WebStart WEPS: WEPS with remote data access and cloud-computing functionality

Authors: Wagner, Larry; Haas, Mark; Fox, Jr, Fred

Submitted to: Journal of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/19/2022
Publication Date: 1/19/2022
Citation: Wagner, L.E., Haas, M.E., Fox, F.A. 2022. WebStart WEPS: WEPS with remote data access and cloud-computing functionality. Journal of the ASABE. 65(2):427-436. https://doi.org/10.13031/ja.14773.
DOI: https://doi.org/10.13031/ja.14773

Interpretive Summary: The Wind Erosion Prediction System (WEPS) is a physical process based wind erosion model developed by the United States Department of Agriculture - Agricultural Research Service (USDA-ARS). WEPS simulates climate and management driven changes to the surface, vegetation and soil on a daily basis and estimates erosion losses when surface conditions and wind speeds are sufficient for soil movement to occur. WEPS was initially released in 2010 to the USDA-Natural Resources Conservation Service (USDA-NRCS) and is now their preferred wind erosion model to assist land managers in controlling wind erosion, establishing field level plans to conserve the soil and to determine wind erosion susceptibility as a part of national conservation programs to preserve the soil resource. The current WEPS interface is designed to allow easy user access to national scale input data. There are only four primary inputs required to setup and execute a WEPS run: a) field dimensions; b) field location, which automatically determines which climate data to use; c) a representative soil specified for the field; and d) the sequence of management practices applied on the site. However, these inputs access a significant amount of data currently provided primarily in locally stored files. Maintaining this data on every computer running WEPS and keeping it current are issues that NRCS and other WEPS users would like to eliminate. To address these concerns, remote access functionality was incorporated into WEPS to deliver the climate and wind data, as well as the operation, crop/residue and management records for the WebStart release of WEPS. A Java WebStart framework was also employed because it provided a simplified delivery mechanism to the WEPS users’ computer that would automatically update to the latest version of the model. In addition, it also allowed the developers to keep the current WEPS desktop interface, leveraging the use of that code base and retaining the interface’s full functionality. WebStart WEPS also features both remote access to the necessary WEPS data as well as the option of running the WEPS science model components remotely, "in the cloud", as web services. Thus, the computational and storage requirements of the WEPS user’s computer are significantly reduced. This presentation also explores the decision making process that went into the selection of the Java WebStart delivery framework and discusses the unique issues that arose while incorporating remote services functionality into WEPS.

Technical Abstract: The Wind Erosion Prediction System (WEPS) is a daily time step, process based wind erosion model developed by the United States Department of Agriculture - Agricultural Research Service (USDA-ARS). WEPS simulates climate and management driven changes to the surface/vegetation/soil state on a daily basis and estimates erosion losses when surface conditions cause the wind surface friction velocity to exceed the computed surface threshold friction velocity. WEPS was released to USDA-Natural Resources Conservation Service (USDA-NRCS) in 2010. WEPS has since been the USDA-NRCS preferred wind erosion model to assist land managers in controlling wind erosion on wind erosion susceptible cropland fields. It is used to establish baseline field-level conservation plans to reduce soil loss as well as to determine if the risk level to wind erosion meets eligibility standards for participation in select USDA national conservation programs. WEPS greatest attribute is its ability to be utilized as a "what-if" tool to evaluate different management scenarios for developing and evaluating alternative wind erosion control practices to apply at the field level. WEPS is currently designed with a modular coded Fortran science model and a multi-platform capable Java based user interface. The interface allows easy user access to national scale weather and soil data as well as crop/residue and operation records representing the most commonly grown crops and field operations used in the United States. There are only four primary inputs required to setup and execute a WEPS run: a) field dimensions (size/shape/orientation); b) field location (where the field lat/lon automatically determines the weather and wind data to be used; c) dominant erosive soil component on the field; and d) the sequence of management practices applied on the site. These inputs, however, access a significant amount of data currently provided in local stored files and databases on the user's computer. Maintaining this data locally on every computer and keeping it current poses two major issues: ensuring sufficient local disk space is available to retain copies of the data and providing a automated means to keep the data current. To address these concerns, remote access has been provided for the climate and wind data used by WEPS as well as the operation, crop/residue and management records for the WebStart release of WEPS. A Java WebStart framework was employed to retain the current WEPS interface, leveraging the current code base and still keeping the full functionality of the current desktop interface, without having to rewrite the interface as a complete web application. WebStart WEPS allows both remote access to the necessary WEPS data as well as the option of running the WEPS science model components remotely, "in the cloud", as a service. Thus, the computational and storage requirements of the client computer are significantly reduced for making WEPS runs, as long as the user has access to these remote services. This presentation explores the decision making process that went into the selection of the Java WebStart delivery framework and discusses the unique issues that arose while incorporating remote services functionality into WEPS, while still preserving its ability to be run offline with locally stored data.