Environmental variables and occurrence points for predicting the distribution of Mythimna separata

This dataset includes occurrence records and environmental variables used for species distribution modeling of an agricultural pest. Occurrence data (presences.csv) were compiled from the Global Biodiversity Information Facility (GBIF) and filtered using a 10 km spatial thinning procedure to reduce sampling bias. Environmental predictors consist of bioclimatic variables (BIO1–BIO19) in ASCII format at a spatial resolution of ~18.5 km, representing baseline climatic conditions (1991–2021). These data were used to develop and compare mechanistic (CLIMEX), correlative (MaxEnt), and deep learning (DNN) models.