CSIR National Institute of Data Science and AI

Climate change has direct effect on rice crop production, resulting in an impact on food security in tropical countries like India. The coastal state of Odisha, situated in the country's eastern region, also suffered from crop production/yield due to the regional climate change. In this research work, the multi-source long term data analysis is being carried out to quantify the climate-crop relation in the state of Odisha. The analysis of rainfall, temperature, Kharif crop (rice) yield and production for the past three decades in Odisha at district scale is carried out. The study based on the cluster analysis of different physiographical regions i.e. coastal plain (CP), river catchment (RC) and mountainous zone (MZ). The region wise climatology, interannual variability, and trend analysis of both rainfall and temperature and the rain/temperature days are computed using the data over 30 (1993–2022) years. The variability of annual and monsoon rainfall is comparatively higher in the CP and RC than in MZ. The monsoon rainfall trend seems to be decreasing in the CP region, increasing in the MZ and no trend over RC. The analysis infers a negligible (0.01 °C/year) temperature trend in the state. The correlation result suggested that the Kharif rice crop yield & production well correlated with the summer monsoon rainfall over RC and MZ at 90% confidence levels. The rice production in the state shows increasing trend in RC as compared to other two regions. The results suggest a good correlation between the rainfall and rice crop production observed in RC and MZ regions. Similarly, temperature also correlates better with the Kharif crop production and yield in CP as compared to other regions. The district scale analysis also indicated that strong correlation (95%) of rice crop yield with temperature is observed in 10 districts and the relation of crop production and rainfall is strong in 12 districts. The impact of large-scale process like ENSO on the district scale Kharif yield/production also assessed. The analysis signifies the global, local, and direct impact of climate patterns on rice production at district and regional scales. This study opens a scope for the potential use of the integrated approach to address the impact of climate change on a state's agriculture and food sector.

In this study, we present machine learning model support vector machine (SVM) and deep learning architectures, including long short-term memory (LSTM), gated recurrent unit (GRU), long- and short-term time series network (LSTNet), and temporal convolution network (TCN) to estimate leaf area index (LAI) utilizing Sentinel-2 multispectral imagery within the Kosi watershed located in the central Himalayan region is based on various vegetation indices. In situ LAI data were collected through a random sampling method using a plant canopy analyzer LI-COR over a continuous field survey period of 25 days for model training and validation. The current study evaluated four variants of the LSTM model, with the bidirectional LSTM emerging as the most effective in estimating LAI. The LAI derived using the Rendvi index was found to be superior compared to those obtained from other vegetation indices. The bidirectional LSTM (BiLSTM) model outperformed the SVM model in LAI retrieval based on all accuracy metrics computed. Specifically, the R² and mean absolute percentage error (MAPE) values for the BiLSTM model were 0.99 and 0.29, respectively, while the SVM model achieved corresponding values of 0.93 and 0.56. Additionally, the performance of the BiLSTM in predicting LAI surpassed that of other deep learning models such as GRU (R² = 0.99, MAPE = 0.82), LSTNet (R² = 0.99, MAPE = 0.74), and TCN (R² = 0.99, MAPE = 0.39). We compared the performance of the BiLSTM with the SVM model and the moderate-resolution imaging spectroradiometer (MODIS) LAI product by validating against in situ LAI measurements. The LAI retrieved using the BiLSTM model (R² = 0.99) demonstrated superior agreement with in situ measurements compared to the MODIS LAI product (R² = 0.0003). Furthermore, seasonal changes in vegetation and interannual variations in LAI over the Kosi watershed were more accurately depicted in the high-resolution LAI retrieved with BiLSTM from Sentinel-2 compared to the MODIS LAI product.