Mismanagement of fresh water is a primary concern that negatively impacts agricultural productivity. Judicious use of water in agriculture is possible by estimating the optimal requirement. The present practice of estimating crop water requirements is using reference evapotranspiration (ET₀) values, which is considered a standard method. Hence, predicting ET₀ is vital in allocating and managing available resources. In this study, different machine learning (ML) algorithms, namely random forests (RF), extreme gradient boosting (XGB), and light gradient boosting (LGB), were optimized using the naturally inspired grey wolf optimizer (GWO) viz. GWORF, GWOXGB, and GWOLGB. The daily meteorological data of 10 locations falling under humid and sub-humid regions of India for different cross-validation stages were employed, using eighteen input scenarios. Besides, different empirical models were also compared with the ML models. The hybrid ML models were found superior in accurately predicting at all the stations than the conventional and empirical models. The reduction in the root mean square error (RMSE) from 0.919 to 0.812 mm/day in the humid region and 1.253 mm/day to 1.154 mm/day in the sub-humid region was seen in the least accurate model using the hyperparameter tuning. The RF models have improved their accuracies substantially using the GWO optimizer than LGB and XGB models.

The degradation of land and increasing water scarcity are existing challenges for agricultural sustainability, necessitating the implementation of improved soil-conservation practices at the watershed scale. The identification and selection of critical/prone areas based on erosion-governing criteria is essential and helps in the execution of the management process for determining priority. This study prioritizes erosion-prone sub-watersheds (alternatives) based on morphometric parameters (multiple criteria) via five Multi-Criteria Decision Analysis (MCDA) approaches, i.e., AHP: Analytical Hierarchy Process; TOPSIS: Technique for Order of Preference by Similarity to Ideal Solution; VIKOR: VIseKriterijumska Optimizacija I Kompromisno Resenje; SAW: Simple Additive Weighting; and CF: Compound Factor. Based on their priority score, 19 sub-watersheds were classified into four priority classes: low priority (0–0.25), moderate priority (0.25–0.50), high priority (0.50–0.75), and very high priority (0.75–1). The results revealed that about 8.34–30.15% area of the Burhanpur watershed is critically prone to erosion, followed by 23.38–52.05% area classed as high priority, 7.47–49.99% area classed as moderate priority, and 10.33–18.28% area classed as low priority. Additionally, four indices—percentage of changes (∆P), intensity of changes (∆I), the Spearman rank correlation coefficient test (SCCT), and the Kendall tau correlation coefficient test (KTCCT)—were employed to compare the models. This study confirms the efficacy of morphometric parameters for prioritizing sub-watersheds to preserve soil and the environment, particularly in areas for which limited information is available.

Precision irrigation with IoT-based decision-making technologies has proven effective in optimizing agricultural production irrigation. The Internet of Things (IoT) is crucial for monitoring the real-time data from sensors and automatically activating irrigation systems. This study evaluated the effectiveness of a drip irrigation system based on IoT and soil moisture sensors in a field experiment with sweet corn between 2020 and 2022. There were nine treatments with three replications: ETc-based drip irrigation (ETc 100%) and IoT-based drip irrigation scheduling with two soil moisture levels under three mulches: black plastic mulch, silver plastic mulch, and control (bare soil). IoT-based drip irrigation scheduling (100% FC) applied irrigation when soil moisture reached a lower threshold (≤ 33.1%) and ended when the field capacity was reached (≥ 43.5). With IoT-based drip irrigation scheduling (80% FC), irrigation was applied when the soil moisture content reached the threshold (≤ 33.1%) and ended when the field capacity reached 80% (≥ 34.8). Growth variables (root biomass, yield, corn length, cob weight, and water productivity) were compared for each irrigation method. Results showed that the ET-based irrigation method was easier to implement with less infrastructure and could result in lower yields than the IoT-based drip irrigation method with 100% FC. Grain and stalk yields increased by more than 12.05% and 14.97% for the IoT irrigation with 100% FC. It was found that IoT-based drip irrigation with 100% FC and 80% FC used 12.7% and 24.5% less irrigation water, respectively, and provided IoT-based drip irrigation with 100% FC, there was a 12.8% increase in marketable yield than ETc and IoT-based drip irrigation with 80% FC. The results show that the developed IoT system can potentially monitor the microclimate of plants in real time under different conditions of using plastic mulch. The IoT system is rugged and water-resistant, making it suitable for outdoor agriculture. Solar panels power the system, so there is no need for cabling and sensor nodes can be efficiently monitored. Research conducted on the IoT system shows that it can record and display environmental parameters to users via the cloud (ThingSpeak).