Recently Involved
Recently Involved
Embedded Files
Working on Publications
Working on Publications
Dharali Flood, Uttarakhand, India (August-2025)
Dharali Flood, Uttarakhand, India (August-2025)
The recent tragedy in Dharali was a devastating reminder that in the Himalayas, where we build is as important as what we build. The disaster was not simply a case of extreme weather but a predictable intersection of natural hazards and human vulnerability.
Settlement on Alluvial Plains: For centuries, communities have settled on alluvial fans—the fan-shaped deposits of sediment where mountain streams meet the main valley. While offering rare flat, fertile land, these fans are the historical pathways for debris flows. The recent event in Dharali, which swept away 20-30 houses and 20 hotels1, tragically highlighted this inherent vulnerability. By building on these plains, communities are placing themselves directly in nature's hazard lane.
Massive Geomorphic Change: The power of the debris flow was immense. It deposited a massive new alluvial fan that extended approximately 150 meters from the old perimeter, completely altering the landscape. This new deposit blocked the stream's natural course, creating secondary risks like damming and potential breach floods for downstream areas.
A Cascading Disaster: The severity of the initial cloudburst was exponentially amplified by a cascading event. Evidence suggests that a high-altitude glacial lake burst (a GLOF) coincided with the extreme rainfall. This released a catastrophic volume of additional water and sediment, transforming what would have been a severe flood into an unstoppable, high-density debris flow.
Understanding these interconnected risks is crucial. We must move beyond reactive relief and towards proactive, risk-informed planning and development, especially in fragile mountain ecosystems.
The study presents an assessment of water quality at the Massanjore Dam, located on the Mayurakshi River in Dumka district, Jharkhand.
Machine Learning based Landslide Induced Road Risk Dashboard showing Road risk with different classes and costing.
Assessing the Impact of Urban Green Cover on Land Surface Temperature Dynamics in Kolkata
Interactive Crime Analytics Dashboard Predictive Decision Support System (updating)
https://ee-souravdey.projects.earthengine.app/view/crimeviz (just for viewer's concept using dummy data)
Temporal Change Pre and Post Preparedness of Kumbh-Tribeni,Prayagraj
Tribeni Sangam (December, 2024)
Data Used: Sentinel-2
Tribeni Sangam (January, 2025) Data Used: Sentinel-2
Tribeni Sangam (December, 2024) Data Used: Sentinel-1
Tribeni Sangam (January, 2025) Data Used: Sentinel-1
During the Kumbh Mela 2025, satellite imagery captured the rapid transformation of Prayagraj. In December, prior to the event, baseline images showcased the natural terrain around the Triveni Sangam. By late January, significant changes were evident: thousands of temporary tents, new roads, and pontoon bridges had been constructed to accommodate millions of pilgrims. The combination of multispectral and Synthetic Aperture Radar (SAR) data provided near-real-time insights into this large-scale transformation, highlighting the effectiveness of geospatial technologies in documenting and managing such monumental events .
Earth Engine Data Catalog | Google for Developers
Tool to monitor Snow and Debris Cover of entire Asia
Tool to monitor Snow and Debris Cover of entire Asia
Explore snow cover (2000-2024) with this interactive tool designed for glaciologists, remote sensing specialists, and cryo-enthusiasts. Track seasonal snow accumulation across Asia, analyze glacier dynamics, and discover snow percentage and debris layer with statistics. MODIS derived threshold (Mityók et al., 2018).
Explore the application and utilize the leverage: snowscapeasia
https://souravdeyee.projects.earthengine.app/view/snowscapeasia
Energy Balance Dynamics of South Lhonak Glacier, Sikkim: A Time Series Analysis (1998-2024) using Google Earth Engine
Energy Balance Dynamics of South Lhonak Glacier, Sikkim: A Time Series Analysis (1998-2024) using Google Earth Engine
The energy balance in meteorological terms refers to the net exchange of energy fluxes between the Earth’s surface and the atmosphere. It governs the thermodynamics of surface systems, including glaciers, where radiative, conductive, and latent heat fluxes drive critical changes in surface temperature and ice melt.The energy balance of the South Lhonak Glacier in Sikkim plays a crucial role in determining its melt rates and mass balance fluctuations.
[figure01: Daily Max aggregated to Yearly Mean, figure02: Daily Min aggregated to Yearly Mean, figure03: Daily Mean aggregated to Yearly Mean]
Using ERA5-Land data and processed via Google Earth Engine, an analysis was conducted from 1998 to 2024, focusing on key variables such as net shortwave radiation (SW), net longwave radiation (LW), sensible heat flux (QH), latent heat flux (QE), and precipitation energy flux (QR). These components formed the basis of the surface energy balance equation:
QM= SWnet+LWnet+QH+QE+QR
(Garph figure unit 'Wm-2')
Shortwave radiation dominated the energy input, modulated by seasonal variations in solar insolation and albedo changes on the glacier surface. Longwave radiation acted as a significant negative flux, reflecting energy loss through thermal radiation, especially in cloudy or nighttime conditions. Sensible heat flux, driven by temperature gradients and boundary layer dynamics, added to the melting potential, while latent heat flux captured the role of sublimation and evaporation in the overall energy exchange. Precipitation energy flux, although smaller in magnitude, contributed during monsoonal periods.
This time series analysis provides essential insights into the glacier's thermodynamic responses to atmospheric forcing, aiding in more accurate predictions of glacier mass balance in a changing Himalayan climate.
Real-time Flood Monitoring of Ganga-Brahmaputra Basin
Real-time Flood Monitoring of Ganga-Brahmaputra Basin
Every monsoon season, the Ganga-Brahmaputra basin (Uttarakhand, Bihar, Assam, etc) faces severe flooding, causing widespread damage and disruption.
To help monitor and visualize the flood intensity and its affect to major socio-economical piller(Crop and Builtup), I've prepared a Earth Engine application. This tool not only shows real-time flood data but also provides crucial statistics and graphs on inundated crops and affected urban areas.
Explore the application and utilize the leverage : gbfm
Samudra tapu Glacier Lake(Chandra basin) expansion inducing the threat of occurring flood(GLOF) :
Samudra tapu Glacier Lake(Chandra basin) expansion inducing the threat of occurring flood(GLOF) :
Previously Worked on
Previously Worked on
Precipitation Trend Analysis
Precipitation Trend Analysis
Graphically visualizing the annual mean rainfall trend over India from 1980 to 2022 using CRU data highlights increasing trends and uncertainties in off-season precipitation, while seasonal rain intensity remains low. This shift impacts water resources, agriculture, and disaster management. Understanding these changes is crucial for developing strategies to adapt to evolving monsoon patterns.
Time series change of South Lohnak Glacier Lake, Sikkim
Time series change of South Lohnak Glacier Lake, Sikkim
The South Lhonak Glacier Lake in Sikkim has been expanding rapidly from 1984 to 2022, posing an increasing threat to local communities. On October 3, 2023, the lake breached, causing a devastating glacial lake outburst flood (GLOF) that resulted in significant loss of life and property. The lake, located at an altitude of 5,200 meters, has been a known hazard, with its area growing significantly over the decades due to the melting glacier. This event highlights the urgent need for monitoring and mitigation efforts to protect the vulnerable populations in the region from future disasters linked to climate change and glacial dynamics.
The South Lhonak Glacier Lake in Sikkim has been expanding rapidly from 1984 to 2022, posing an increasing threat to local communities. On October 3, 2023, the lake breached, causing a devastating glacial lake outburst flood (GLOF) that resulted in significant loss of life and property. The lake, located at an altitude of 5,200 meters, has been a known hazard, with its area growing significantly over the decades due to the melting glacier. This event highlights the urgent need for monitoring and mitigation efforts to protect the vulnerable populations in the region from future disasters linked to climate change and glacial dynamics.
Delineating Snow and Ice Using Principal Component Analysis
Delineating Snow and Ice Using Principal Component Analysis
Drung Drang Glacier, located in the Zanskar Valley, Indian Himalayas, is a significant glacier contributing to the region's hydrology and ecosystem.
Principal Component Analysis (PCA) is a statistical technique used to emphasize variation and bring out strong patterns in a dataset, transforming it into principal components for easier analysis. By analyzing PCA outputs (PCA1, PCA2, PCA3), we can effectively delineate and discriminate between glacier ice and debris based on their spectral signatures, aiding in more accurate mapping and study of glacial environments / inventory.
Classification of Snow, Ice, and Debris using SWIR Reflectance Spectra.
Classification of Snow, Ice, and Debris using SWIR Reflectance Spectra.
The Normalized Difference Snow Index (NDSI) is a powerful tool for remote sensing applications, particularly in environmental monitoring and natural disaster management. This technique leverages the unique properties of the green and short-wave InfraRed (SWIR) reflectance spectrums, which is sensitive to snow and water due to their absorption and reflection patterns. The green band (around 550-570 nm) is highly reflective for snow and ice, while the SWIR band (around 1.5-2.5 μm) is more absorptive for water. In contrast, the SWIR band is less affected by atmospheric scattering, making it ideal for masking cloud-covered pixels. By calculating the NDSI, pixels are discriminated into nine distinct classes. This innovative technique ensures accurate classification results by minimizing the impact of clouds and atmospheric interference. The SWIR band is used to mask cloud-covered pixels, allowing for a clear distinction between snow, ice, and debris. With this approach, researchers can improve our understanding of environmental changes and better respond to natural glacier such as shift in ELA, mass balance, debrish cover etc . The accuracy of NDSI-based classification has significant implications for various applications, including climate modeling, natural resource management, and emergency response planning. By harnessing the power of green and SWIR reflectance spectroscopy, we can gain valuable insights into our planet's dynamic environment.
The Normalized Difference Snow Index (NDSI) is a powerful tool for remote sensing applications, particularly in environmental monitoring and natural disaster management. This technique leverages the unique properties of the green and short-wave InfraRed (SWIR) reflectance spectrums, which is sensitive to snow and water due to their absorption and reflection patterns. The green band (around 550-570 nm) is highly reflective for snow and ice, while the SWIR band (around 1.5-2.5 μm) is more absorptive for water. In contrast, the SWIR band is less affected by atmospheric scattering, making it ideal for masking cloud-covered pixels. By calculating the NDSI, pixels are discriminated into nine distinct classes. This innovative technique ensures accurate classification results by minimizing the impact of clouds and atmospheric interference. The SWIR band is used to mask cloud-covered pixels, allowing for a clear distinction between snow, ice, and debris. With this approach, researchers can improve our understanding of environmental changes and better respond to natural glacier such as shift in ELA, mass balance, debrish cover etc . The accuracy of NDSI-based classification has significant implications for various applications, including climate modeling, natural resource management, and emergency response planning. By harnessing the power of green and SWIR reflectance spectroscopy, we can gain valuable insights into our planet's dynamic environment.
ForestFire Hotspot Severity Map, Nagaland, India
ForestFire Hotspot Severity Map, Nagaland, India
The Forest Fire Hotspot Severity Map for Nagaland, India, utilizes satellite-based inputs and FIRMS forest fire point data to identify and assess areas at high risk of wildfires.
This mapping is crucial for understanding fire distribution and intensity, aiding in effective resource allocation and fire mitigation. It supports researchers, policymakers, and conservationists in developing strategies to protect biodiversity and local communities. Ultimately, it enhances forest management and prevention efforts in the region.
Unraveling the Waterlogged Truth: A Sentinel-1 Synthetic Aperture Radar Analysis of Bihar's Ganga Floodplain
Using Sentinel-1 Synthetic Aperture Radar (SAR) data, we embarked on a mission to identify the permanent waterbodies and recurring waterlogging pockets in Bihar's Ganga floodplain. Through time-series analysis, we extracted the major areas that remain waterlogged throughout the monsoon season.
Our findings reveal a stark reality of persistent waterlogging in certain regions, warranting targeted interventions. The Sentinel-1 SAR data provided high-resolution insights into the flood-prone areas, enabling a more accurate assessment of the affected regions. By mapping these waterlogged pockets, we can now prioritize efforts to mitigate the impact of floods and improve flood resilience in Bihar.
Arsenic Contamination in Ganga Plain Leveraging EO and Field Inputs: Geological Explanation(Case Study).
Arsenic Contamination in Ganga Plain Leveraging EO and Field Inputs: Geological Explanation(Case Study).
During our M.Sc intermediate semester, our group conducted a field project titled "A Case Study of Site Suitability of Getalsud Dam and Study the Physical Factors of the Getalsud Reservoir." We visited the Getalsud Dam and reservoir in Jharkhand, collected remote sensing and GIS data, and validated our findings with ground information. This comprehensive analysis provided valuable insights into the geological suitability and management of the reservoir site.
A Case Study of Dam-Site Suitability
Spatial Wind Vectord(Wind Speed and Wind Direction) of the Region
Spatial and Hourly Temporal Variation of Temperature at 2 Meter
[Within the scope of the M.Sc. intermediate semester assignment, this study aimed to investigate the conditions conducive to fog formation by analyzing the most dense fog occurrence day, as recorded in news archives.]
Abstract:
This study analyzed the climatic variables influencing dense fog formation in the Indo-Gangetic Plains (IGP) region, focusing on Delhi and Haryana. MODIS Aerosol Optical Depth (AOD) data showed high presence of nuclei in December, coinciding with the peak winter season. The study used ERA5 hourly data for wind speed and direction, GIOVANNI temperature data, and NASA/POWER CERES/MERRA2 relative humidity data. R programming software was used to extract and visualize the data.
Key Findings
• High presence of atmospheric nuclei in December, a favorable condition for fog formation
• Near-constant or less fluctuating relative humidity during the winter season
• Diurnal temperature pattern with a decrease in temperature at night and an increase during the day
• Stable wind direction, which helps to precipitate atmospheric moisture as fog
Implications:
• Improved fog forecasting models can be developed using individual climatic variables
• Remote sensing data can be used to estimate fog-inducing factors without direct measurements
• Understanding the impact of climatic variables on fog formation can help reduce the effects of fog on transportation and daily activities
drafts......
Page updated
Google Sites
Report abuse