An index for precipitation on the north coast of Peru using logistic regression

Abstract

The northern coast of Peru has a desert-like climate. Since precipitation is so scarce, convective rainfall events have a major impact. However, little is known about these events, and their prediction is complex. To date, anomalous convective activity has mainly been associated with warm sea surface temperature anomalies near the Peruvian coast. However, a more comprehensive analysis of atmospheric variables could shed light on how these precipitation events are triggered. To address this need, this study presents a new diagnostic index of precipitation using logistic regression. Satellite radar data are used as predictands, and ERA5 reanalysis parameters are used as predictors. The new index includes the mixing ratio and divergence at different levels (950, 700, and 250 hPa) and the Gálvez-Davison Index. This combination yields a logistic regression equation that ultimately takes the form of a new index, which we call RAMI (Rivas, Anderson-Frey, McMurdie Index). The RAMI is useful for diagnosing rainfall on the northern coast of Peru and could be useful for forecasting in this region, which is devoid of surface radars or other severe weather instruments.