High Mountain Asian glacier response to climate revealed by multi-temporal satellite observations since the 1960s

High Mountain Asia (HMA) harbors the largest concentration of glaciers outside the polar regions, acting as a vital water tower for major Asian rivers. However, climate change threatens this water source, with projected temperature increases driving glacier recession and ultimately reducing runoff. While models project future glacier behavior, uncertainties remain due to gaps in our understanding of regional glacier responses to climate change. Long-term mass balance data are crucial for model calibration and validation, but in-situ measurements are scarce in HMA (only about 30 out of 80,000 glaciers have sufficient data). Geodetic methods using satellite data offer an alternative for large-scale assessment. Studies using US spy satellite imagery (Hexagon KH-9) have provided insights since the mid-1970s, revealing spatial heterogeneity in glacier mass changes and increased loss rates after 2000. However, limitations in image quality and data voids affect accuracy. Declassified Corona KH-4 satellite imagery from the 1960s offers high-resolution stereo data to extend the observation period. This study uses multi-temporal geodetic glacier mass budgets from various satellite sources (Corona KH-4, Hexagon KH-9, Pléiades, etc.) spanning nearly six decades across seven climatically diverse HMA regions to characterize mass budget variability and identify climate drivers using ERA5 Land and in-situ weather station data. The aim is to improve our understanding of HMA glacier response to climate change.

Existing literature highlights the importance of HMA glaciers as water towers and their vulnerability to climate change. Previous studies using in-situ glaciological methods have been limited by the scarcity of long-term data. Satellite-based geodetic methods, particularly using Hexagon KH-9 imagery, have improved our understanding of glacier mass changes since the mid-1970s, revealing spatial heterogeneity and accelerated mass loss after 2000. However, these studies had limitations in terms of data coverage, temporal resolution and accuracy. The 'Karakoram anomaly', a region exhibiting near-balanced mass budgets until recently, has been a focus of research. The use of declassified Corona KH-4 imagery provides a unique opportunity to extend the record back to the 1960s and improve the accuracy of mass balance estimates. Previous studies using Corona KH-4 data were limited to localized areas. This study builds upon these previous works by providing a comprehensive, multi-regional analysis spanning several decades, bridging the gap in our understanding of HMA glacier behavior.

The study analyzed multi-temporal geodetic glacier mass budgets in seven climatically distinct HMA regions using DEMs derived from Corona KH-4, Hexagon KH-9, Pléiades, ASTER, and TerraSAR-X satellite data. The methodology involved manual adjustment of glacier outlines using various image sources, DEM generation using different software packages (RSG, LPS, PCI Geomatica, Ames Stereo Pipeline, GAMMA), co-registration and outlier removal, and gap filling using a combination of moving window and median hypsometric methods. The impact of radar penetration in TanDEM-X data was addressed through comparisons with optical data. A seasonality correction was applied to account for differences in image acquisition dates. Uncertainty analysis was performed, incorporating uncertainties related to elevation changes, area changes, and ice density. Climate data were obtained from ERA5 Land reanalysis and in-situ weather stations. Correlation analysis was performed between glacier mass balance estimates and climate variables (summer temperature and solid precipitation) to identify climate drivers. The study also considered non-climatic factors, such as glacier surges and proglacial lake expansion, which may influence glacier mass balance.

The study revealed substantial glacier mass loss in all seven regions across HMA, although temporal variability existed. Several regions (Northern Tien Shan, Western Nyainqentanglha, Poiqu/Langtang) experienced consistently increasing mass loss rates. Regions previously showing near-balanced mass budgets (Muztagh Ata, Gurla Mandhata, Purogangri Ice Cap) transitioned to significant mass loss in recent years. The mean mass budget from 1964 to 2004 was -0.23 ± 0.10 m w.e.a⁻¹, approximately doubling to -0.50 ± 0.11 m w.e.a⁻¹ in the most recent period (2004-2019). Highest area loss rates were observed in Northern Tien Shan (-0.60 ± 0.07% a⁻¹) and Western Nyainqentanglha (-0.30 ± 0.02% a⁻¹). Analysis of ERA5 Land and in-situ weather station data indicated that increased summer temperature (SumT) anomalies are the primary driver of enhanced ice loss, although subtle decreases in solid precipitation (SolP) might have also contributed in some regions. Correlation analysis confirmed SumT as the main control on glacier mass budgets in Northern Tien Shan (r = 0.97, p = 0.009). In cold, dry regions, SolP showed a strong correlation (r² = 0.79, p = 0.0007) with glacier mass balance. In humid regions, both SumT and SolP significantly influenced glacier mass budgets. Regions showing a shift from balanced to negative budgets (Gurla Mandhata, Muztagh Ata) were highly sensitive to SumT increases, outweighing the influence of SolP. Comparisons with previous studies showed general consistency in contemporary mass balance estimates but highlighted the value of temporally resolved data for understanding the evolution of glacier mass loss. Non-climatic factors, such as glacier surges and proglacial lake expansion, also influenced mass balance variability.

The findings demonstrate that the pervasive increase in summer temperatures is the primary driver of accelerated glacier mass loss across HMA. The transition from near-balanced to negative mass budgets in some regions emphasizes the sensitivity of these glaciers to even modest temperature increases. The study highlights the limitations of relying solely on climate variables to explain mass balance variability, particularly in regions affected by glacier surges or proglacial lake expansion. The study's extensive temporal coverage (six decades) reveals the long-term trend of accelerating mass loss and underscores the need for continuous monitoring to improve projections of future glacier behavior and its implications for water resources in Asia. The strong correlation between summer temperatures and mass loss suggests that even modest future warming will significantly impact HMA glaciers.

This study, using a multi-decadal, multi-source satellite data set, reveals a widespread and accelerating loss of glacier mass across diverse High Mountain Asian regions. Increased summer temperatures are the dominant driver of this loss, even surpassing the effects of precipitation changes in several regions. The transition of previously near-balanced glacier mass budgets to negative mass budgets in recent years underscores the urgency of monitoring and modelling efforts. Future research should focus on improving high-elevation precipitation measurements, refining models that incorporate non-climatic factors affecting glacier mass balance (e.g., surges and glacial lake expansion), and strengthening the integration of in-situ and remote sensing data for more comprehensive analysis.

The study's reliance on satellite data introduces uncertainties related to image quality, data voids, and the accuracy of DEM generation. In-situ climate data were limited in spatial extent and temporal coverage, affecting the precision of the climate-mass balance correlations. The analysis did not fully quantify the combined impact of all non-climatic factors affecting glacier behavior, which could vary regionally and temporally. The use of different software and methods for DEM generation from different datasets, while rigorously addressed, still introduces potential uncertainties.