Crop yield forecasting before harvesting is critical for the creation, implementation, and optimization of policies related to food safety as well as for agro-product storage and marketing. Crop growth and development are influenced by the weather. Therefore, models using weather variables can provide reliable predictions of crop yields. It can be tough to select the best crop production forecasting model. Therefore, in this study, five alternative models, viz., stepwise multiple linear regression (SMLR), an artificial neural network (ANN), the least absolute shrinkage and selection operator (LASSO), an elastic net (ELNET), and ridge regression, were compared in order to discover the best model for rice yield prediction. The outputs from individual models were used to build ensemble models using the generalized linear model (GLM), random forest (RF), cubist and ELNET methods. For the previous 21 years, historical rice yield statistics and meteorological data were collected for three districts under three separate agro-climatic zones of Chhattisgarh, viz., Raipur in the Chhattisgarh plains, Surguja in the northern hills, and Bastar in the southern plateau. The models were calibrated using 80% of these datasets, and the remaining 20% was used for the validation of models. The present study concluded that for rice crop yield forecasting, the performance of the ANN was good for the Raipur (Rcal2 = 1, Rval2= 1 and RMSEcal = 0.002, RMSEval = 0.003) and Surguja (Rcal2 = 1, Rval2= 0.99 and RMSEcal = 0.004, RMSEval = 0.214) districts as compared to the other models, whereas for Bastar, ELNET (Rcal2 = 90, Rval2= 0.48) and LASSO (Rcal2 = 93, Rval2= 0.568) performed better. The performance of the ensemble model was better compared to the individual models. For Raipur and Surguja, the performance of all the ensemble methods was comparable, whereas for Bastar, random forest (RF) performed better, with R2 = 0.85 and 0.81 for calibration and validation, respectively, as compared to the GLM, cubist, and ELNET approach.

The prediction of crop yield before harvest is crucial for facilitating the formulation and implementation of policies about food safety, transportation cost, and import-export, storage and marketing of agro-products. The weather plays a crucial role in crop growth and development. Therefore, models using weather variables can provide reliable forecasts for crop yield and choosing the right model for crop production forecasts can be difficult. Therefore in the present study, an attempt was made to find the best model for wheat yield forecast by using five different techniques viz. Stepwise Multiple Linear Regression (SMLR), Artificial Neural Network (ANN), Least Absolute Shrinkage and Selection Operator (LASSO), Elastic Net (ELNET) and Ridge regression. Historical wheat yield data (taken from the Directorate of Economics and Statistics, Ministry of Agriculture and Farmers Welfare) and weather data of past 18-20 years were collected for seven different districts of Uttarakhand. Analysis was carried out by fixing 80% of the data for calibration and remaining dataset for validation. The present study concluded that the performance of ANN was good for crop yield forecasting as compared to the other models based on the value of RMSE (0.005 - 0.474) and nRMSE (0.166 - 26.171).

No abstract

Background: Climate change has become a major challenge in cultivation of chickpea and productivity. Negative impacts of climate change are likely to result from the effects of high temperature, low temperature, drought and excessive moisture and these factors affect crop yield ultimately. Methods: Keeping in view for quantifying the effects an experiment was laid out in split plot design. The experiment with three dates of sowing i.e. 12th December 2018 (D1), 22nd December 2019 (D2) and 2nd January 2019 (D3) as main plot treatments and the four microclimatic regimes viz. open field (T1), Open roof (T2), perforated roof (T3) and closed or packed (T4) by 100 GSM plastic film as sub plot treatments was laid to analyse the impact of temperature variation. Result: The major finding of the study is that the chickpea crop sown on 12th Dec. (D1) found highest grain yield (1200 kg ha-1) as compared to 22nd Dec. (845 kg ha-1) and 2nd Jan. (638 kg ha-1). This may be mainly attributed to congenial weather during the entire growing period. By studying the role of weather variables on chickpea in terms of seed yield, it is noticed that best performance of Packed subset (T4) i.e. 1044 kg ha-1 was observed in all dates of sowing followed by Perforated (T3) i.e. 955 kg ha-1, Open roof (T2) i.e. 813 kg ha-1 and Open field (T1) i.e. 765 kg ha-1.