1 | INTRODUCTION

With the combination of western disturbance (WDs), Indian summer monsoon (ISM), and Tibetan anticyclone principal synoptic systems, the climate pattern of the Hindukush-Karakoram-Western Himalaya (HKKH) mountain ranges are different from other adjacent Tibetan Plateau (TP) region, hence HKKH comprises the most heavily glacierized watershed outside the poles (Bishop et al., 2010; Farhan, Zhang, Ma, Guo & Ma, 2015; Hewitt, 1998; Palazzi, Hardenberg & Provenzale, 2013; Treydte et al., 2006). The Indus river emerges from the HKKH ranges, and the upper Indus basin (UIB) is known as ”Pakistan Water Tower” (Farinotti, 2017; Immerzeel, Ludovicus & Bierkens, 2010). Comprehensive analysis of the Water Tower Index indicated that UIB ranked top of the most critical and vulnerable unit in Asia even globally with abundant water resources (Immerzeel et al., 2019). More than 40% of freshwater in UIB depends on the seasonal ice/snow melting, it satisfies municipal and industrial water usage for more than 237 million people living in the downstream area. Global warming, along with the heavy irrigation and densely population are creating ever-greater pressure on exploiting mountain glacier resources, which makes the Indus aquifer the most overstressed aquifer worldwide (Lutz, Immerzeel, Shrestha & Bierkens, 2014; Pritchard, 2019). Synthesizing knowledge of the ice reserves is becoming a fundamental prerequisite for the glacier runoff project for water resources allocation and socio-economic development (Cogley, 2012; Radić & Hock, 2011; Vaughan et al., 2013). Most research was concentrated on glacier change, the precise ice reserves estimates in the UIB region remain unclear due to limited accessibility (Bolch et al., 2012; Khan, Naz & Bowling, 2015).
Multiple approaches have been proposed for deducing ice reserves recently, including the geodetic surveying, semi-empirical formula, physical models, and other methods (Farinotti et al., 2019; 2017). Ice core drilling, gravimeter instrument, and ground-penetrating radar (GPR) technique can be classified as geodetic surveying, the high cost makes drilling technique just used for auxiliary medium to glacier structure although it is the most intuitive method with the highest precision (Garzonio et al., 2018); Gravimeter instrument was once tested on Batura glacier in northern Pakistan (BGIG, 1979); GPR technique is a radio-echo detection method that using electromagnetic wave to acquire ice thickness, its high-resolution accuracy and efficiency in noninvasive subsurface makes it is improving continually. During the past three decades, GPR has been applied in the inner and eastern TP widely (Gergan, Dobhal & Kaushik, 1999; Ma et al., 2000; Marussi, 1964; Singh et al., 2012). The second category is the semi-empirical formula, V-A scaling constructs the correlation relationship by linear geostatistical regression, considering that the ice flow and surface landform remain stable, it is suitable for populations of glaciers (Radić & Hock, 2013). The physical model can be sorted as the third category, which consists of shallow ice approximation (SIA), mass conserving, minimization, and velocity-based model. Glacier Bed Topography version 2 (GlabTop2) has been conducted widely for its operability, efficient, and theoretical reliability that originated from the classic laminar flow hypothesis (Linsbauer et al., 2016), it is developed from GlabTop and avoids the laborious process of drawing branch lines manually. Mass conserving, minimization, and velocity-based models were applied to the unique glacier through surface characteristics (Fürst et al., 2017; Maussion et al., 2019). Even the valley glacier fill thickness can be estimated using artificial neural networks that corresponding to the fourth category.
Numerous investigators have applied multiple methods to assess ice reserves in HKKH region, including V-A scaling law, slope-depth, HF-model, GlabTop2, GlabTop2_IITB, Open Global Glacier Model (OGGM) (Bahr, Pfeffer & Kaser, 2015; Chen & Ohmura, 1990; Farinotti et al., 2019; Huss & Farinotti, 2012; Liu, Shen, Sun & Li, 2002). The evaluated ice volume of the entire Karakoram-Western Himalaya region (KH) is 2.2-3.5 × 103 km3 with an ice area of 16.89 × 103 km2approximately (Frey et al., 2014), it was calibrated by GPR-measured points on Baltoro and Chhota Shigri glaciers (Azam et al., 2012; Marussi, 1964). GPR work is the precious verification materials for ice reserves estimation, rare ice thickness and long-term glacial change measurements were involved in the South Asia West district (WGI Partition), in particular of the UIB region (Gärtner et al., 2014; Grinsted, 2013; Hewitt, 2016). Few mapping and further ground-based investigations were carried out before 1995 by the Batura Glacier Investigation Group (BGIG) (BGIG, 1979; Zhang, Chen & Wang, 1996).
In light of the unique geographical and strategic position of the UIB region, the ancient Silk Road from China to South Asia and the modern Karakoram highway both pass through this kernel zone, research on ice reserves can protect against extreme water stress on seasonal and longer time scales, even to maintain social development and geopolitical stability. The main objectives of this paper are to (1) present GPR-surveyed ice thickness of typical glaciers in the UIB sub-basins; (2) compare the plots and profiles of the GPR-measured ice bed elevation versus GlabTop2-estimated results, integrated with GlabTop, Volta model, and the IDW interpolated results, then select the optimized parametric scheme of GlabTop2; (3) simulate the ice reserves of the entire UIB and its sub-catchments, and assess the potential hydrological effect of glacier resources.