HydroGEN Forecast Generator

Case ID:

HydroGEN is a web-based platform for hydrologic simulation utilizing advanced machine learning algorithms and current data to provide accurate forecasts to water managers and other decision makers. The HydroGEN Forecast Generator is a workflow that builds a hydrologic model for any watershed in the United States and seasonal runs forecasts using a machine learning emulator. This workflow makes it possible to generate a seasonal forecast for any watershed in the US on demand. The hydrological simulated models go beyond streamflow scenarios and are designed to provide complete watershed systems including groundwater, soil moisture, streamflow, and plant water use. 

Since 1980, the U.S has sustained 258 weather and climate disasters. Among these 26 have been droughts, costing the nation at least $249 billion, $9.6 billion per incurred drought. Furthermore, floods cost the nation $17 billion yearly, according to the Federal Emergency Management Agency. Lastly, federal wildfire suppression has cost the nation an average of $1.6 billion yearly. Water is the driving force behind these extreme events, damaging the nation’s people and economically straining the US. Hydrological research, analysis, and forecasts to predict these events to mitigate and prevent damages have been very inaccurate utilizing historical events as the basis of the prediction. This is because of the no-analog future correlation when predicting these water-based events. HydroGEN’s model uses current data and specific machine-learning computational system to forecast hydrological scenarios and solve complex groundwater issues more accurately. 


  • Forecasting hydrological scenarios
  • Analyzing groundwater problems
  • Risk management for insurance purposes
  • Aid for wildfire planners, forest management, and water management


  • Physical models generated 
  • Advance models generated
  • Web-based (easier and on-demand access)
  • Advanced software
Patent Information:
Contact For More Information:
Jay Martin
Licensing Associate, Software and Copyright
The University of Arizona
Lead Inventor(s):
Laura Condon
Nirav Merchant
Amy Johnson
Reed Maxwell
Peter Melchior
Jill Williams
Will Lytle
Edwin Skidmore
Yueling Ma
Elena Leonarduzzi
Andrew Bennett
Luis De la Fuente