Predicting chlorophyll and nitrogen content in rice using multiple regression models

Chlorophyll and nitrogen (N) content in rice at critical crop growth stages are important input parameters for crop growth and nutrient uptake models used to optimize N application. Hence, the study was conducted to predict chlorophyll and N content in rice using multiple regression-based models involving normalized difference vegetation index (NDVI), Soil Plant Analysis Development (SPAD), Customized leaf color chart (CLCC), chlorophyll and N content. An experiment was conducted with six varieties i.e. CR Dhan 312, CR Dhan 310, Lalat, Shatabdi, Swarna Shreya, CR Dhan 206 with four N levels (i.e. 0, 80, 100 and 120 kg N ha-1) in split plot design and replicated thrice. The NDVI, SPAD and CLCC readings were taken at 45, 49 and 54 days after transplanting (DAT). Plant sampling was done for estimating the chlorophyll and nitrogen content in laboratory following standard methodology. Multiple linear regressions were performed between NDVI, SPAD and CLCC readings and chlorophyll & N content in rice in different variables combinations. Multiple linear regression (MLR) models with two variables i.e. NDVI and SPAD predicted the chlorophyll and N content of rice varieties. The predicted values for nitrogen and chlorophyll were comparable to those obtained using models from individual varieties when combined data from six different varieties were used to create MLR models. The study concludes that variations in chlorophyll and N concentrations in rice can be estimated using data from optical sensors with notable accuracies.