Provide comprehensive support to undergraduate and postgraduate students in their independent research projects, including dissertations and thesis.Assisting with planning, preparing, and delivering teaching materials for practical classes and lectures at all levels of taught courses related to GIS and other fields.Provide GIS expertise and consultation to the wider School of Environment Education and Development research activities.

MSc Thesis on "Deep Learning Based Multimodal Fusion of Sentinel-1 and Sentinel-2 Data for Deforestation Detection in Amazon Rainforest"

Project Title: Deep Learning based wildfire prediction using time‑series climatic variablesDescription: Used ERA5 time series reanalysis data indexed using Uber’s open source hexagonal H3 indexing system as an input for LSTM‑based neural network for wildfire prediction.

Developed custom pipelines for processing Sentinel‑2 MSI data, MODIS NDVI data for agricultural and vegetation monitoring and Sentinel‑1 GRD data for flood mapping using open source tools (GDAL, python, QGIS, SNAP, PostgreSQL).Support and assistance are provided to scientists and interns with various GIS tools and techniques related to GIS and Remote Sensing problems, analysis, visualization, data handling, acquiring and managing, and data processing. This includes automating tools and processes using traditional GIS software and programming languages.

Project title: Dynamic Analysis of Time‑series Vegetation Indices in WebGIS Environment to improve crop ManagementDescription: Used signal smoothing technique (Modified Savitzky‑Golay filter and the Whittaker Smoother) on time‑series OCM and Proba‑V NDVI dataset for extracting vegetation phenology The majority of India’s population sustains agricultural activity. Here irrigation and multiple cropping agriculture is a crucial component for economic development and poverty alleviation. Therefore, information regarding crop phenology is essential for crop management. The study is done for the Gujrat district of India using the OCM NDVI dataset and Proba-V NDVI dataset for the agricultural year of 2016-2017. Two smoothing techniques have been used for smoothing the erroneous behaviour of NDVI time-series: Modified Savitzky-Golay filter and the Whittaker Smoother. The result obtained from these two techniques is then evaluated using AIC, BIC, and RMSE. And spatiotemporal analysis of the smoothed output annual vegetation phenology is extracted for each particular point on the study area.