Chauhan et al. (2025) Snow cover analysis using NDSI and SWI indices in Pindari-Kafni Glacier valleys, Kumaon Himalaya

Identification

• Journal: Applied Geomatics
• Year: 2025
• Date: 2025-11-18
• Authors: Pankaj Chauhan, Ram L. Ray, S. Samanta, Dharmaveer Singh, Rajib Shaw, Nirmal Kumar
• DOI: 10.1007/s12518-025-00667-x

Research Groups

• Wadia Institute of Himalayan Geology, Dehradun, India
• Cooperative Agricultural Research Center, College of Agriculture, Food, and Natural Resources, Prairie View A&M University, Prairie View, TX, USA
• Symbiosis International (Deemed University), Pune, Maharashtra, India
• Graduate School of Media and Governance, Keio University, Kanagawa, Japan

Short Summary

This study analyzed seasonal snow cover area (SCA) dynamics in the Pindari and Kafni glacier valleys, Kumaon Himalaya, over two decades using Landsat imagery and comparing NDSI and the newly applied SWI, revealing an overall increasing SCA trend and superior performance of SWI in cloud and water discrimination.

Objective

• To quantify the snow cover area (SCA) and map its temporal distribution monthly and annually.
• To compare the analysis of Normalized Difference Snow Index (NDSI) and Snow Water Index (SWI) techniques for assessing SCA using satellite imagery and field investigation.
• To compare the analysis of SCA from Pindari and Kafni glacier valleys in Kumaon Himalaya.
• To understand the trend and magnitude of snow mass and heterogeneity of the SCA.

Study Configuration

• Spatial Scale: Pindari and Kafni glacier valleys, Kumaon Himalaya, India. Pindari Glacier valley: approximately 111 square kilometers; Kafni Glacier valley: approximately 62 square kilometers. Total study area: approximately 173 square kilometers, covering a catchment from 2500 meters to 6200 meters above sea level.
• Temporal Scale: Accumulation periods (November–December and January–April) for the years 2008–2009, 2015–2016, and 2021–2022. Meteorological data spanned 2008–2022.

Methodology and Data

• Models used:
  • Normalized Difference Snow Index (NDSI)
  • Snow Water Index (SWI)
• Data sources:
  • Satellite imagery: 63 multispectral images from Landsat-5 (TM) and Landsat-8 (OLI) (path/row: 145/039) from USGS Earth Explorer.
  • Reanalysis data: ERA5-land (for rainfall, temperature, and wind speed), ERA-5 (for snow cover fraction).
  • Modeled data: MODIS-modeled snow cover fraction.
  • Topographic data: Shuttle Radar Topography Mission (SRTM) at 30 meters spatial resolution.
  • Validation data: Field-based observations (34 Ground Control Points collected with a Garmin eTrex32 GPS with 3 meters accuracy), high-resolution Google Earth imagery.

Main Results

• The overall accuracy for NDSI was 70%, and for SWI, it was 72%, indicating SWI's marginal superiority, particularly in managing cloud contaminants and water-snow discrimination.
• Year-wise SCA exhibited an increasing trend: 1.3 times higher in 2021–2022 and 1.2 times higher in 2015–2016 compared to 2008–2009.
• Both NDSI and SWI analyses generally showed minimum SCA in December and maximum SCA in April, with a notable exception in 2021–2022, which recorded minimum SCA (14.49%) in April and maximum SCA (20.52%) in January.
• The seasonal sum of SCA using NDSI increased from 461.17 square kilometers (2008–2009) to 491.87 square kilometers (2015–2016) and 601.01 square kilometers (2021–2022).
• The seasonal sum of SCA using SWI increased from 291.34 square kilometers (2008–2009) to 301.23 square kilometers (2015–2016) and 364.05 square kilometers (2021–2022).
• Meteorological data revealed considerable variation in precipitation and temperature patterns, with mean temperatures below -5 degrees Celsius and lowest temperatures below -15 degrees Celsius in December, January, and February.
• The observed increasing SCA trend is attributed to a combination of climatic (e.g., increased precipitation, Western Disturbances) and non-climatic factors (e.g., elevation, topography) and orographic effects.
• Mid-elevation areas (3000–4000 meters altitude) experienced an increase in SCA during the pre-monsoon season (March to May).

Contributions

• This study marks the first application and comparison of the Snow Water Index (SWI) with NDSI for snow cover area (SCA) estimation in the Pindari-Kafni region of the Kumaon Himalaya.
• It demonstrates SWI's superior performance over NDSI in managing cloud contaminants and water-snow discrimination in complex mountainous terrain.
• The research provides new insights into seasonal and inter-annual SCA dynamics in the central Himalayan region, highlighting an overall increasing trend despite rapid snowpack melting influenced by heterogeneity, atmospheric dynamics, and Rain-on-Snow (ROS) events.
• The findings contribute to a better understanding of glacier mass balance, hydrological processes, and climate change impacts in the Himalayas, with significant implications for regional water resource management.

Funding

• Science and Engineering Research Board (SERB), Department of Science and Technology, Government of India (File No. SIR/2022/000972).

Citation

@article{Chauhan2025Snow,
  author = {Chauhan, Pankaj and Ray, Ram L. and Samanta, S. and Singh, Dharmaveer and Shaw, Rajib and Kumar, Nirmal},
  title = {Snow cover analysis using NDSI and SWI indices in Pindari-Kafni Glacier valleys, Kumaon Himalaya},
  journal = {Applied Geomatics},
  year = {2025},
  doi = {10.1007/s12518-025-00667-x},
  url = {https://doi.org/10.1007/s12518-025-00667-x}
}