Coffee Leaf Rust Incidence Dataset

This dataset contains secondary, fully anonymized field observations on coffee leaf rust (Hemileia vastatrix) collected by the Coffee Research Institute (CRI) under the Kenya Agricultural and Livestock Research Organization (KALRO), with technical support from World Coffee Research (WCR). Data were compiled between 2018 and 2023 under established institutional data-sharing agreements. No personal or identifiable farmer information is included, and no direct human participation occurred; therefore, ethical approval was not required.

The dataset represents six major Arabica-producing counties in Kenya—Bungoma, Kericho, Kiambu, Kirinyaga, Murang’a, and Nyeri—capturing broad variation in altitude, microclimate, and management practices. This ecological diversity provides a robust foundation for modeling disease dynamics across smallholder coffee systems.

The primary response variable is coffee leaf rust incidence, recorded as a binary indicator denoting the presence (1) or absence (0) of visible infection at the plot level. Rust severity, measured as the percentage of leaf area affected, is also included for descriptive purposes.

Predictor variables encompass environmental, spatial, and agronomic factors known to influence disease development. These include daily relative humidity, daily temperature, precipitation, leaf-wetness duration, elevation, NDVI, coffee variety, plant age, shade percentage, fungicide use and application frequency, past outbreak history, lagged incidence from the preceding week, and distance to the nearest infected farm.

This dataset supports research on coffee disease ecology, climate–disease interactions, and the development of predictive models for early warning systems in smallholder coffee production.