Keywords

Cloud computing, Web application, Ecological risk

Location

Session A3: Innovative Architectures and Approaches of Cloud and Mobile Technology for Environmental Modeling

Start Date

18-6-2014 9:00 AM

End Date

18-6-2014 10:20 AM

Abstract

The U.S. Environmental Protection Agency (USEPA) relies on a number of ecological risk assessment models that have been developed over 30-plus years of regulating pesticide exposure and risks under Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) and the Endangered Species Act. Since computing technology have changed dramatically over this time period, constituent legacy models often contain algorithms based on source code with defunct dependencies and/or have been integrated with graphical user interface elements no longer compatible with current operating systems. Model migration to modern web applications creates integration challenges for back-end science model code residing on a server. An example of a legacy FORTRAN code that we integrated into our regulatory web application (the übertool) is the Pesticide Root Zone Model (PRZM). PRZM is comprised of three components: a cloud-based server to host a browser based graphic user interface (GUI) as a front-end, a database to store model results, and a REST API which invokes the model. A significant advantage of cloud-based PRZM is that hardware operating system compatibility issues (an important obstacle in running legacy FORTRAN models) are avoided because simulations are executed on the cloud server. In addition, migrating to the cloud reduces development and maintenance costs, facilities model coupling and offers instant deployment of updates and new versions. With the database element, users can save simulation outputs or retrieve previous assessments for comparison.

Share

COinS
 
Jun 18th, 9:00 AM Jun 18th, 10:20 AM

Back-end Science Model Integration for Ecological Risk Assessment

Session A3: Innovative Architectures and Approaches of Cloud and Mobile Technology for Environmental Modeling

The U.S. Environmental Protection Agency (USEPA) relies on a number of ecological risk assessment models that have been developed over 30-plus years of regulating pesticide exposure and risks under Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) and the Endangered Species Act. Since computing technology have changed dramatically over this time period, constituent legacy models often contain algorithms based on source code with defunct dependencies and/or have been integrated with graphical user interface elements no longer compatible with current operating systems. Model migration to modern web applications creates integration challenges for back-end science model code residing on a server. An example of a legacy FORTRAN code that we integrated into our regulatory web application (the übertool) is the Pesticide Root Zone Model (PRZM). PRZM is comprised of three components: a cloud-based server to host a browser based graphic user interface (GUI) as a front-end, a database to store model results, and a REST API which invokes the model. A significant advantage of cloud-based PRZM is that hardware operating system compatibility issues (an important obstacle in running legacy FORTRAN models) are avoided because simulations are executed on the cloud server. In addition, migrating to the cloud reduces development and maintenance costs, facilities model coupling and offers instant deployment of updates and new versions. With the database element, users can save simulation outputs or retrieve previous assessments for comparison.