Examining spatiotemporal trends of drought in the conterminous United States using self-organizing maps

Droughts are a natural, recurrent climate extreme that can inflict long-lasting devastation on natural ecosystems and socio-economic sectors. Unlike other hazards, drought onset is insidious often affects greater spatial extent over prolonged temporal scale. In the United States evolution of its impacts typically region-specific intensified precipitation variability may obscure how manifest. this study, we examine spatiotemporal trends using self-organizing maps (SOM), competitive learning subset artificial neural networks (ANN), requiring unsupervised training inputs. We introduced monthly Palmer Drought Severity Index (PDSI) values to SOM identify existing clusters wetting drying patterns from 1895 2016. After training, created cartographic visualizations output conducted subsequent time-series analysis link with geographic drought. Over last 40 years, in Northeast, Midwest, upper Great Plains across several nodes. Across majority patterns, identified no significant changes or century for part CONUS.