Page Banner

United States Department of Agriculture

Agricultural Research Service

Research Project: Spatial Data and Implementation Tools for Distributed Watershed Modeling Addressing NRCS Resource Concerns

Location: Agricultural Systems Research Unit

2012 Annual Report


1a.Objectives (from AD-416):
1. Mainstream USDA-ARS Agricultural Systems Research Unit (ASRU) modeling activities via the Object Modeling System (OMS) into NRCS Information Technology (IT) database access processes, corporate data stores, geospatial data access, and other data oriented activities; 2. Align ASRU modeling activities via OMS, particularly those related to the AgroEcoSystem-Watershed (AgES-W) model, with the NRCS Conservation Delivery Streamlining Initiative (CDSI) user interface architecture/infrastructure; 3. Implement AgES-W in the Cloud Services Innovation Platform (CSIP) and related OMS framework enhancements, as needed; and 4. Support the delivery of AgES-W model and tools in OMS to support CDSI regarding assessment of selected NRCS identified resource concerns.


1b.Approach (from AD-416):
NRCS data oriented activities are progressing under the Conservation Delivery Streamline Initiative (CSDI), which will be leveraged to provide geospatial and temporal data access and delivery processes to run ASRU distributed watershed models, particularly the AgroEcoSystem-Watershed (AgES-W), and address NRCS Resource Concerns. The SCA will link data provisioning and model deployment to parallel activities in other ASRU-CSU agreements and NRCS activities. Potential synergies will be explored with the new NRCS Land Management Operational Database (LMOD) and Basin Analysis Geographical Information Systems (BAGIS).

The Cloud Services Innovation Platform (CSIP) will be tested and used to deploy AgES-W on nested watershed models in the lower Cache la Poudre Watershed, where agricultural water and nutrient management are key drivers of water quantity and quality. Targeted spatial conservation practices will also be simulated in Iowa at field to watershed scales, where explicit spatial modelling and interactions between hydrological response units should be essential to capture the integrated responses that impact selected Resource Concerns. Using these two regional case studies, the data provisioning tools will be integrated with parameter estimation methods (being developed in parallel through other SCAs with CSU) to demonstrate general data management processes and their linkages with simulation in OMS. AgES-W model outputs will be mapped to selected NRCS Resource Concerns, where ARS and CSU will inform NRCS of progress through quarterly meetings and annual reports, and receive feedback and possible adjustments to NRCS priorities and CDSI tools and databases.

Tools delivered in OMS and CSIP will be nationally scoped in terms of their abilities to be deployed over the continental USA using the same data provisioning processes and modeling tools. National application of the tools will be the future work of NRCS, but their input into the tool/framework development will be ensured, primarily through regular communication. The deliverables for each year (see Mutual Agreements below) will be demonstrated through public seminars and archived as simulation projects in Javaforge for future reference and development. Thus, the Colorado and Iowa case studies will serve as openly available templates for broader regional and national applications by stakeholders, including but not limited to NRCS.


3.Progress Report:

A preliminary model input data provisioning tool was developed to access the Natural Resources Conservation Service (NRCS) Land Management Operations Database (LMOD). This access uses queries from existing internet services to get an LMOD data array in a specific convention called JavaScript Object Notation (JSON). The data provisioning tool then converts this JSON data array to the model-specific input sets. The prototype tool implements recent advances and functionality developed by NRCS for the web-based Unified Rotation Builder (URB). Thus our progress complements concurrent work on URB for national deployment. To regionalize an area of interest, the data provisioning tool uses information from each geographical Crop Management Zone (CMZ). An LMOD database query-and-answer produces a JSON data array, which includes all of the available crop management and operation details from a specific CMZ necessary for model input file generation. The vegetation or crop names in the LMOD database do not necessarily correspond with the available crop names in a specific model. Thus a translator is required, which generates model-specific crop parameter description files. For this purpose, crop names in LMOD have been mapped (pointed) to data names in the model input files. The primary development is focused on the AgroEcoSystem-Watershed (AgES-W), as a precursor to multi-model data provisioning for models like the Soil and Water Assessment Tool (SWAT) and the Agricultural Policy Environmental Extender (APEX). Information and data needed to run AgES-W that are not available in LMOD are also being identified. The Hydrological Response Unit (HRU) delineation tool was adapted from a tool developed by the Friedrich-Schiller University Jena/Germany to generate model entities and model input files in the data format needed in AgES-W. The adaptations in the HRU-tool are mainly in the part of the tool that creates model input files. The tool now automatically generates AgES-W model entities described in a comma separated values (csv) file. Also to link HRUs spatially, the hydrological routing scheme generates a multiple-HRU linkage description. The original tool was completely in the German language. Now the USDA-ARS has a version translated into English. The HRU delineation tool was tested on a watershed at the field scale. The Scott-East watershed (56 ha) on the Drake Farm is located in a sub-catchment of the Cache la Poudre Watershed, east of Fort Collins and northwest of Greeley. The generated HRUs describe the Scott-East watershed using its location in the landform configuration, a soil type value, a land-use value, a hydro-geology property value and a connection to a sub-catchment that is inherent from a delineation layer, for instance a soil moisture measurement network. The intent is to calibrate, test and improve AgES-W, by comparing and analyzing the measured and modeled soil water at 16 locations and 4 depths and watershed runoff data. An Application Programmer Interface (API) for the Cloud Services Innovation Platform (CSIP)was created for uploading model input parameters, model execution, and downloading model outputs. AgES-W Model Service execution was prototyped with native parameter files using this API. All of these activities enable the goal of demonstrating the use of available NRCS data stores and services to run complex watershed models such as AgES-W.


Last Modified: 10/21/2014
Footer Content Back to Top of Page