Sengupta et al. (2025) How seas whisper to snow: teleconnections drive spatio–temporal variability of snow cover in Western Himalayas

Identification

• Journal: Scientific Reports
• Year: 2025
• Date: 2025-10-06
• Authors: Shairik Sengupta, Rajarshi Das Bhowmik
• DOI: 10.1038/s41598-025-18606-6

Research Groups

• Interdisciplinary Center for Water Research, Indian Institute of Science, Bengaluru, India

Short Summary

This study investigates the spatio-temporal variability of snow cover (SC) in six Western Himalayan watersheds, identifying key local meteorological and large-scale oceanic-atmospheric teleconnection drivers influencing these variations across different seasons and timescales, providing crucial insights for season-ahead SC predictions.

Objective

• To investigate the spatiotemporal variation of snow cover (SC) for Western Himalayan basins.
• To understand the long-term and seasonal drivers of SC extent, and how local meteorological variables and climate variability influence the SC extent.

Study Configuration

• Spatial Scale: Six high-mountainous watersheds in the Western Himalayas (Nubra, Spiti, Chandra-Bhaga, Ravi, Beas, Baspa), ranging from 998 square kilometers to 12,393 square kilometers.
• Temporal Scale: Historical record from August 1989 to February 2025.

Methodology and Data

• Models used: Principal Component Analysis (PCA), Wavelet Coherence Analysis, Pearson/Spearman/Kendall correlation analysis, Composite analysis, Best Subset Regression.
• Data sources:
  • Satellite: Landsat (missions 4, 5, 7, 8, 9) Level 2, Collection 2 images (30 meter spatial resolution) for Normalized Difference Snow Index (NDSI).
  • Observation: Daily gridded precipitation (0.25° × 0.25°) and gridded temperature (1° × 1° maximum and minimum) from the India Meteorological Department (IMD).
  • Reanalysis/Indices: Monthly climate variability modes (oceanic-atmospheric indices) from NOAA repository, including Arctic Oscillation (AO), North Atlantic Oscillation (NAO), Pacific Decadal Oscillation (PDO), Southern Oscillation Index (SOI), NINO 1+2, NINO 3, NINO 3.4, NINO 4 (for El Niño Southern Oscillations), and Dipole Moment Index (DMI) for Indian Ocean Dipole.
  • Topography: Global Satellite Radar Topography Mission (SRTM) Digital Elevation Model (DEM) from USGS Earth Explorer.

Main Results

• Snow cover (SC) in the Western Himalayas exhibits significant spatio-temporal variability, with distinct geographical areas responding to separate local and remote factors, challenging the traditional assumption of uniform behavior.
• Three characteristics of SC extent were derived: Fractional SC (FSC), Fractional Temporary SC (TSC), and Principal Components (PC) of NDSI, which revealed spatial patterns of SC variability linked to basin topography (e.g., high altitude versus river channels, south-facing versus high altitudes, high ridges versus north-facing regions).
• Interannual Scale: FSC is coherent with NAO at a yearly frequency (4-6 year cycles). TSC is driven by Nino 1+2 at a biennial frequency. PC1 (SC difference between high altitude and river channels) is modulated by PDO at an annual frequency. PC2 (SC in south-facing regions versus high altitudes) is influenced by Nino 4 (4-5 year timescale). PC3 (SC in north-facing regions versus high ridges) is synchronized with NAO at a yearly frequency.
• Seasonal Scale:
  • Winter: PC1 (mid-altitudes) is primarily driven by NAO and PDO, with a slight AO contribution. PC2 (south-facing mid-altitudes) is modulated by Nino 4 and PDO. PC3 (lowest altitudes) is influenced by AO. DMI affects PC1 in higher/drier basins (Nubra, Chandrabhaga, Spiti) at 0-3 month lags, and PC2 in southern lower altitude basins (Ravi, Beas, Baspa) at 6-month lags.
  • Summer: PC1 (channel tributaries) is influenced by NAO and AO (lag 5). PC3 (south-facing slopes) is driven by PDO (lag 5). TSC is influenced by DMI (0-3 month lag) and SOI (higher lags).
  • Monsoon: Local precipitation and temperature are major drivers of PC1. PC2 and PC3 are driven by PDO, DMI, and Nino 1+2. Nino 3.4 drives various SC characteristics at approximately 8 months lag. AO and NAO also emerge as common drivers for PC3 across basins.
  • Fall: Local precipitation is a dominant driver for all PCs. PC1 is influenced by AO, PDO, and DMI. PC2 and PC3 are influenced by PDO and AO, respectively. Temperature does not consistently emerge as a driver.
• Composite analysis confirmed that the associations between SC and climate variability indices are amplified during periods of heightened activity of the relevant climate variability modes.

Contributions

• This is the first study to apply integrated Principal Component Analysis (PCA) with high-resolution satellite data to group areas by snow cover (SC) behavior in the Himalayas, revealing fundamental geomorphological units influencing SC variability.
• Introduced a novel metric, Fractional Temporary Snow Cover (TSC), to differentiate between persistent and temporary snow, highlighting its hydrological and ecological significance.
• Demonstrated that large-scale ocean-atmosphere couplings (teleconnections) are major drivers of SC extent in the Western Himalayas across both interannual and seasonal timescales, and that these influences are spatially varied within river basins.
• Provided critical insights for improving season-ahead SC predictions and water resource management strategies in the Western Himalayas.

Funding

• PhD stipend for Shairik Sengupta was provided by the people of India.

Citation

@article{Sengupta2025How,
  author = {Sengupta, Shairik and Bhowmik, Rajarshi Das},
  title = {How seas whisper to snow: teleconnections drive spatio–temporal variability of snow cover in Western Himalayas},
  journal = {Scientific Reports},
  year = {2025},
  doi = {10.1038/s41598-025-18606-6},
  url = {https://doi.org/10.1038/s41598-025-18606-6}
}