**1. Introduction**

In the past two decades, with global warming, grea<sup>t</sup> changes have taken place in global glaciers [1–3]. A recent study based on satellite data and modeling has revealed that

**Citation:** Shen, C.; Jia, L.; Ren, S. Inter- and Intra-Annual Glacier Elevation Change in High Mountain Asia Region Based on ICESat-1&2 Data Using Elevation-Aspect Bin Analysis Method. *Remote Sens.* **2022**, *14*, 1630. https://doi.org/10.3390/ rs14071630

Academic Editors: Gareth Rees and Peter Romanov

Received: 24 January 2022 Accepted: 16 March 2022 Published: 29 March 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

the potential contribution of the world's melting glaciers to sea level rise is 257 ± 85 mm in 2017–2018 [4]. High Mountain Asia (HMA) is the region with the largest distribution of glaciers and diverse glacier types in the middle latitudes of the earth. It is the birthplace of major rivers in Asia, such as the Amu Darya River, the Syr Darya River, the Yangtze River, the Yellow River, the Salween River, the Brahmaputra River and the Indus River [5–7]. To date, the glaciers in the HMA region have undergone significant changes, triggering a series of disasters and having a major impact on people's production and lives. The glacier changes not only directly drive changes in the natural environment in the HMA region, but also have feedbacks to climate change in the whole northern hemisphere and even the whole world. Therefore, accurate monitoring of glacier changes in the HMA region is crucial for studying global glacier and climate change, as well as for understanding the potential impact of glacier retreat [8–11].

The method based on satellite stereoscopic imagery has good spatial coverage and is a common method to monitor glacier elevation changes [12,13]. Existing studies have shown that the glaciers in the HMA are generally in a melting state, but the rate of change varies in different regions and different time spans [9,14,15]. For example, in the Nyanqentanglha Mountains, Ren et al. [16] analyzed the ZY-3 satellite stereo image pair data and found that the glacier thinning rate in 2013–2017 was faster than that in 2000–2013. The results of Brun et al. [17] showed that the glacier thickness across the HMA region decreased by 0.21 m per year from 2000 to 2016. However, glaciers in East Pamir, Karakoram and West Kunlun are in a state of equilibrium or slightly rising, which is called the "Karakoram Anomaly" [18–21].

Lidar altimetry data, such as ICESat-1&2 data, have higher vertical accuracy than satellite-based stereo image pair data. Using lidar data to monitor glacier elevation change is a hot research topic [22]. At present, ICESat-1 equipped with the Geoscience Laser Altimeter System (GLAS), as the first generation of space-borne laser point cloud satellite, can provide global laser point cloud data from 2003 to 2009. The data have a high point frequency (40 Hz) and are widely used in ice sheet monitoring in the Arctic and Antarctic regions [23]. ICESat-1 GLAS data were also used to explore the mass balance changes of glaciers in the Tibetan Plateau, and found that the annual mass loss of glaciers from 2003 to 2009 was −26 ± 12 Gt [24]. However, ICESat-1 data points in the mid-latitude region are sparse, which limits its use in the HMA. In addition, ICESat-1 stopped operation after 2009. ICESat-2 is a new generation of spaceborne lidar satellite equipped with an Advanced Topographic Laser Altimeter System (ATLAS). Since 2018, the ATLAS has provided abundant laser point data every year. Compared with ICESat-1 data, ICESat-2 data not only improves the observation accuracy but can also observe more glaciers in the same time span, thus providing more complete and accurate glacier elevation information [25–27]. Therefore, ICESat-2 provides a new perspective for monitoring inter-annual glacier elevation changes.

At the same time, it would be interesting to explore the intra-annual changes of glacier elevation. Seasonal meltwater from glaciers is a guarantee of water resources in the surrounding and downstream regions of the HMA, in particular in arid and semi-arid regions [28]. Many studies on glacier elevation changes in the HMA region have focused on the inter-annual variability, and studies on monthly/seasonal changes in glacier elevation are still lacking. Although many studies have estimated the intra-annual loss of glaciers, most studies infer monthly/seasonal glacier elevation changes based on relationships with precipitation rather than directly extracting glacier elevation information from satellites observations. Existing studies have shown that the timing of glacier accumulation or melting varies widely in different regions. For example, Ageta [29] found that most of the glaciers in the inner Tibetan Plateau thickened in summer based on continuous ground observations of some specific glaciers, but the spatial patterns still need to be understood to a large extent. Maussion et al. [30] discovered that most glaciers in Pamir and Spiti Lahaul thickened in winter, but they used a clustering algorithm based on monthly precipitation data to calculate monthly changes in glacier elevation, rather than using direct observations of monthly glacier elevation. Wang et al. [31] utilized ICESat-1 and Gravity

Recovery and Climate Experiment (GRACE) data to explore seasonal changes of glaciers in the HMA. However, due to sensor failure, the ICESat-1 data could only be obtained two to three times every year [32], and most of them were concentrated in March, June and October, and could not provide observations for each month. However, studying monthly/seasonal changes of glaciers is of grea<sup>t</sup> significance for disaster warning and an in-depth understanding of the mechanism of glacier changes [31]. In short, none of the existing studies have analyzed glacier elevation changes on a monthly/seasonal scale [33]. The lack of existing observations and effective monitoring methods has limited researchers to accurately analyze the intra-annual variation of glacier elevation. The rich data volume of ICESat-2 data, along with the increase of laser point emission rate (10 KHz) and the number of laser point beams (six beams), provides a new perspective for exploring the intra-annual changes of glacier elevation.

Terrain factors, i.e., elevation, slope and aspect, have impacts on the glacier mass balance and changes [34]. In general, the lower the altitude, the more glacier mass is lost, and vice versa. Existing studies show that precipitation affects glacier accumulation [35]. It is found that the glaciers on the windward side of the mountain retreated less, which was attributed to the effective recharge produced by more precipitation than that on the leeward side of the mountain [36–38]. For example, Wang et al. [39] found that due to the influence of climatic conditions, the glaciers on the northern slope of the Tienshan Mountains decreased more than that on the southern slope. This may be because the precipitation brought by the westerly wind supplemented the mass loss of the glacier on the southern slope, resulting in less glacier mass loss. Therefore, it is important to consider terrain aspect and elevation when monitoring glacier elevation changes using laser point data. Previous studies have rarely considered topographical aspect effects when extracting changes in glacier elevation using ICESat-1&2 data.

The objectives of this study are:


### **2. Study Area and Datasets**

### *2.1. Study Area*

The HMA (26◦N~47◦N, 65◦E~104◦E), located in the central Asia (Figure 1), with an average altitude of more than 4000 m. It is the birthplace of many major rivers in central Asia, East Asia, Southeast Asia and South Asia [40]. The HMA is located in the intersection of various climates. The south of HMA is mainly affected by the Indian monsoon [41], the west and northwest of HMA are mainly affected by the westerly wind, the east is mainly affected by the East Asian monsoon, and the central region is mainly affected by the continental climate. The HMA is sensitive to climate change. Previous studies have shown that a 1.5 ◦C increase in global temperature will lead to a 2.1 ◦C increase in the temperature in the Tibetan Plateau [11]. In addition, various types of glaciers are formed in the HMA region due to the dense mountains and complex topography. According to the Randolph Glacier Inventory (RGI) 6.0 glacier catalog data, there are about 100,000 glaciers in the HMA region, covering an area of nearly 100,000 square kilometers [42]. As these glaciers are less affected by human activities, their changes are largely driven by natural factors. Therefore, glacier change is an important indicator of global warming, and it is of grea<sup>t</sup> significance to explore the changes in the HMA glacier and its relationship with the climate variations.

**Figure 1.** Overview of the High Mountain Asia with background image as elevation above sea level. The white regions in the HMA are the glacier areas. The blue, black and red arrows show the climate-influencing sphere of the Westerly, East Asian monsoon and Indian monsoon, respectively. (**a**) The HMA sub-regions according to the regional division of Brun et al. [17]; (**b**) Major river basins in the HMA according to [43].
