The research team, led by Abhinav Rai and Pankaj Kumar of the Delhi School of Economics at the University of Delhi, together with colleagues at Miranda House, examined five key cryo-climatic parameters across the Chandra Basin in Himachal Pradesh: temperature, total precipitation, rainfall, snowfall, and snow depth. Their dataset came from ERA5-Land, a high-resolution reanalysis product from the Copernicus Climate Change Service that provides monthly land-surface variables at roughly 9-kilometer resolution. Covering the period from 1950 to 2024, the record allowed the researchers to track changes at monthly, seasonal, and annual timescales, both across the basin as a whole and grid cell by grid cell.

The Chandra Basin is no ordinary mountain catchment. Spanning roughly 2,449 square kilometers, with elevations ranging from about 2,800 to 6,550 meters, it feeds the Chandra River, which joins the Bhaga at Tandi to form the Chandrabhaga, later becoming the Chenab, the largest tributary of the Indus system. Approximately 80 percent of the Indus River’s discharge originates from glacier ice and seasonal snowmelt, and peak discharge in the Chandra can reach 260.7 cubic meters per second during the ablation season. Drinking water, irrigation, and hydropower for communities across the region all depend on the timing and reliability of this meltwater, making the basin’s frozen stores a matter of direct economic and social consequence.

Because high-altitude weather stations are scarce in the Himalaya, the researchers first had to establish that the reanalysis data could be trusted. They compared ERA5-Land temperature estimates against four automatic weather stations operated within the basin by the National Centre for Polar and Ocean Research, at Himansh, Gepang Gath, Sutri Dhaka, and Bara Shigri. The temporal correlations exceeded 0.90 at all four stations. The dataset showed a systematic cold bias, but because the study’s primary statistical tools are rank-based, such uniform bias does not distort trend detection. Cross-checks against CRU, NASA-GLDAS, IMD, APHRODITE, and GPM-IMERG datasets confirmed consistent warming directions, and snow depth variability matched MODIS satellite observations of snow-cover area.

The statistical framework itself was deliberately multi-pronged. The Mann-Kendall test, a non-parametric workhorse of trend detection, identified the presence and direction of monotonic trends, while Sen’s slope estimator quantified their magnitude robustly against outliers. The Innovative Trend Analysis method, which plots the first half of a time series against the second and compares the result to a no-trend diagonal, revealed sub-trends hidden within the aggregate statistics. Finally, the Sequential Mann-Kendall test located change points by intersecting forward and backward trend series, pinpointing when the basin’s climate regime actually began to shift.

The headline finding is unambiguous warming. Annual mean temperature in the basin has risen at approximately 0.015 degrees Celsius per year, equivalent to about 0.15 degrees per decade over the study period. Seasonally, the monsoon months warmed fastest at 0.02 degrees per year, followed by winter and the pre-monsoon season. July, November, and June saw the most intense monthly warming. More striking still, the Sequential Mann-Kendall analysis identified a change point around 1985 for annual temperature, after which the warming trajectory steepened and became statistically significant from roughly 1990 onward. For the monsoon season, the inflection came around 1988, with the forward trend series crossing the 95 percent confidence limit in the final two decades of the record, a signal the authors describe as a fundamental shift in the basin’s thermal regime.

Beneath the warming signal lies a more consequential transformation: a phase shift in precipitation itself. Annual rainfall has climbed significantly at about 1.15 millimeters per year, with significant increases recorded in seven individual months, including the peak winter months of January and February, when precipitation historically arrived as snow. Meanwhile, annual snowfall has declined substantially at 3.0 millimeters per year, with significant losses in the pre-monsoon and post-monsoon seasons. The snow fraction, the proportion of total precipitation falling as snow, is eroding at 0.12 percent per year, a cumulative loss of roughly nine percentage points over the 74-year record. Snowfall that once made up 70 to 90 percent of total precipitation is steadily ceding ground to rain.

The consequences are already visible in the snowpack. Mean snow depth is falling at 4.2 millimeters per year during the critical pre-monsoon season of March through May, when snow reserves act as a natural reservoir that releases water gradually into the growing season. Sharp declines in April, May, June, and July point to an earlier onset and faster pace of spring melt. A change point around 1995, with the decline becoming statistically significant after about 2010, indicates that snowpack loss has accelerated in recent decades. Notably, even in June, where snowfall showed a localized significant increase, snow depth still declined, underscoring that warming temperatures, not precipitation inputs, now dominate the fate of the snowpack.

Spatially, the changes are far from uniform. Grid-wise analysis reveals that the eastern, higher-elevation, glacierized portions of the basin are experiencing significant drying, precisely where the glaciers that nourish the river are concentrated. In contrast, monsoon rainfall has increased in the lower-elevation western and central areas. This elevation-dependent split suggests that warming is pushing the freezing level upward, so that orographic precipitation increasingly falls as rain at lower elevations while the high ice-covered zones receive less replenishment. The authors link the declining snowfall to weakening winter western disturbances, which supply 60 to 70 percent of the basin’s annual snowfall, and note that mechanisms such as snow-albedo feedback, in which shrinking reflective snow cover amplifies absorption of solar radiation, may further accelerate the warming.

For the millions of people who depend on the Indus system, the implications are sobering. Earlier and more rapid snowmelt shifts river peak flows toward spring, potentially leaving less water during late summer and autumn, when agriculture and hydropower need it most. Persistent declines in snowfall and snow depth prevent glaciers from accumulating mass, driving thinning and negative mass balance, while early exposure of underlying ice to solar radiation speeds retreat further. The study’s authors caution that reanalysis products carry inherent uncertainties in complex terrain and that in-situ records remain limited, but the consistency of trend directions across multiple independent datasets and methods lends strong confidence to their conclusions. Their message is clear: the Chandra Basin’s cryosphere is degrading on multiple fronts simultaneously, and proactive water resource management and adaptation strategies are urgently needed before the water tower of the western Himalaya loses its capacity to store and deliver the meltwater on which so many depend.

Subject of Research: Long-term spatiotemporal trends in temperature, precipitation, snowfall, and snow depth in the high-altitude Chandra Basin of the western Himalaya

Article Title: Spatiotemporal trends of cryo-climatic parameters in the Chandra Basin, Himachal Pradesh, India

Article References: Rai, A., Kumar, P., Singh, S., & Rai, A. K. (2026). Spatiotemporal trends of cryo-climatic parameters in the Chandra Basin, Himachal Pradesh, India. Discover Geoscience, 4(1), Article 310. https://doi.org/10.1007/s44288-026-00681-5

Image Credits: AI Generated

DOI: 10.1007/s44288-026-00681-5

Keywords: Himalayan cryosphere, Chandra Basin, climate change, snowfall decline, precipitation phase shift, ERA5-Land, Mann-Kendall test, snow depth, glacier retreat, water security, western Himalaya, trend analysis