Mr. Suman Kumar Saha will present his MTP2 as per the details below:

Date: 23rd June 2026

Time: 1100 – 1200 hrs.

Venue: C-TARA Conference Room No.1

Topic: Development of a remote sensing data-driven AI tool for identifying alternative crops for marginalized farmers and increasing livelihood.

Guide: Prof. Pennan Chinnasamy           

Examiners: Prof. Chaaruchandra Korde, Prof. Raaj Ramsankaran

Abstract:

Agriculture is increasingly facing several water-related challenges that include water shortage, climate variability, and inefficiencies in water use. Small and marginal farmers have been the hardest hit by these problems as many cannot afford the latest technology. Improvement in the CWP has been identified as one of the important factors required to achieve climate-smart agriculture.

This study attempts to examine the viability of using remote sensing, machine learning, and digital agriculture in improving water-use efficiency in crop production. It has adopted a multidisciplinary approach involving a literature review, field studies, stakeholder consultation, data modelling, and website development. The field study was carried out in Datli and Musalgaon villages located in the Nashik district, and the main objective was to examine farmers' level of digital access, online information-seeking behaviour, and awareness regarding technological advancements in the field of agriculture. In order to predict the CWP value, climatic factors as well as environmental factors were considered in the process of creating machine learning models. In addition, data augmentation strategies were used to make sure that the models were robust. Various machine learning models were considered, such as Linear Regression, Random Forest, Support Vector Machine, Extreme Gradient Boosting, Bagged Trees, and Gaussian Process Regression. The findings suggest the possibility of developing reliable CWP prediction models using machine learning algorithms. For the implementation of research outcomes, a web portal called SmartFarm was developed. It combines CWP prediction, weather forecasts, prices at mandis, water source monitoring, farmers' records, task management and welfare schemes into one platform for the benefit of both farmers and extension workers.

In conclusion, the research illustrates the potential for the integration of remote sensing technology, machine learning and digital advisory systems in improving the water-use efficiency and climate change resilience, thus fostering the sustainable development of agriculture.

Keywords: Crop Water Productivity (CWP), Remote Sensing, Machine Learning, Digital Agriculture, SmartFarm, Water Use Efficiency, Climate-Resilient Agriculture.