Catchment natural driving factors and prediction of baseflow index for Continental United States based on Random Forest technique

Baseflow plays a critical role in maintaining the aquatic environmental health. However, driving factors and predictions of baseflow have not been rigorously investigated on large scale, partly preventing hydrologist from deeply understanding runoff generation. To this end, Lyne–Hollick digital filter method automatic identification technique were used to estimate long-term seasonal index (BFI) 619 catchments across Continental United States (CONUS) 1981 2014. Six natural are selected 31 catchment attributes about topography location, soil, geology, land cover, climate characteristics. The Random Forest (RF) was predict BFI with six as predictors. Results show that average 0.49, value different four seasons, highest 0.55 winter lowest 0.46 autumn. forest fraction, clay proportion snow fraction most powerful affecting BFI. RF predicts sites CONUS R2 0.59 after Leave-One-Location cross-validation, which more satisfactory than multiple linear regression method. This study can provide deep insight into generation variation guide annual prediction for water resources management.