SHGR-Sugarcane-TSH

Description This dataset supports the research article "Semiparametric Heteroscedastic Gamma Regression for Sugarcane Tons of Sugar per Hectare (TSH) in India: A Distributional Modelling Framework Integrating Remote Sensing, Stress Indices, and Leaf Nutrient Status." The dataset was collected from sugarcane cultivation plots located in Sirsa, Masjidwa, and Matiaria villages of West Champaran district, Bihar, India. It is designed to support prediction of Tons of Sugar per Hectare (TSH) during different sugarcane growth stages using a semiparametric heteroscedastic gamma regression (SHGR) framework. The dataset integrates multiple sources of information, including: Agronomic data (village, cultivar, production cycle, plot information) TSH records (response variable) Landsat 8/9 derived Green Normalized Difference Vegetation Index (GNDVI) Daily precipitation and maximum temperature Abiotic stress indices (drought, excess rainfall, heat stress, soil degradation) Biotic stress indices (pest and disease pressure) Leaf nutrient measurements (Nitrogen, Phosphorus, Potassium) Nutrient balance index Chlorophyll and fluorescence measurements Microclimate observations Stomatal physiology measurements These variables were collected from field surveys, laboratory analysis, remote sensing products, meteorological records, and agricultural research stations to support distributional modelling of both the expected TSH and its variability. The dataset can be used for: Sugarcane yield and sugar productivity prediction Precision agriculture applications Remote sensing-based crop monitoring Abiotic and biotic stress assessment Statistical and machine learning research Reproducibility of the published SHGR model