Machine Learning  Modelling of the Relationship between Weather and Paddy Yield in Sri Lanka

Abstract

This paper presents the development of crop-weather models for the paddy yield in Sri Lanka based on nine weather indices, namely, rainfall, relative humidity (minimum and maximum), temperature (minimum and maximum), wind speed (morning and evening), evaporation, and sunshine hours. The statistics of seven geographical regions, which contribute to about two-thirds of the country’s total paddy production, were used for this study. The significance of the weather indices on the paddy yield was explored by employing Random Forest (RF) and the variable importance of each of them was determined. Pearson’s correlation and Spearman’s correlation were used to identify the behavior of correlation in a positive or negative direction. Further, the pairwise correlation among the weather indices was examined. The results indicate that the minimum relative humidity and the maximum temperature during the paddy cultivation period are the most influential weather indices. Moreover, RF was used to develop a paddy yield prediction model and four more techniques, namely, Power Regression (PR), Multiple Linear Regression (MLR) with stepwise selection, forward (step-up) selection, and backward (step-down) elimination, were used to benchmark the performance of the machine learning technique. Their performances were compared in terms of the Root Mean Squared Error (RMSE), Correlation Coefficient (R), Mean Absolute Error (MAE), and the Mean Absolute Percentage Error (MAPE). As per the results, RF is a reliable and accurate model for the prediction of paddy yield in Sri Lanka, demonstrating a very high R of 0.99 and the least MAPE of 1.4%.

Two-Stage Spatiotemporal Time Series Modelling Approach for Rice Yield Prediction & Advanced Agroecosystem Management

“…In general, nonlinearity or mixture of linear and nonlinear pattern in rice yield can be caused due to (i) different cropping management strategies (fertilization, pest and/or weed management, irrigation), (ii) agroecological (soil type, therein development, exposition) and/or (iii) climatological (air or soil temperature, air humidity, solar radiation, total precipitation) conditions [42][43][44]. Machine learning techniques have been used widely to develop crop-environmental models, and some of them highlighted the accuracy of some approaches (e.g., Random Forest regression) while attributing their superiority in handling data to multicollinearity and extracting nonlinear interactions [44]. Thus, agronomists, agrometeorologists, crop data modelers and statisticians have attempted to quantitatively assess the effects of various biotic and/or abiotic impacts on crop (rice) development rate [44,45].…”

“…Machine learning techniques have been used widely to develop crop-environmental models, and some of them highlighted the accuracy of some approaches (e.g., Random Forest regression) while attributing their superiority in handling data to multicollinearity and extracting nonlinear interactions [44]. Thus, agronomists, agrometeorologists, crop data modelers and statisticians have attempted to quantitatively assess the effects of various biotic and/or abiotic impacts on crop (rice) development rate [44,45]. For instance, rice yield can be defined as a function of weed density and the duration of the weed-crop interference, where relation between weed density and crop yield is probably caused by the availability of solar radiation, phytonutrients, and/or water.…”

Multi-Objective Optimization with Artificial Neural Network Based Robust Paddy Yield Prediction Model

“…Accuracy of each generated rule was measured. Ekanayake et al [10] designed a machine learning model that predicts the impact in paddy cultivation. The author studies the correlation of parameters such as maximum and minimum temperature, humidity, and wind speed using regression techniques such as power, multiple, stepwise forward selection, and stepwise backward elimination methods.…”

Hybrid model in machine learning–robust regression applied for sustainability agriculture and food security

“…The sustainability agriculture and food security are more reliable, capable, and help to boost productivity [66]. Random forest has shown a reliable and accurate model to predict paddy showing a very high accuracy, which is aimed for sustainability agricultural and food security [67]. The random forest-Hampel's provides the most relevant data of the result which applied for sustainability agriculture and food security.…”