Development of an Environmental Mitigation Database and Statistical Models for Predicting Likely FERC License Mitigation Requirements at Hydropower Projects

Licensing of hydropower projects in the US is often conducted with a great deal of uncertainty around potential environmental mitigation that may be required as license conditions. To reduce this uncertainty we reviewed over 300 FERC licenses to create a geospatial database of environmental mitigation requirements at projects across the US and used those data to developing statistical models that predict likely mitigation for any proposed or existing site based on project design, environmental conditions, species presence, and various landscape and geo-political characteristics. Data extraction was limited to licenses issued in the past 15 years based on the assumption that recently required mitigation is the better predictor of future mitigation requirements. The 309 licenses, all issued since 1998, included 442 hydropower plants. A 3-tiered hierarchical database was developed to group mitigation requirements into six Tier I categories (fish passage, hydrology, water quality, biodiversity, habitat, and recreation/cultural), 20 Tier II categories nested within Tier I, and over 120 Tier III categories. Logistic regression was used to predict the probability that a particular mitigation would be required at an existing or proposed site. Various combinations of predictor variables were tested to develop the best predictive models.