading potentials into streams and other surface water resources. Research addresses the NP 201 Conservation Practice Assessment problem area.

11. Improved root-zone water balance prediction via the assimilation of a thermal remote sensing soil moisture proxy; Rainfall-driven water balance models are frequency used to operationally monitor the availability of root-zone (surface to 1-meter) soil water availability for agricultural crops. Unfortunately, these models are prone to wide range of errors. Such errors, however, could be reduced through the assimilation of root-zone soil moisture estimates independently acquired using thermal-based remote sensing techniques. In this research, a data assimilation technique was developed to optimally integrate such remote sensing estimates into a rainfall-driven soil moisture model. The data assimilation effectively eliminated the degrading impact of accumulating water balance errors associated with poor rainfall observation and/or the inaccurate prediction of evapotranspiration. The ability of such an approach to improve the monitoring of root-zone soil water availability was clearly demonstrated. Wide-scale application of such an approach has the potential to improve irrigation scheduling, drought monitoring and the detection of hydrologic impacts of global change. Research addresses the NP 201 Watershed Management, Water Availability, and Ecosystem Restoration problem area.

12. A new management practice for reducing nitrate loss in drainage waters: In the Choptank River basin (MD), excess nutrient transport from agricultural fields is primarily through extensive drainage open ditches. Studies have shown that annually an average of 6% of nitrate applied to agricultural fields can be transported in drainage water to receiving surface waters. One of the best management practices (BMPs) being used in the open ditches has been the installation of a water control structure at a drainage outlet. These control structures can be placed in field ditch to control water levels from agricultural fields to reduce water flow and nutrient loss. As part of Choptank Conservation Effects Assessment Program (CEAP) activities, several control drainage systems have been implemented for monitoring water flow and sampling. One potential management schedule is to increase water elevation at the outlet elevation to just below the root zone during the growing season and lower to the bottom of the drainage ditch at the outlet (free drainage) during planting and harvesting operations. Preliminary studies have shown that this schedule can reduce nitrate losses up to 40%. Elevating the drainage outlet to just below the root zone during the growing season can potentially reduce nitrate losses in drainage water from 15% to 30%. Findings from this study will provide quantitative efficiencies for both water and nitrate reductions and better management strategies for more efficient use of these control drainage BMPs. Research addresses the NP 201 Conservation Practice Assessment problem area.

13. Improved large scale energy flux modeling Using a Large Eddy Simulation model: The Large Eddy Simulation (LES) model that simulates the turbulent exchange of heat, water vapor and momentum (wind energy) between the land surface and lower atmosphere has been coupled with remote sensing-based land surface model and applied to agricultural study regions in the Texas High Plains, Southern Great Plains, and the Desert Southwest. Preliminary analyses from the suite of simulations indicate the strength of surface contrasts in surface temperature, canopy cover and moisture has a significant affect on surface-air coupling and resulting spatial distribution of near-surface air properties (wind speed, air temperature and humidity). This in turn can have a measurable impact on flux computations since typically land surface models assume uniform atmospheric properties over the landscape. However, in many cases the results indicate that variations in air temperature and wind speed tend to cancel their individual effects on model-derived fluxes, and as a result local variations in air temperature and wind speed often have a minor affect on the flux fields. On the other hand, using atmospheric inputs from a weather station located in non-representative area can still cause significant errors in heat flux estimation. Therefore, a simple scheme is under development to estimate spatial fields of air properties using remote sensing data. From this work, we will be able to account for the effects of non-uniform air properties on flux estimation and for the quality of land surface model output affected by non-representative weather station data input. This will ultimately improve the capability of monitoring water use and crop conditions at regional scales. Research addresses the Irrigation Water Management and Watershed Management, Water Availability, and Ecosystem Restoration NP 201 problem areas.

14. Creation of watershed-scale, ground-based soil moisture datasets required for validation of satellite-based soil moisture estimates: Large-scale, ground-based observations of surface soil moisture are difficult to acquire but vitally important for efforts to validate satellite-based soil moisture estimates. In response, scientists have completed archiving of four watershed-scale soil moisture databases from 2002 to 2005 Data from large-scale networks imbedded with the Little River (GA) Reynolds Creek (ID) and Little Washita (OK) watersheds are ingested daily and watershed averaged soil moistures are computed for comparison to satellite remote sensing estimates. The Walnut Gulch (AZ) is off-line currently and the transmission system is being modified for implementation in October 2007. The networks have been quality controlled to produce a stable estimate of large-scale soil moisture dynamics. These estimates will be used to compare to soil moisture retrieval algorithms from existing satellite. Research addresses the NP 201 Watershed Management, Water Availability, and Ecosystem Restoration problem area.

15. Successful application of a new technique for the mapping of sub-surface water flow pathways: Subsurface hydrologic processes are difficult to quantify and therefore poorly understood - introducing uncertainty in efforts to trace and/or predict pollutant transport from agricultural fields. Detailed ground-penetrating radar was collected every 2-m and analyzed over a 1.78 ha field and subsequently linked with a kinematic GPS elevation data to quantify soil structures that could influence subsurface water and chemical fluxes exiting this parcel of land. It was observed that subsurface water converged into several discrete subsurface flow pathways. Some of these pathways were narrow (1-3 m wide) while others were several meters wide (> 10 m). With the capacity to accurately determine the location of these pathways protocols for monitoring water and chemical fluxes exiting agricultural land are being developed which will improve water quality monitoring and models efforts. Research addresses the NP 201 Watershed Management, Water Availability, and Ecosystem Restoration problem area.

16. The novel application of remote sensing to invasive weed monitoring: Worldwide, invasive weeds cause billions of dollars worth of damage annually in economic losses. Remote sensing can be used to monitor many species of invasive weeds but it is unknown which sensors can be used under specific conditions. Researchers combined simulation models with spectral libraries of invasive weeds and tested the output using leafy spurge in Northeastern Wyoming. The methodology can be used to develop a target list of invasive species that can be monitored by remote sensing, potentially saving millions of dollars in ground monitoring costs. Research addresses the NP 201 Watershed Management, Water Availability, and Ecosystem Restoration problem area.

17. Better characterization of biochemical N processing: Movement of agricultural nitrogen (N) into riparian buffers often occurs within discreet seepage or upwelling zones which can limit ability of the ecosystem to process the nutrient. The biogeochemical processing of N was characterized within these zones to assess the ability of riparian wetlands to remove nitrate from soil water. The stratification pattern of microbial populations and DEA was found to be consistent with new carbon inputs to the ecosystem being most important driver of biogeochemical activities such as de-nitrification. This information will permit better management of riparian buffers for effective removal of agricultural nitrate entering the ecosystem from groundwater. Research addresses the NP 201 Conservation Practice Assessment problem area.

18. Understanding energy and water budgets in arid regions: Comparison of reflectance measured in the visible, near infrared, and short wave infrared wavelengths by the Advanced Spaceborne Thermal Emission Reflection Radiometer (ASTER), MODIS/ASTER Airborne Simulator (MASTER), and ground based Analytical Spectral Devices Spectroradiometer (ASD) in a semiarid area of the northern Chihuahuan desert in southern New Mexico USA were made for three dates (May 12, 2001, October 6, 2002, and May 2, 2003) and three different vegetation communities (grass, shrub, and shrub-grass transition). A strong positive correlation between the measurements indicated that the three sensors were measuring similar reflectance values for the three dates and vegetation communities. Reflectance was highest from the shrub and shrub-grass transition communities and lowest from the grass community and was related to the amount of vegetation cover present. This study has implications for understanding the energy and water budgets in this region of the Chihuahuan desert where shrub communities with low ground cover are invading and replacing grass communities. Research addresses the NP 201 Watershed Management, Water Availability, and Ecosystem Restoration problem area.


5.Significant Activities that Support Special Target Populations
1. Collaboration with the EPA Chesapeake Bay Program (USEPA CBP) office: In April, 2006, BARC's Associate Area Director at Beltsville, in negotiations with the USEPA/CPB Director of Science, established a 6-month detail for one of our scientist to work closely with CBP scientists in the areas of mutual interests. ARS’ primary involvement have been in the areas of: i) manure technology and their implementations within the Chesapeake Bay watershed; ii) improvement in Chesapeake Bay modeling activities in terms of their side-by-side validations with USDA agricultural watershed models; and iii) the improvement in the quantification of efficiency estimates for the best management practices (BMPs) applied within the Choptank as well as other Chesapeake Bay watersheds. Our scientist is currently involved in the CBP Pillar activities carried out under Strategic Implementation Planning (SIP). The development of SIP was in response to a recent request from Senators Mikulski, Sarbanes and Warner which results in a Government Accountability Office (GAO) audit. The GAO examined (1) the extent to which this office established appropriate measures for assessing restoration progress; (2) the extent to which the reporting mechanisms the CBP uses clearly and accurately describe the Bay’s overall health; (3) how much funding was provided for restoring the Chesapeake Bay for fiscal years 1995 through 2004 and for what purposes; and (4) how effectively the restoration is being coordinated and managed. In its evaluation, GAO found that the CBP does not have a comprehensive, coordinated implementation strategy which will enable it to achieve the goals outlined in the Chesapeake 2000 plan. The findings also cite the lack of consistent, realistic plans for the 10 keystone commitments as a problem. The GAO recommended that the Administrator of EPA instruct the CBP Office to “develop an overall, coordinated implementation strategy that unifies the program’s various planning documents and to establish a means to better target its limited resources to ensure that the most effective and realistic work plans are developed and implemented.” In September 2006, the CBP regional leaders enthusiastically adopted the CBP’s Five Pillar framework (consisted of the 10 keystone commitments) as the platform for the next five years for reviewing federal program involvement in, and support for, the CBP. ARS involvement is within Pillar 4 - entitled “Maintaining Healthy Watersheds.”

2. Development of satellite-base crop yield monitoring tools for NASS: The official USDA forecasts of domestic crop production, yield assessments, crop progress, and condition monitoring are reported by the National Agricultural Statistics Service (NASS). This responsibility is federally mandated, and the reports are regarded as the most influential reports used by the food and fiber sector world-wide throughout the year. Yield assessments rely on reports from surveyed farmers or actual field observations. This operational procedure is relatively expensive to collect, labor-intensive, inconsistent and lacking in spatial variability. There is a need for a more economically efficient and spatially complete data and tools to assess major crop progress, conditions and yields over large areas accurately and rapidly. This research is being conducted in cooperation with NASS personnel at the State and National offices. The preliminary results (2003-2006) from this research in Iowa and Illinois were very encouraging to NASS statisticians in these states. They have requested our cooperation and training to develop operational crop condition and yields at their state offices.

3. Delivery of soil moisture and energy balance remote sensing products to NOAA: In collaboration with scientists in NOAA/NESDIS (National Environmental Satellite Data Information Service), plans are underway to begin implementing the Atmosphere-Land Exchange Inverse (ALEXI) surface energy balance model, developed by HRSL scientists, and supporting model infrastructure at NESDIS starting in FY08. This would constitute a first step towards transitioning this model to operational status as a continental-scale evapotranspiration product supported by NOAA and linked to standard NOAA meteorological data fields.

4. Development of invasive species detection tools for the U.S. Forest Service: ARS scientists worked with U.S. Forest Service personnel in Salt Lake City, Utah to implement comprehensive methods for monitoring leafy spurge by remote sensing and ground crews. Preliminary maps of suitable habitat for leafy spurge were delivered to the Fishlake National Forest in Utah.


6.Technology Transfer

Number of web sites managed1
Number of non-peer reviewed presentations and proceedings71
Number of newspaper articles and other presentations for non-science audiences2

Review Publications
Cosh, M.H., Stedinger, J., Ou, S.C., Liou, K.-N., Brutsaert, W. 2007. Evolution of the variability of surface temperature and vegetation density in the Great Plains. Advances in Water Resources. 30(5):1094-1104.

Anderson, M.C., Norman, J.M., Mecikalski, J.R., Otkin, J.A., Kustas, W.P. 2007. A climatological study of evapotranspiration and moisture stress across the continental U.S. based on thermal remote sensing. I. Model formulation. Journal of Geophysical Research. 112, D10117. http://dx.doi.org/10.10/29/2006JD007506.

Agam, N., Kustas, W.P., Anderson, M.C., Li, F., Neale, C. 2007. A vegetation index based technique for spatial sharpening of thermal imagery. Remote Sensing of Environment. 107:545-558.

Timmermans, W.J., Kustas, W.P., Anderson, M.C., French, A.N. 2007. An intercomparison of the surface energy balance algorithm for land (SEBAL) and the two-source energy balance (TSEB) modeling schemes. Remote Sensing of Environment. 108: 369-284.

Choi, M., Jacobs, J.M., Cosh, M.H. 2007. Scaled spatial variability of soil moisture fields. Geophysical Research Letters. 34(L01401). http://dx.doi.org/10.1029/2006GL028247.

Vivoni, E.R., Guiterrex-Jurado, H.A., Aragon, C.A., Mendez-Barroso, L.A., Reinhart, A.J., Wyckoff, R.L., Rodriguez, J.C., Watts, C.J., Bolten, J.D., Lakshmi, V., Jackson, T.J. 2007. Variation of hydometerological conditions along a topographic transect in northwestern Mexico during the North American Monsoon. Journal of Climate. 20:1792-1809.

Shi, J., Jiang, L.M., Zhang, L.X., Chen, K.S., Wigneron, J.P., Chanzy, A., Jackson, T.J. 2006. Physically based estimation of bare surface soil moisture with the passive radiometers. IEEE Transactions on Geoscience and Remote Sensing. 44:3145-3153.

Last Modified: 10/26/2014
Footer Content Back to Top of Page