Karuppasamy Sudalaimuthu

Assessment of yield-limiting factors in paddy production should be carried out properly in order to improve productivity under varying agro-climatic conditions. This research paper introduces a combined geospatial and machine learning system to estimate stage-based paddy yield stress during Kuruvai 2023 in the Thanjavur region using vegetation indices and agrometeorological variables. Five different satellite-derived vegetation indices were calibrated using spectroradiometer data during the vegetative, reproductive, and ripening stages. The field- and satellite-derived indices showed good agreement in calibration (R²=0.85-0.92). Correlation analysis showed that there were significant associations between rainfall_vegetation, humidity_reproductive, and yield variability, whereas the canopy vigor represented by ripening_EVI, SAVI correlated positively with grain filling and final productivity. Yield prediction before and after calibration was performed using machine learning models such as Multiple Linear Regression (MLR), K-Nearest Neighbors (KNN), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost). Among these, XGBoost resulted in high performance after calibration (R²=0.908, RMSE=179.728 kg/ha, MAE=143.487 kg/ha). Feature importance revealed that reproductive-stage humidity, rainfall, and ripening-stage soil moisture were the most important yield drivers. This model can be used to provide a powerful decision-support tool for spatial stress mapping, accurate irrigation regulation, and climate-resilient crop strategies in irrigated rice environments.

Seawater intrusion into groundwater aquifers is a growing global concern due to factors such as groundwater exploitation, rising sea levels, and climate change. This study focuses on the Chennai south coastal region and incorporates an urban sprawl perspective to delineate vulnerable zones of seawater intrusion into littoral aquifers. The study uses land use/land cover (LULC) classification to measure the expansion of impervious surfaces in the region from 2000 to 2020. The GALDIT-B model is modified and performed using the Geographic Information System (GIS) based -Analytical Heirarchy Process (AHP) to assign weights and ratings to each factor. The resulting model identifies five categories of vulnerable zones, ranging from very high to very low, based on factors such as groundwater level and the impact of existing seawater intrusion. The study found that the coastal and deltaic geomorphological formations in the region, which consist of clayey and sandy aquifer regions with built-up areas, are the most vulnerable to saltwater intrusion. The vulnerability of the very high vulnerable zone increased by 6% between pre-monsoon 2000 and pre-monsoon 2010 and by 16% between pre-monsoon 2010 and pre-monsoon 2020. The model was cross-validated using major hydrochemical parameters and the hydrochemical facies evolution (HFE) diagram of groundwater elements. The highest concentration of NaCl was found in the extremely vulnerable zones. The Piper trilinear diagram was used to identify the presence of saltwater in coastal aquifers. The modified GALDIT-B model is a useful tool for decision-makers involved in the development of coastal urbanization.

Spectral indices-based soil prediction models derived from multispectral datasets are too intricate in terms of accuracy as well as resolution. Complications arise while incorporating multispectral datasets for regional-scale spatial assessment of soil macronutrients. Sporadically satellite image fusion techniques have been used for soil nutrient interpolation to circumvent the complications. The fusion of multispectral bands encompasses precise soil information that cannot be observed as accurate with single satellite dataset. In this study, fusion of near infrared regions of Landsat 8 Operational Land Imager and Sentinel 2 has been observed for its contribution on soil macronutrient assessments. Area-to-point regression Kriging (ATPRK) approach is followed in fusing the two satellite imagery and in situ soil spectral have used for the validation of the resultant. Comparative statistical analysis on Landsat 8 OLI band 5 (wavelength: 845–885 nm), Sentine-2 band 8,8A (wavelength: 785–900 nm) datasets and fused satellite bands provides R² values of 0.8209, 0.8436, and 0.8763 respectively. Regression models y = (0.25006 ± 0.00754) + (0.0000313)x, y = (0.25252 ± 0.0062) + (0.0000810)x, and y = (0.23715 ± 0.0062) + (0.0001210)x for nitrogen, phosphorus, and potassium respectively aids for soil macronutrient interpolation and assessments. Computations reveals the ranges of nitrogen, phosphorus and potassium that floats from 48 to 295 kg/ha, 5.0 to 37 kg/ha, and 32 to 455 kg/ha in the study area. Fusion of satellite imagery by ATPRK approaches in soil macronutrient study at regional scale brings the novelty of the study.

This study incorporates the Groundwater Quality Index (GQI) to alter the conventional groundwater potential zoning to quantity-quality potential zoning in the Chennai coastal region, which is experiencing groundwater decline and quality deterioration. Using Analytical Hierarchy Process (AHP), weights for twelve thematic layers were assigned and are superimposed with a weighted overlay technique. Pre- and post-monsoon groundwater quantity-quality potential zone (GQQPZ) are classified into poor, moderate, good, and very good zone using pre-and post-monsoon GQI, respectively. Furthermore, the study revealed that seasonal changes also influence the quantity-quality potential zone of the study area. According to the study, the poor zone areas changed from 3.58% to 0.94% due to reduced concentration of pollutants and increased quantity of groundwater during monsoon rain. The validation is carried out with pre and post-monsoon Electrical Conductivity (EC) for pre and post-GQQPZ and resulted in an accuracy of 88.5% and 89.3%, respectively.

Meticulous analysis on intertwined interaction among the soil nutrients and microbial communities brings fruitful outcomes on horticulture. This study focuses on identifying well compatible bacterial genera in enhancing soil primary macro-nutrients for sustainable vegetable crops cultivation. The biochemical tests were executed for identification of bacterial genera. Eventually, mathematical models among available NPK nutrients (Nitrogen, Phosphorus and Potassium) and identified bacterial genera were derived to determine most suited bacterial genus for nutrients inhibition. This Study reveals the present nutrient’s status of soils at Anaimalai block covering 12,424 ha of Coimbatore district in Tamil Nadu. Seven utilitarian bacterial genera were identified which inhibit the plant nutrients. Among them, Azotobacter, Arthrobacter and Achromobacter actively inhibit available NPK in the soil. Present study of correlating soil nutrients with bacterial components enriches successful conservation of biosphere through adopting these innovative technologies in horticulture.