FUTUREWATER INDIANA: A SCIENCE GATEWAY FOR MODELING SPATIAL-TEMPORAL HYDROLOGICAL CHANGES AT THE REGIONAL SCALE

Traditionally, we develop a hydrological model and publish it in a journal. This is the end of the story. However, the challenges of sustainability and climate adaptation require us to change our way of doing and communicating sciences. In this presentation, we will describe the FutureWater Science Gateway. The core of the gateway is coupled surface-water and groundwater models of the Wabash River basin, which simulate the water cycle to the year 2100 under two Representative Concentration Pathways (4.5 and 8.5) climate scenarios. The outputs of the hydrological models are visualized with detailed maps and interactive graphs. Data are available for download, and models are available to other researchers, state and NGO staff to run their own versions. The model is live—it is updated when new data become available.

The FutureWater gateway is built based on the Apache Airavata gateway middleware framework and hosted under the SciGaP project at Indiana University. The visualization involves geographical map integration and dynamic data provisioning using R-Shiny application deployed in the gateway. The gateway provides an integrated infrastructure for simulations based on the parallelized Soil and Water Assessment Tool (SWAT) and SWAT-MODFLOW software execution on Extreme Science and Engineering Discovery Environment (XSEDE) and Indiana University’s (IU’s) HPC resources. It organizes data in optimized relational databases and enables intuitive simulational result exploration. The gateway also serves as a cyberPlatform for teaching. The climate change concept is exceptionally amorphous, which presents a well-known cognitive challenge. The common examples of climate change impacts are polar bears on breaking away ice, sea level rise on the coasts, and the melting of Greenland glacial, which have been proven to be ineffective. Four teaching modules have been developed to allow students to explore water availability on the gateway under various climate change scenarios near their hometowns. These modules are distributed to the science teachers in the state. A large number of faculty, staff, and students have contributed to this project. Research collaboration is underway with Indiana University and Purdue University computer scientists to explore applications of graph neural networks to predict hydrological extremes.