Skip to main content
Presentation
A software development kit for creating hydrologic apps in the cloud
Spring Runoff Conference
  • Nathan Swain
  • Scott Christensen
  • Jim Nelson
  • Norman Jones
Location
Eccles Conference Center
Event Website
http://water.usu.edu
Start Date
4-1-2014 7:00 PM
End Date
4-1-2014 7:20 PM
Description

Water resources stakeholders and decision makers often use hydrologic simulations to estimate and analyze watershed responses to specific scenarios. Such decisions are formed largely on the basis of static reports generated from hydrologic simulation analysis and forecasts. One problem with this genesis of decisions is that at the time a report is generated from the simulation model it is impossible to analyze all current or even conceive of future scenarios for which the model may be applicable. This severely limits the utility of the model in making good decisions in the first place. A web-based or cloud environment capable of housing more sophisticated but data rich spatially distributed simulations is ideal for creating “living models” that can be used to look at a wide range of existing and possible future scenarios that might be critical in making important water resources decisions. A software development kit (SDK) for creating hydrologic web apps is currently being developed as part of the National Science Foundation (NSF) sponsored Cyber Infrastructure Water (CI-Water) project. This SDK can be used to develop narrowly focused scenario exploration workflows in the form of web applications or web “apps” that will enable clients to create scenarios for existing models, run them in a cloud environment, and post-process and visualize the results. The end result is that previously unusable models become accessible to clients for scenario exploration exercises in a controlled environment as living models. The SDK synthesizes many state-of-the-art web projects to provide the functionality needed to build hydrologic modeling apps. The SDK is a plugin for the CKAN data management system. Apps are developed using Python and the Model View Controller (MVC) development paradigm and hosted as part of a customized version of CKAN. On advantage of using Python for development is that all of the scientific Python modules can be incorporated into the apps. For spatial components of hydrologic models, the SDK offers tools for working with the PostgreSQL with PostGIS extension for storing spatial data, 52North WPS with GRASS and Sextante geoprocessing libraries, and Google Earth and Google Maps for spatial visualization. For visualizing tabular data, the SDK includes the popular HighCharts plotting library. With the SDK plots, maps, and other common user interface elements can be inserted into an app using only a few lines of code to allow rapid development of the apps. This material is based upon work supported by the National Science Foundation under Grant No. 1135482

Citation Information
Nathan Swain, Scott Christensen, Jim Nelson and Norman Jones. "A software development kit for creating hydrologic apps in the cloud" (2014)
Available at: http://works.bepress.com/norman-jones/4/