Glacial responses to climate change exhibit considerable heterogeneity. Although global glaciers are generally thinning and retreat, glaciers in the Karakoram region are distinct in their surging or advancing, exhibiting nearly zero or positive mass balance-a phenomenon known as the Karakoram Anomaly. This anomaly has sparked significant scientific interest, prompting extensive research into glacier anomalies. However, the dynamics of the Karakoram anomaly, particularly its evolution and persistence, remain insufficiently explored. In this study, we employed Landsat reflectance data and Moderate Resolution Imaging Spectroradiometer (MODIS) MCD43A3 albedo products to developed high-resolution albedo retrieval models using two machine learning (ML) regressions--random forest regression (RFR) and back-propagation neural network regression (BPNNR). The optimal BPNNR model (Pearson correlation coefficient [r] = 0.77-0.97, unbiased root mean squared error [ubRMSE] = 0.056-0.077, RMSE = 0.055-0.168, Bias = -0.149 similar to -0.001) was implemented on the Google Earth Engine cloud-based platform to estimate summer albedo at a 30-m resolution for the Karakoram region from 1990 to 2021. Validation against in-situ albedo measurements on three glaciers (Batura, Mulungutti and Yala Glacier) demonstrated that the model achieved an average ubRMSE of 0.069 (p < 0.001), with RMSE and ubRMSE improvements of 0.027 compared to MODIS albedo products. The high-resolution data was then used to identify firn/snow extents using a 0.37 threshold, facilitating the extraction of long-term firn-line altitudes (FLA) to indicate the glacier dynamics. Our findings revealed that a consistent decline in summer albedo across the Karakoram over the past three decades, signifying a darkening of glacier surfaces that increased solar radiation absorption and intensified melting. The reduction in albedo showed spatial heterogeneity, with slower reductions in the western and central Karakoram (-0.0005-0.0005 yr(-1)) compared to the eastern Karakoram (-0.006 similar to -0.01 yr(-1)). Notably, surge- or advance-type glaciers, avalanche-fed glaciers and debris-covered glaciers exhibited slower albedo reduction rates, which decreased further with increasing glacier size. Additionally, albedo reduction accelerated with altitude, peaking near the equilibrium-line altitude. Fluctuations in the albedo-derived FLAs suggest a transition in the dynamics of Karakoram glaciers from anomalous behavior to retreat. Most glaciers exhibited anomalous behavior from 1995 to 2010, peaking in 2003, but they have shown signs of retreat since the 2010s, marking the end of the Karakoram anomaly. These insights deepen our understanding of the Karakoram anomaly and provide a theoretical basis for assessing the effect of glacier anomaly to retreat dynamics on the water resources and adaptation strategies for the Indus and Tarim Rivers.

Study region: The northern region of the Karakoram Range. Study focus: Karakoram is a region in High Mountain Asia with many surge-type glaciers. This study employed over 200 high-temporal-resolution remote sensing images and investigated the variations in elevation and velocity of the Ghujerab River Head Glacier (GRHG) from 2019 to 2023. Furthermore, we elucidated the potential controlling mechanisms. New hydrological insights for the region: Our findings revealed that the GRHG, akin to typical surgetype glaciers in Karakoram, started to surge in the spring and finished surging in the summer, with a duration of less than two years. Throughout the surging process, the glacier transferred a mass of 0.11 +/- 0.003 km3 from the reservoir area to the receiving area, resulting in a thickening of 91.59 +/- 1.04 m at the glacier terminus and thinning of 11.78 +/- 1.04 m in the upper glacier. By analysing the mass balance and glacier surface albedo during surging, we proposed that climatic disturbances in the glacier region provided essential material inputs for the surge. Additionally, based on the seasonal evolution pattern of glacier flow velocity, we inferred a close correlation between surging and variations in subglacial hydrology. The duration of acceleration and deceleration during glacier surging, as well as a comparison with existing studies, further support our conclusion. Future research integrating multi-source remote sensing and onsite observations can support numerical simulations to quantitatively reveal the key processes occurring beneath and within glaciers during surge events.

The Karakoram mountain range is prone to natural disasters such as glacial surging and glacial lake outburst flood (GLOF) events. In this study, we aimed to document and reconstruct the sequence of events caused by glacial debris flows that dammed the Immit River in the Hindu Kush Karakoram Range on 17 July 2018. We used satellite remote sensing and field data to conduct the analyses. The order of the events in the disaster chain were determined as follows: glacial meltwater from the G2 glacier (ID: G074052E36491N) transported ice and debris that dammed the meltwater at the snout of the G1 glacier (ID: G074103E36480N), then the debris flow dammed the Immit River and caused Lake Badswat to expand. We surveyed the extent of these events using remote sensing imagery. We analyzed the glaciers' responses to this event chain and found that the glacial debris flow induced G1 to exhibit accelerating ice flow in parts of the region from 25 July 2018 to 4 August 2018. According to the records from reanalysis data and data from the automatic weather station located 75 km from Lake Badswat, the occurrence of this disaster chain was related to high temperatures recorded after 15 July 2018. The chains of events caused by glacially related disasters makes such hazards more complex and dangerous. Therefore, this study is useful not only for understanding the formation of glacial disaster chains, but also for framing mitigation plans to reduce the risks for vulnerable downstream/upstream residents.

The China-Pakistan Economic Corridor (CPEC), a key hub for trade, is susceptible to glacial lake outburst floods. The distributions and types of glacial lakes in the CPEC are not well documented. In this study, cloud-free imagery acquired using the Landsat 8 Operational Land Imager during 2016-2018 was used to delineate the extent of glacial lakes in the mountainous terrain of the CPEC. In the study domain, 1341 glacial lakes (size >= 0.01 km(2)) with a total area of 109.76 +/- 9.82 km(2) were delineated through the normalized difference water index threshold method, slope analysis, and a manual rectification process. On the basis of the formation mechanisms and characteristics of glacial lakes, four major classes and eight subclasses of lakes were identified. In all, 492 blocked lakes (162 end moraine-dammed lakes, 17 lateral moraine-dammed lakes, 312 other moraine-dammed lakes, and 1 ice-blocked lake), 723 erosion lakes (123 cirque lakes and 600 other erosion lakes), 86 supraglacial lakes, and 40 other glacial lakes were identified. All lakes were distributed between 2220 and 5119 m a.s.l. At higher latitudes, the predominate lake type changed from moraine-related to erosion. From among the Gez, Taxkorgan, Hunza, Gilgit, and Indus basins, most glacial lakes were located in the Indus Basin. The number and area of glacial lakes were larger on the southern slopes of the Karakoram range.

Investigating the trends in the major climatic variables over the Upper Indus Basin (UIB) region is difficult for many reasons, including highly complex terrain with heterogeneous spatial precipitation patterns and a scarcity of gauge stations. The Weather Research and Forecasting (WRF) model was applied to simulate the spatiotemporal variability of precipitation and temperature over the Indus Basin from 2000 through 2015 with boundary conditions derived from the Climate Forecast System Reanalysis (CFSR) data. The WRF model was configured with three nested domains (d01-d03) with horizontal resolutions increasing inward from 36 km to 12 km to 4 km horizontal resolution, respectively. These simulations were a continuous run with a spin-up year (i.e., 2000) to equilibrate the soil moisture, snow cover, and temperature at the beginning of the simulation. The simulations were then compared with TRMM and station data for the same time period using root mean squared error (RMSE), percentage bias (PBIAS), mean bias error (MBE), and the Pearson correlation coefficient. The results showed that the precipitation and temperature simulations were largely improved from d01 to d03. However, WRF tended to overestimate precipitation and underestimate temperature in all domains. This study presents high-resolution climatological datasets, which could be useful for the study of climate change and hydrological processes in this data-sparse region.