GHOST: A WATERSHED-SCALE HYDROLOGIC MODEL
IWA researchers at IIHR—Hydroscience & Engineering (the parent organization of the Iowa Flood Center) developed and implemented a watershed-scale hydrologic model, the Generic Hydrologic Overland-Subsurface Toolkit (GHOST), to estimate watershed responses to rainfall events. GHOST considers Iowa’s varied topography, soils, and land use to simulate the hydrologic responses at watersheds over time periods on the order of decades.
GHOST is based on the open source hydrologic code MM-pihm (Qu and Duffy 2007, Yu et al. 2013), which fully couples surface and subsurface water systems to predict streamflow and groundwater movement for normal and extreme rainfall and snowmelt events. GHOST is a reformulation of MM-pihm designed to perform multi-year simulations in large basins. IIHR modelers developed modifications and modules in GHOST specifically to:
- Increase performance
- Couple fluxes capable of reproducing long-term continuous basin behavior
- Capture physical processes in Iowa watersheds
- Create robust algorithms
GHOST breaks the watershed down into manageable and representative numerical elements (40 acres for a HUC-8; 5–10 acres for a HUC-12). The model also quantifies the impact of existing and potential best management practices (BMPs).
USING GHOST TO PREDICT FLOODING IN IOWA
Each one of Iowa’s 99 counties has been impacted by flooding events that exceeded the state’s capacity to respond and that ultimately led to flood-related presidential disaster declarations (FRDD). In the last three decades, the total number of FRDD in Iowa counties has exceeded 900, making flooding one of the most prominent environmental challenges that Iowa faces.
IWA modelers used GHOST to evaluate the flood reduction benefits expected from both nature-based and structural mitigation strategies for all IWA watersheds. The model was calibrated and validated comparing simulated streamflow hydrographs against measurements taken at USGS stations. The model reproduced the measured hydrographs satisfactorily, as well as seasonal and annual trends. We determined model baseline conditions using a 15-year continuous simulation with hourly climatological data. We modified model parameters to simulate the implementation of cover crops and native vegetation (e.g., tall-grass prairie) in the study area. In addition, the watershed model evaluated the flood reduction benefits associated to a system of distributed storage built with ponds located in the watershed’s headwater catchments. Details of this hydrological modeling are available on the IWA website: