Characterization and Retrieval of Snow Grain Size in the Upper Himalayan Region Using Hyperspectral Prisma Data

Characterization and Retrieval of Snow Grain Size in the Upper Himalayan Region Using Hyperspectral Prisma Data | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Characterization and Retrieval of Snow Grain Size in the Upper Himalayan Region Using Hyperspectral Prisma Data Manish Rawat, Ashish Pandey, Dhananjay Paswan Das, Praveen Kumar Gupta This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5031527/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 14 Apr, 2025 Read the published version in Applied Geomatics → Version 1 posted 10 You are reading this latest preprint version Abstract Rapid urbanization have significantly increased freshwater consumption, leading researchers to focus on accurately predicting snowmelt-derived streamflow using hydrological models. The glacierized basins of the Himalayan region are significantly vulnerable to climate change. The understanding of physical characterization of snow such as snow cover and snow grain size, remains challenging due to inaccessible terrains which creates hindrance for in-situ data collection. The hyperspectral remote sensing datasets are most promising for monitoring and retrieving the snow properties at the micro and macro levels. In this study, the PRISMA hyperspectral dataset was used for the retrieval of different snow grain sizes in the Bhilangana basin of the upper Himalayan region using the Spectral Angle Mapper (SAM) classification method and the Snow Grain Index (SGI) method. The spectral reflectance of different types of glacier features was generated using the SAM classification technique on PRISMA hyperspectral imagery and validated from the USGS spectral library for mapping. The results demonstrated that there is good qualitative agreement observed between the class-wise grain size classes using the grain index and SAM method. Additionally, the overall accuracy of the SAM and the grain size index classification methods for retrieving the grain size of different classes was approximately 88% .The outcomes of the study discloses the competency of PRISMA data for snow grain size mapping over the Mountainous region. The estimated parameters could be applied to climatology, hydrology, and mountain hazard mapping over the Himalayan region. Detailed snow grain size mapping can enhance mountain hazard assessments, including the prediction and management of snow avalanches, thereby contributing to the safety and sustainability of mountainous areas in the Himalayan region. hyperspectral imagery Grain Size Index Snow cover classification endmembers and Eigenvalues Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 1. Introduction Snow and glaciers are valuable natural resources in mountainous regions, influencing major rivers, snowmelt runoff, regional climate, and snow avalanches. Snow plays a significant role in hydrological and climatic models, especially in a snowmelt runoff simulation. However, due to rugged terrain and harsh weather, snow mapping and its physical characteristics with field-based surveys is extremely challenging. In such an environment, remote sensing plays an important role because it gives high temporal and spatial information about the surface of the earth and its physical characteristics. Significant mass loss of glaciers has been caused by continuous climate change in high and rough terrains throughout the world, and several of them are currently at risk (Kumar et al. 2020 ; Ahmed et al. 2021 a; Sarkar et al. 2020 ). The snow cover extent and its spatial extent, as well as albedo, are the most analytical snow parameters for energy-mass balance modeling (Bloschl, 1991; Dozier and Painter 2004 ). During the winter season, more than 40% of the northern hemisphere region is covered by snow, therefore the pertaining snow cover information can be used for various applications i.e. snow melt run-off modeling, snow water equivalent estimation, and climatic modeling. The complex meshing of climate and geological processes is responsible for the degradation of natural resources in the Himalayan ecosystem (Mudbhari et al., 2022 ). Due to the Himalayan challenging terrain, data collection by conventional methods has its limitations, therefore real-time satellite data is a useful tool for mapping the extent of snow cover. Seasonal snow cover variations are a key indicator of climate change in the Himalayan region. At the surface, the assessment of snow grain size holds significant importance in the modeling of snow albedo, which serves as a primary determinant of both snow energy balance and the timing of snowmelt (Marks and Dozier 1992 ). However, due to various landscapes and heterogeneous snow and vegetation cover, the snowpack frequently gets poorly sampled. Therefore, it is necessary to develop an optimal framework for mapping snow grain size and snow cover spatially using advanced remote sensing techniques. One of the most valuable snow physical parameters is snow grain size which indicates the advancement in snow metamorphism. Snow grain size represents a fundamental characteristic of snow that influences its reflectivity (Wiscombe and Warren 1980 ), and it serves as a means to describe the processes of snow metamorphism and stratigraphy (Colbeck 1991 ). In the snow avalanches, snowpack phenomenon stability has been analyzed with the help of snow grain size condition, as fine-grain size snow cover has more strength than coarse-grain snow cover. Snow metamorphism is related to the transformation and structural changes that snow undergoes over time as a result of environmental conditions such as temperature, humidity, and wind. Hyperspectral remote sensing is one of the most advanced technology for surface features extraction and its mapping is based on the identification of the spectral signature of different materials. For example, hyperspectral imagery has been used for the identification of distinct snow properties such as snow cover extent, snow grain size, and different surface material present on the glacier region that can be collected by the hyperspectral sensor, and this strengthens the identification of different features among similar land cover class than the multispectral data. Assessment of hyperspectral data is a quite challenging process due to extensive spatial variability of the different spectral signatures of different land use land cover classes, atmospheric implications, and large data dimensionality (Moughal 2013 ). To overcome this challenging process, semi-automatic techniques (SAM), bi-spectral methods, and machine learning techniques like Support Vector Machines (SVM) can be adopted. The hyperspectral dataset contains multiple narrow continuous spectral bands from the range of visible to the shortwave infrared region, acquiring a huge amount of spectral information at each wavelength region for feature identification (Petropoulos et al. 2012 ). The albedo of snow varies with the snow surface's physical and textural properties that influence the snowmelt phenomenon (Casacchia et al. 2001 ). The reflectance of fresh snow in the visible region is observed to be approximately 90% and will decrease continuously at longer wavelengths (Warren and Winscombe 1980). Fresh snow has a higher albedo in the visible range of the electromagnetic spectrum and gradually decreases with snow aging (Negi et al. 2010 ). The snow melting causes grains to grow clusters and act as a large single grain (Dozier et al. 1981 ; Warren 1982 ). Water content presence between ice crystals leads snow to act as optically large grain size (Colbeck S.C 1979). In the Himalayan region, snow grain size is broadly classified by estimating the snow grain index method using the central band wavelength at 440 and 1030 nm based on ground-based Spectroradiometer data (Negi et al. 2010 ). Multispectral and hyperspectral remote sensing, covering a range of wavelengths from 0.4 to 15 µm, enables the retrieval of various properties such as snow-covered area, albedo, grain size, the presence of liquid water near the surface, and temperature (Dozier and Painter 2004 ) Finer the grain size, the higher is the reflectance observed at this wavelength (Painter et al. 2003 ). In (Negi et al. 2013 ) compared snow grain sizes in the Himalayan region using three different methods: the spectral angle method, the grain index, (SAM), and the ART theory method, based on Hyperion data. Snow reflectance in the visible region is strongly influenced by absorbing impurities and is virtually independent of grain size. In contrast, snow reflectance in the NIR region primarily depends on snow grain size, with a decrease in reflectance occurring as grain size increases. Snow reflectance is highly sensitive to grain size within a wavelength range of 1.0- 1.3µm, extending the diagnostic ice absorption features at 1.03µm and 1.26µm (Negi and Kokhanovsky 2011 ). The maximum deviation in reflectance or sensitive wavelength for snow grain size was examined in the NIR wavelength region as 1030, 1050, and 1240 nm, as the wavelength is more absorptive due to ice presence. According to (Negi and Kokhanovsky 2011 ), these wavelengths have been used to estimate the size of the snow grains. The snow grain index was proposed by (Kokhanovsky et al. 2013) and is a bi-spectral technique, with one region in the visible and the other in the NIR region (440 and 1030/1240). Using AVIRIS data for the US region, a method was developed for quantitatively retrieving snow characteristics (Green et al. 2002 ). Snow Grain size mapping was done using NDSGI techniques between MODIS band 1 (620–670 nm) and MODIS Band 2 (841– 876) radiances of pure snow pixels (Scambos et al. 2007 ). The snow grain size was retrieved from AVIRIS hyperspectral data using multiple wavelengths of 860, 1050, 1240, and 1730nm respectively (Le et al. 2001). The objective of this study is to utilize the PRISMA hyperspectral dataset to retrieve various snow grain sizes in the Bhilangana basin of the upper Himalayan region in Uttarakhand on February 18, 2020. This was done using the Spectral Angle Mapper (SAM) classification method and the Snow Grain Index (SGI) method. Additionally, the spectral reflectance of different glacier features was generated using the SAM classification technique on PRISMA hyperspectral imagery and validated against spectra from the USGS spectral library for accurate mapping. The estimated parameters derived from this data can be instrumental in advancing our understanding of regional climatology by providing insights into snowpack characteristics and their influence on local and global climate patterns. 2. Study area The study area is located in the upper Himalayan range. It is located between the latitude of 31°1'39.31"N to 30°42' 8.30"N and longitude of 78°43'29.61"E to 79°50'45.55"E, with an elevation ranging from 4000 to 5800 meters. This study focuses on the PRISMA data scene within the Bhilangana basin in the upper Himalayan region, which falls under the Uttarkashi district of Uttarakhand, India (Fig. 1 ), covering an area of 886.56 km 2 to retrieve different snow grain size characteristics. This region experiences a cold and arid climate and experiences heavy snowfall during the winter season while most of the terrain in this area lacks significant vegetation cover, which can have implications for the study of snow grain size characteristics. Additionally, the terrain in this area is largely devoid of substantial vegetation, which influences snow dynamics. The absence of vegetation affects snow accumulation patterns, snow cover duration, and the physical properties of the snowpack. These factors are crucial for understanding snow grain size characteristics and their implications for hydrological and climatic processes in this high-altitude environment. 3. Data used and sensor description The study has been carried out by using PRISMA (PRecursore IperSpettrale Della Missione Applicativa), the latest Hyperspectral satellite sensor of the Italian Space Agency (ASI), which was launched into orbital on March 22, 2019, and is positioned in a sun-synchronous Low Earth Orbit at an altitude of 620 km. It operates on a repeat cycle of approximately 29 days. The PRISMA hyperspectral payload consists of a push broom sensor having 230 contiguous spectral bands (400 to 2500 nm wavelength region) of which 66 bands are in the Visible Near Infrared (VNIR) range and 164 bands in the Short Wave Infrared (SWIR) range with coherent, and having a spectral resolution is about 12 nm narrow bandwidth and a 30 m spatial resolution with a swath width of 30 km. The PRISMA L1 data products distributed with HDF5 file format were downloaded from the PRISMA portal ( https://prisma.asi.it ) and re-projects with a geographic look-up table (GLT) for the correction of Bowtie artifacts associated with missing data from the ENVI software. 4. Methodology The first step in processing the PRISMA Hyperspectral dataset involved converting it from HDF5 to HDR format using the R package within ENVI software. This dataset contains a continuous spectrum with a high-dimensional volume of spectral data. Due to the large numbers of the band available in the PRISMA spectral dataset, there are radiometric interferences, low Signal-to-Noise Ratio (SNR), and heavy water absorption influences in several spectral bands, as a result, a total of 24 bands were dropped from 230 original bands, Finally, 206 bands are calibrated in this dataset, which requires accurate pre-processing for operating the noise control in several optimized bands. During the pre-processing phase, the visual interpretation of the remaining 206 spectral bands involved removing uncalibrated bands, de-striping a few SWIR bands, and applying atmospheric corrections using ENVI's FLAASH model. Before performing atmospheric correction on the PRISMA data, we eliminated uncalibrated images or bands in the SWIR region that were either noisy or contained no data. We selected radiometrically calibrated bands with band numbers ranging from 9 to 57 in the VNIR range and from 78 to 220 in the SWIR wavelength range, resulting in a total of 158 spectral channels used in the PRISMA Hyperspectral dataset. The primary objective of this study is to analyze the spectral signatures of snow-glacier features within the PRISMA scene of the Bhilangana basin in the upper Himalayan region and ascertain in mapping different snow features using hyperspectral remote sensing techniques. Given the challenging conditions of the high rugged terrain and extremely cold climate during the winter season, ground truth data is lacking. Therefore, in this study, we compared the snow grain size acquired through the Snow Grain Size Index method with the results of the spectral angle classification method. The comprehensive methodology employed in this study is depicted in Fig 2. 4.1. Elimination of Bad Band in Dataset During the pre-processing of PRISMA hyperspectral imagery, certain bands were identified as 'bad bands. These bands exhibited either a lack of data or strong absorption due to water vapor. The noise management strategy for the bands involves identifying 'bad' pixels, mitigating vertical striping effects, and compensating for atmospheric distortions in the data. It also aims to minimize the compounding effects through image processing (Datt et al. 2003).To identify and remove these undesirable bands, we performed manual visual inspection, carefully examining each spectral band for anomalies. Out of the initial 230 bands present in the PRISMA data, a total of 24 bands were identified as 'bad bands' and only 206 bands were considered for further analysis. Table 1 shows the list of eliminated bands. Table 1. List of unusual bands of the PRISMA sensor Bands number Condition 1, 48,84,85 Vertical stripping bands 100-104 Vertical stripping bands 136 Vertical stripping band 149 - 156 Vertical stripping bands 159, 203 Vertical stripping bands 161,164, 229,230 Water vapour absorption bands 4.2 Destriping of PRISMA Bands There are vertical stripes observed in a few bands of the PRISMA dataset consisting of white or dark striping pixels. The bright pixels exhibit a significantly higher sensor response, while the dark vertical strips show minimal or no sensor response. These vertical stripes accommodating abnormal pixels have lower DN values or no information when compared to neighbouring pixels (Fig 3). These distorted vertical strips can be replaced by their adjacent DN pixels value using the spatial editor tool available in ENVI software . This is a critical issue in remote sensing, as it can affect the accuracy of interpretations and analyses derived from the PRISMA dataset . 4.3. Radiometric Correction Rescaling of dataset PRISMA data consists of 230 spectral bands ranging from 402.40 to 2496.86 nanometres (nm) with spectral resolution of 12 nm and 30m spatial resolution. The sensor consists of two array detectors, VNIR (Visible Near Infrared) and SWIR (Short Wave Infrared) bands. Among the 230 bands, only 206 bands are calibrated for the VNIR, which wavelength region from 402.40 nm to 998.36 nm, and a SWIR range from 919.13 to 2496.86 nm wavelength regions. To convert Digital Numbers (DN) to radiance values, the data needs to be rescaled. As a result, the VNIR and SWIR bands in the PRISMA image have considered two scaling factors. The scaling factors of 40 for the VNIR band and 80 for the SWIR bands are provided as input for the FLAASH Model. When converting DNs to radiance values for subsequent analysis and interpretation, calibrating the bands and incorporating scaling factors are essential to ensure the accuracy and reliability of the data. This procedure is an important part of the data processing workflow. 4.4. FLAASH for Atmospheric Correction FLAASH (Fast Line-of-Sight Atmospheric Analysis of Spectral Hypercubes) is an atmospheric correction model that uses a radiative transfer program to reduce atmospheric inaccuracies. The FLAASH model based on the Moderate Resolution Atmospheric Transmission version 4 (MODTRAN4) radiative transfer algorithm was used to perform atmospheric correction (Anderson et al. 1999; Cooley et al. 2002). The input radiance image for FLAASH should contain a radiometrically calibrated radiance image. This model is rigorous and requires various parameters for data scaling as well as to define other information such as scene centre location, sensor information, atmospheric, and aerosol model that are to be used in MODTRAN-4 radiative transfer code which is used in FLAASH, the various parameters used for the atmospheric correction are listed below in Table 2. FLAASH allows us to understand the different standard and derived atmospheric and aerosol models attempted for the Hyperion dataset (Matthew et al. 2000) and provides water vapor estimates for each atmospheric model as shown in Table 3. It has been observed that the water vapor data derived from the atmospheric correction model, particularly the water vapor absorption feature at 820 nm wavelength, was sufficient for estimating the water vapor quantity column for each pixel. As a result, the atmospheric model (Mid-latitude winter) was used to determine the water vapor quantity for each pixel in the image, and a 2-band KT aerosol model was employed for atmospheric correction (Kaufman et al. 1997). Consequently, we selected the atmospheric model 'Mid-latitude summer' and the aerosol model 'rural' for our specific analysis. The FLAASH atmospheric correction results for the PRISMA image are displayed in their respective spectral profiles in Fig. 4 with snow pixel Spectral profile (Z-profile), a) before Atmospheric corrections and b) after Atmospheric FLAASH corrections. Table 2 . Various parameters were used in the FLAASH model Scene location 31°1’39.79 N and 78°43’29.13 E Sensor altitude 620.00 km Pixel size 30.00 m Acquisition date 18-Feb-2020 Atmospheric model Mid-latitude Winter Water absorption feature 820 nanometers Aerosol model Rural Aerosol retrieval 2-Band(K-T) Initial visibility 40 km Width polishing 9 Aerosol scale height 1.50 km Modtran multiscatter model Scaled DISCORT Number of Discort stream 8 Table 3 . Different atmospheric model parameters were attempted to correct the atmospheric effects in PRISMA image of the study area Study area Atmospheric model Aerosol model Obtained water vapors(gm/cm 2 ) PRISMA scene in Bhinlgana basin (Feb 2020) Mid-Latitude summer 2- Band KT 0.1286 Sub-Arctic summer 2- Band KT 0.0474 Sub-Arctic winter 2- Band KT 0.0483 Mid-Latitude winter (1135nm) 2- Band KT 0.0511 Mid- Latitude winter(820nm) 2- Band KT 0.0535 Tropical 1135nm 2- Band KT 0.0831 4.5. Snow Grain Size Extraction Snow grain size is one of the important physical parameters of snow characteristics that is accountable for snow metamorphic characteristics such as spectral reflectance, snow avalanches, snow melting rate, snow depth, etc. Snow grain size is a vital parameter whose understanding is required across different disciplines, from snow chemistry and gas interactions to Earth's radiation balance. In this study, the Grain Size Index (GSI) and Spectral Angle Mapping (SAM) were employed to classify snow grain size. The snow grain size map generated by the GSI method, as proposed by (Negi et al. 2010) is based on ground truth data collected by a hyperspectral sensor and was further used for validation against the SAM-classified image. These two methods SAM and GSI are qualitative. This implies that they provide relative assessments of snow particle size, allowing researchers to classify snow grains as fine, medium, and coarse based on their spectral properties. 4.5.1 Snow Grain size estimation using Grain Index method In studies related to snow avalanches, snow grain size is closely linked to the stability of the snowpack. Finer snow grain sizes tend to exhibit greater stability compared to coarser snow grains. It has been observed that the reflectance of snow in the near-infrared wavelength range decreases as the snow grain size increases. The spectral reflectance of snow in the wavelength spectrum range of 1030 nm to 1045 nm is found to be the most sensitive region for the estimation of snow grain size variation. In this range, finer snow particles result in higher reflectance measurements (Nolin and Dozier 2000; Painter et al. 2003). The interpreted snow reflectance characteristics to understand the effect of soil contamination and snow grain size. To analyze the shape and depth of the absorption peak between 1025 nm and 1040 nm as snow grain size increased. Reflectance in the visible region is significantly affected by absorbing impurities and shallow snow depth. However, in this wavelength range, snow grain size variability is nearly independent. Contamination reduces snow reflectance in the visible region, while surface snow grain size is more susceptible to changes in the near-infrared region (Singh et al. 2010). They measured different snow grain sizes using snow reflectance at 440–590 nm in the visible range and 1040–1050 nm in the NIR spectrum (Negi et al. 2010). They found that as grain index threshold values increased, there was a corresponding increase in snow particle size. Additionally, it was observed that due to snow aging and variations in snow grain size, a diagnostic ice absorption depth feature appeared near 1030 nm, as shown in Fig 5. In this wavelength region, snow grain size exhibited higher sensitivity for reflectance. Snow grain size mapping in this study has been derived from the grain size index (GSI) method. The snow grain index model is based on the bi-spectral method, which consists of visible (dependence on contamination) and near-infrared (dependence on grain size) regions for retrieving snow grain size. This method is suitable for clean, dry snow in the upper Himalayan region, which is heterogeneous in terms of spatial-temporal changes in snow grain size. In this research, snow grain size formulation has been done by using the PRISMA bands 6 (visible channel wavelength 441.63 nm) and 69 (NIR channel wavelength 1028.79 nm) as shown in the following equation: Grain size was classified into fine, medium, and coarse classes having snow grain index threshold values as follows: 0.000-0.170, 0.170-0.260, and 0.260-0.350, as recommended by (Negi et al., 2010) and given in Table 4. The extraction of snow grain size using the Grain Index and its classified map, based on threshold values of the Grain Index, are displayed in Figure 7 (a and b). Table 4. The snow grain size was categorised into three different classes S. No Class Range of snow grain size (mm) Threshold for snow grain size index 1 Fine < 0.5 0.0 – 0.17 2 Medium 0.5 – 1.0 0.17 -0.26 3 Coarse 1.0 -2.0 0.26 -0.35 4 Unclassified Non-snow-covered area NA In this study, a decision tree algorithm, a non-parametric classifier, and a machine learning technique were employed for snow grain size classification using the bi-spectral grain index method, as illustrated in Fig 6. Decision tree algorithms feature a tree-like structure comprising root nodes, internal nodes, and leaves. They evaluate one or more input features in various combinations to aid in the classification process. Using different grain size threshold values, as detailed in Table 4, the decision tree successfully classified snow grain sizes into fine, medium, and coarse classes. 4.5.2. Band Histogram statistic The wavelengths 1028 nm and 441 nm in hyperspectral images are associated with PRISMA band numbers 69 and 06, respectively. After performing an atmospheric correction on the PRISMA imagery, statistical analysis of the imagery data was performed in this study. This analysis brings valuable insight into the physical characteristics of the imagery that can aid in understanding the information contained in the images. The histogram statistics of the PRISMA imagery after atmosphere correction show the average value was 0.266, this statistic shows the mean value of the pixel in the image. It offers an indication of the overall brightness or reflectivity of the scene, and the highest value was 1.00 this value represents the highest pixel intensity within the image. It indicates the highest reflectance or radiance at the given wavelengths. The standard deviation was observed to be 0.225, the standard deviation quantifies the variability or dispersion of pixel intensities in the image. Fig 8 depicts a histogram of snow grain index. A histogram is a useful tool for evaluating an image's general properties, such as contrast, brightness, and the presence of different materials or features. It is indispensable to obtain a better understanding of the nature of the PRISMA imagery after atmospheric changes by examining these data and visualizing the histogram. 4.6. Estimation of Snow Cover Area Snow cover classification mapping is challenging to apply in the Himalayan region due to mountain shadows and cloud cover over the snow region (Kulkarni et al. 2002). To address these challenges, the normalized difference snow index (NDSI) can be used to overcome this problem. One more advantage of NDSI techniques is the ability to remove snow from cumulus clouds (Negi et al. 2009). In this study, we pre-processed PRISMA data and employed three wavelength channel combinations consisting of (587.79, 1646.96), (554.53, 1646.96), and (500, 1726.43) nm were used for the NDSI. Generally, NDSI threshold values are set at 0.4 for snow cover mapping. However, to mitigate heavy contamination in snow, the threshold value was raised to 0.6 in this study (Negi, Kulkarni, and Semwal Citation 2009; Negi et al. 2010). Normalized difference snow index (NDSI) was calculated using green and SWIR bands ratio to produce the snow cover map in mountain shadows (Kulkarni et al. 2002). To further refine our analysis, we used a higher threshold value of 0.7 to identify pure snow pixels, as illustrated in Fig 9. Any pixels with values below 0.7 were masked and considered as non-snow-covered regions. Fig 9 displays two distinct snow cover maps of the study area, each generated with different NDSI threshold values: 0.40 and 0.70. As shown in Table 5, the NDSI threshold increases from 0.40 to 0.70, the snow cover area decreases significantly across all band combinations. This is because higher thresholds filter out pixels with lower snow reflectance or mixed land cover, identifying only pure snow areas. The NDSI thresholds used to generate snow cover areas at different wavelength regions and their details are given in Table 5. The NDSI formula shown in given equation: Table 5. Snow cover area for different spectral ratios at different NDSI threshold values Bands 25(587.79nm), 127(1646.96nm) Bands 21(554.53nm), 127(1646.96nm) Bands 14(500.10nm), 135(1726.43nm) NDSI threshold values Snow cover area (sq. km) NDSI threshold values Snow cover area (sq. km) NDSI threshold values Snow cover area (sq. km) 0.40 822.06 0.40 815.04 0.40 778.33 0.50 785.06 0.50 777.27 0.50 676.67 0.60 705.82 0.60 696.81 0.60 452.24 0.70 518.75 0.70 491.70 0.70 234.00 The combination of Bands 25 (587.79 nm) and 127 (1646.96 nm) consistently shows the largest snow cover area across all thresholds, indicating that this band pairing is more inclusive of snow cover. The combination of Bands 14 (500.10 nm) and 135 (1726.43 nm) shows the smallest snow cover areas, especially at higher NDSI thresholds, suggesting that this pairing may be better for identifying only the cleanest or freshest snow. 4.7. Land Cover Classification Hyperspectral Image Analysis The following hyperspectral analysis processes, including the Minimum Noise Fraction (MNF) for spectral data reduction, Pixel Purity Index (PPI) for determining spectrally pure pixel values, and subsequently n-dimensional Visualizer for classifying the endmembers directly from the image were applied to the PRISMA dataset. The approach of the Spectral angle mapper (SAM) method has been employed to determine the snow cover features in the glacier region. 4.7.1. Minimum Noise Fraction Transformation (MNF) The minimum Noise Fraction transform technique is an improved version of the Principal Components transform that produces output results by decreasing the signal-to-noise ratio in hyperspectral data. Hyperspectral data consists of different spectral bands that are affected by the low signal-to-noise ratio, especially in the short-wave infrared wavelength spectrum. Noise in hyperspectral images is one of the most common problems that has been observed, it causes interference with feature identification and abundance calculation, so it is necessary to eliminate this complication to acquire good matching spectra. For this determination, Minimum Noise Fraction is the most common processing technique. MNF transformation disintegrates noise from data and determines the inherent dimensionality of the dataset, helping to identify which bands are suitable for further data analysis. There are two steps in the data transformation process. First, the computed covariance matrix is decorrelated to remove noise without band-to-band correlation. Subsequently, the Eigenvalues of each band are calculated, and bands with higher Eigenvalues are selected. Generally, more than one Eigenvalue contains coherent images having valuable information, while the coherent images associated with small eigenvalues contain erroneous information that is dominated by noise (Jensen 2005). The image pixels are presented by Eigenvalues. The MNF technique reduces the dimensionality of large data by considering the coherent images for forwarding data analysis. Fig 10 shows the relation between Eigen number and eigenvalues i.e. MNF band images used to assess the dimensionality of the data and identify the noise-affected band (Qiu et al. 2006). In this study, the first 20 MNF bands containing the most coherent data have been chosen based on the Eigenvalues, while the remaining MNF bands would have noise-affected data that make them unsuitable for data analysis . After visualization of each MNF transformed image, only 20 MNF bands have been selected for subsequent analysis as they consist of higher Eigen containing more feature information. 4.7.2 Pixel Purity Index (Spatial data reduction technique) The processing of the Pixel Purity Index in ENVI has been used to identify the purest pixel from the mass majority of pixels representing mixed pixels in the hyperspectral image. These pure pixels are also known as spectral endmember and their pure spectral signatures. Pixel spectra are considered as an image endmember that lies in an n-dimensional space (Rogge et al. 2007). After eliminating noise and dimensionality reduction of data from MNF transformation techniques. The top 20 coherent MNF bands have been incorporated into the Pixel Purity Index (PPI) algorithm. Fig 11 shows the results of the PPI curve which indicates both the number and distribution of pure pixels in our study area. Fig 11 shows the PPI iteration process was 5000 which obtained 6000 number of pure pixels for endmember extraction in hyperspectral image analysis. This comprehensive approach allows us to gain a deeper understanding of the spectral characteristics and composition of the studied area, contributing to more accurate and insightful analysis and interpretation. 4.7.3 N-Dimensional Visualizer Technique The N-Dimensional Visualizer is an interactive approach for enhancing the process of selecting spectrally pure pixels from a 2-dimensional PPI (Pure Pixel Index) image. In remote sensing and hyperspectral data analysis, identifying pure pixels are the spectral characteristics of individual elements or structures on Earth's surface. In the present analysis, each end member has been assigned to a distinct glacier feature and then obtained using the 3-D scatter plot that is shown in Fig 12. This 3D scatter plot allows for a more tangible and interpretable visualization of the end members, which can help in understanding the spectral characteristics and variations of glacier features. The spectra of image end-members have been used as reference spectra for different snow surface characteristics. These reference spectra are employed in conjunction with the classified Spectral Angle Mapper (SAM) method. SAM is a spectral classification technique that compares the similarity of the spectra in the hyperspectral data to the reference spectra. This method helps classify different surface materials or features in the study area based on their spectral characteristics. 4.7.4 Spectral Angle Mapper (SAM) The classification of hyperspectral images by a spectral angle mapper is one of the finest techniques. It is a pixel-based automated classification algorithm. The SAM technique compares the spectral signature of each pixel in an image to reference spectra of various feature classes extracted from a spectral library or in the field with a spectrometer. The reference spectra of different features considered as endmembers can be obtained primarily through field measurements, the laboratory, or extracted directly from the hyperspectral imagery. The field spectral signature was used to reproduce the snow grain size map by SAM, as described in Negi et al. (2013). The selected spectra generated from the image were further used as reference spectra using the SAM method for grain size mapping (Rowan and Mars 2003) The SAM method determines the spectral similarity between image pixel spectra and reference spectra by computing the angle between two vectors containing these spectral signatures (Kruse et al. 1993) as shown in fig 13. The obtained image spectra from the SAM technique were used for identification and comparison with a spectral signature that was collected from the USGS spectral library built in ENVI software. The following formula calculates the spectral angle (α). 5. Results and discussion 5.1. Snow grain size mapping 5.1.1 Grain size mapping using the SAM method Based on their spectral signature, PRISMA data from the central Himalayan region was used to estimate different snow grain sizes. The spectral signature for fine, medium, and coarse snow grain has been obtained using Z-profiling from the selected endmember extract from the SAM-classified image, as shown in Fig 14. The generated spectral signature has been used for snow grain size mapping. These size classes represent the relative size and shape of individual snow grains, which can be indicative of snowpack properties such as density and temperature. The reflectance of various snow grain sizes was observed to decrease as contamination levels increased, as illustrated in Fig 14. Snow reflectance at 1030 nm has been found to drop by 15% when metamorphic characteristics of grain size change from fine to medium and a 10% decrease when transitioning from medium to coarse metamorphic grain characteristics. However, at visible wavelengths (ranging from 400 nm to 700 nm), reflectance remained relatively insensitive to changes in grain size. Reflectance in the near-infrared region, predominantly at 1030 nm, exhibited a high sensitivity to grain size, as previously reported by (Nolin and Dozier 2000, Dozier 1988). The presence of contamination on snow particles significantly influences the reflectance within the visible region, which can be examined using a hyperspectral remote sensing method. The spectral signatures of the glacier region were extracted from PRISMA data and are displayed in Fig 16. These figures illustrate how the high reflectance of snow changes in the visible region in response to varying types of contamination. It has been observed that fresh snow particles exhibit the highest reflectance in the visible spectrum, with reflectance decreasing almost to zero beyond the shortwave infrared (SWIR) region due to the pronounced ice absorption characteristics. It is also observed that in the near-infrared region, snow contamination has less effect on snow grain size. This classification process involves collecting spectra from the image and identifying each distinct class. Fig 15 shows the SAM-classified image for the snow-glacier region. However, in this classification approach, some areas were not classified because of low spectral signatures. This happened for high altitude areas, backside shadows of hilly regions. As shown in Table 6, the maximum covered area in the study region is observed in the class of less contamination snow, which spans 264.94 km², followed by Fresh snow with a coverage area of 190.33 km². In contrast, the minimum coverage areas are attributed to Barren Rocky and Vegetation, which encompass 7.83 km² and 7.95 km², respectively. The High contamination snow class covers an area of approximately 72 km². Table 6. Retrieved of glacier features using SAM Classification Class Area (Km 2 ) Fresh snow 190.33 Wet snow 51.84 Less contamination snow 264.94 High contamination snow 72.1 Glacier ice 102.05 Barren Rocky 7.83 Vegetation 7.95 Shadow region 104.4 The retrieval of snow grain size was performed using the GSI method and compared with the SAM algorithm for a specific region in the greater Himalayas range. Fig. 17(a and b) show the spatial distribution of snow grain sizes extracted from SAM and GSI respectively. In this study, it has been observed that a large portion snow cover area is covered with fine and medium grain size due to accumulation season in the month of mid-February when temperatures are low and rainfall is scarce. These conditions affect the snow properties in higher altitude regions. Table 7 shows the regional distribution of fine and medium snow particle sizes, indicating a remarkable uniformity between the results of the particle Size Index (GSI) and the SAM-classified imagery. The determination of the SAM angle threshold value helps avoid misclassification over steep slopes and dark or low-illuminated areas, which are marked as unclassified (black) areas. In a few places, a small deviation may be due to the selection of grain size threshold values used for the conversion of grain index to grain size classes. A comparison of the two datasets indicates that the SAM-classified image shows an increase of 13.69 km² in fine grain size snow cover area, a decrease of 23.16 km² in medium grain size snow, and also decrease of 71.48 km² in coarse grain size snow. Table 7. Retrieval of snow grain size using SAM, and Grain Index method Snow grain class Grain size (mm) Area (Km 2 ) (GI method) Area (Km 2 ) (SAM method) Fine snow grain 0.00 - 0.50 211.87 225.56 Medium snow grain 0.50 - 1.00 236.58 213.42 Coarse snow grain 1.00 - 2.00 139.61 68.13 The SAM and G.S.I qualitative grain size methods studied the classification of fine, medium, and coarse grain size classes in the snow glacier region. Such classifications are important in studying the avalanches, snow melting rate as well as snow depth. The two-grain size methods G.S.I and SAM have their advantage as well as limitations. The grain index method consists of two spectral bands that classify the snow grain based on the threshold values. This ratio technique reduces the topographic effects, which helps identify the snow grain under the mountainous shadow regions. The second classified Spectral angle mapper method uses selected maximum band information data to obtain qualitative snow grain sizes. The main advantage of incorporating all bands is the retrieval of snow grain and other features of the snow glacier region based on the identification of spectral signatures. The current methodology for mapping snow grain sizes using SAM is well-suited for the Himalayan region at varying altitudes, ranging from 4000 meters and above. 5.1.2 Classification evaluation for PRISMA data The few random training samples were selected by using certain pixels of the PRISMA image that endmember spectra closely matched to the endmember spectra of the features found in the USGS spectral library and some reference spectra signatures have been taken from different literature. The confusion matrix analysis from the grain size classified map, generated using hyperspectral data, was utilized to assess classification accuracy in ENVI, as presented in Table 8. As a result, the overall accuracy and Kappa coefficient were computed. Classification accuracy assessment of Grain index map using SAM technique for glacier: Overall accuracy = (1107166/ 1249122)/100 = 88.63% Kappa coefficient = 0.82 To enhance classification accuracy and ensure the reliability of the results, SAM classification outcomes were incorporated into a confusion matrix concerning user-defined ROI end members. The classification accuracy is presented in the Table 9. In this study area, the classification accuracy results reveal that the producer accuracy for coarse snow grain is 69.02%, while the user accuracy for coarse snow grain is 59.21%, suggesting a significant commission error in the coarse snow grain class. The user accuracy for fine snow grain is 85.07%, and for medium grain, it is 76.85%. Table 8 . Theoretical error matrix of snow grain size classification Class Unclassified Fine snow grain Medium snow grain Coarse snow grain Total Unclassified 614047 2461 5052 84 621644 Fine snow grain 4022 254337 40592 21 298972 Medium snow grain 34418 3153 192835 20510 250916 Coarse snow grain 4192 18387 9064 45947 77590 Total 656679 278338 247543 66562 1249122 Table 9. Accuracy assessment of different classification Class Production Accuracy (percent) User Accuracy (percent) Unclassified 93.50 98.77 Fine snow grain 91.37 85.07 Medium snow grain 77.89 76.85 Coarse snow grain 69.02 59.21 The snow grain size classified map created using the Spectral Angle Mapper (SAM) method demonstrated an overall matching area of approximately 88.63% when compared to the Grain Size Index (GSI) classified maps. This shows that SAM is a more effective technique for classifying snow grain size, providing valuable information for various applications, and contributing to the understanding of several important factors, such as snow contamination, fractional snow cover area, and albedo. However, it is essential to emphasize that a quantitative estimation of snow grain size features was not feasible in this study owing to the lack of ground truth data. These parameters are essential for accurately mapping snow cover, predicting the timing and magnitude of snowmelt, and better estimating future water resources and climate effects. Additionally, accurate snow cover mapping is crucial for applications related to glaciers, including distinguishing between clean and debris-covered glaciers, assessing glacial hazards, and conducting research in climatology, hydrology, and snow avalanche hazard analysis applications in the Himalayan region. In comparison to standard remote sensing data, hyperspectral data provides comprehensive spectral information that can considerably increase the accuracy of snow-related studies. By comparing the Spectral Angle Mapper method with bi-spectral satellite data, the study successfully captures changes in snow grain size dynamics, contributing to our understanding of snowpack characteristics in the upper Himalayan region. The application of this methodology is particularly beneficial in regions where snow conditions remain relatively consistent, as it allows for the accurate and timely assessment of snow properties. This, in turn, supports critical applications such as avalanche forecasting, water resource management, and climate research, all of which are essential for both scientific investigations and practical decision-making in the Himalayan region, where snow cover plays a vital role in various aspects of life and the environment. 6. Conclusions In this research, we examined the variations in spectral properties of different snow types within the upper Himalayan region using state-of-the-art hyperspectral PRISMA data. We employed two distinct classification methods: SAM and GI. These methods offer critical insights into snow attributes like grain size and snow cover, which are of utmost importance for hydrological modeling. These classification approaches serve as valuable tools for estimating snow grain size in scenarios where qualitative information about grain size is needed, such as in the context of melting snow, glacier forecasting, and climate studies, among others. While both SAM and GI techniques were utilized to assess snow particle size, the SAM method demonstrated superior effectiveness in identifying unique snow characteristics within the glacier region. This advantage stems from the SAM classifier's ability to utilize a select few spectral bands that yield more informative results with reduced noise. In contrast, the grain index method relies on just two spectral bands. Additionally, the ratio approach stands out due to its capacity to account for topographical influences. Spectral wavelengths around 441 nanometers and 1043 nanometers emerge as the most sensitive indicators of snow grain size. However, extracting endmembers from PRISMA data, characterized by a 30-meter resolution and a broad swath width, poses a challenging task when estimating snow grain size. Furthermore, the PRISMA hyperspectral data proves invaluable for studies in glacier regions. Detecting vertical variations in the snowpack through grain retrievals utilizing various ice absorption channels can significantly contribute to the understanding of snowpack stability in Himalayan snow avalanche research. Declarations Acknowledgment The study was carried out using PRISMA Products freely provided by the Italian Space Agency (ASI), and delivered under an ASI License. The Italian Space Agency is grateful for the PRISMA hyperspectral data. The authors are also thankful to the Department of Water Resources Development and Management (WRD&M), IIT Roorkee, for providing all necessary facilities and constant encouragement for doing this research work. Funding This research received no funding Conflicts of interest All authors have read and approved the final manuscript and have no conflicts of interest related to this research work. Author Contributions Manish Rawat- conceptualized, formalized, and interpreted the results, and writing—original draft preparation.; Ashish Pandey- Supervision, results analysis, review, editing.; Dhananjay Paswan Das—review and Praveen Kumar Gupta- Supervision, results analysis, review, editing. All authors have read and approved the final manuscript. References Ahmed, R., Wani, G. F., Ahmad, S. T., Sahana, M., Singh, H., & Ahmed, P. (2021). A review of glacial lake expansion and associated glacial lake outburst floods in the Himalayan region. Earth Systems and Environment, 5(3), 695-708 . https://doi.org/10.1007/s41748-021-00230-9 Anderson GP, Pukall B, Allred CL, Jeong LS, Hoke MA, Chetwynd JH, Adler-Golden SM, Berk A, Bernstein LS, Richtsmeier SC, Acharya PK (1999) FLAASH and MODTRAN4: state-of-the-art atmospheric correction for hyperspectral data. In1999 IEEE Aerospace Conference. Proceedings (Cat. No. 99TH8403) Mar 7 (Vol. 4, pp. 177-181). IEEE. https://doi.org/10.1109/AERO.1999.792088 Bloschl G, Kirnbauer R, Gutknecht D (1991) Distributed snowmelt simulations in an alpine catchment: 1. Model evaluation on the basis of snow cover patterns. Water resources research 27(12):3171-9. https://doi.org/10.1029/91WR02250 Cooley T, Anderson GP, Felde GW, Hoke ML, Ratkowski AJ, Chetwynd JH, Gardner JA, Adler-Golden SM, Matthew MW, Berk A, Bernstein LS (2002) FLAASH, a MODTRAN4-based atmospheric correction algorithm, its application and validation. InIEEE international geoscience and remote sensing symposium. 1414-1418. IEEE. https://doi.org/10.1109/IGARSS.2002.1026134 Colbeck SC (1991) The layered character of snow covers. Reviews of Geophysics 29(1):81-96. https://doi.org/10.1029/90RG02351 Colbeck SC (1979) Water flow through heterogeneous snow. Cold Regions Science and Technology 1(1):37-45. https://doi.org/10.1016/0165-232X(79)90017-X Casacchia R, Lauta F, Salvatori R, Cagnati A, Valt M, Ørbæk JB (2001) Radiometric investigation of different snow covers in Svalbard. Polar Research 20(1):13-22. https://doi.org/10.1111/j.1751-8369.2001.tb00035.x Dozier J, Painter TH (2004) Multispectral and hyperspectral remote sensing of alpine snow properties. Annu. Rev. Earth Planet 32:465-94. Dozier J, Schneider SR, McGinnis Jr DF (1981) Effect of grain size and snowpack water equivalence on visible and near‐infrared satellite observations of snow. Water Resources Research 17(4):1213-21. https://doi.org/10.1029/WR017i004p01213 Dozier J, Davis RE, Chang AT, Brown K (1988) The spectral bidirectional reflectance of snow. In Spectral Signatures of Objects in Remote Sensing (Vol. 287, p. 87). Dozier J, Painter TH (2004) Multispectral and hyperspectral remote sensing of alpine snow properties. Annu. Rev. Earth Planet. Sci.32:465-94. Datt B, McVicar TR, Van Niel TG, Jupp DL, Pearlman JS (2003) Preprocessing EO-1 Hyperion hyperspectral data to support the application of agricultural indexes. IEEE Transactions on Geoscience and Remote Sensing 41(6):1246-59. https://doi.org/10.1109/TGRS.2003.813206 Green RO, Dozier J, Roberts D, Painter T (2002) Spectral snow-reflectance models for grain-size and liquid-water fraction in melting snow for the solar-reflected spectrum. Annals of Glaciology 34:71-3. https://doi.org/10.3189/172756402781817987 Jenson, J.R., (2005) Introductory Digital Image Processing: a Remote Sensing Perspective.,(Pearson Prentice Hall: Upper Saddle River, NJ. Kokhanovsky A (2013) Spectral reflectance of solar light from dirty snow: a simple theoretical model and its validation. The Cryosphere 7(4):1325-31. https://doi.org/10.5194/tc-7-1325-2013 Kaufman YJ, Wald AE, Remer LA, Gao BC, Li RR, Flynn L (1997) The MODIS 2.1-/spl mu/m channel-correlation with visible reflectance for use in remote sensing of aerosol. IEEE transactions on Geoscience and Remote Sensing 35(5):1286-98. https://doi.org/10.1109/36.628795 Kulkarni AV, Srinivasulu J, Manjul SS, P M (2002) Field based spectral reflectance studies to develop NDSI method for snow cover monitoring. Journal of the Indian Society of Remote Sensing 30:73-80. https://doi.org/10.1007/BF02989978 Kruse FA, Lefkoff AB, Boardman JW, Heidebrecht KB, Shapiro AT, Barloon PJ, Goetz AF (1993) The spectral image processing system (SIPS)—interactive visualization and analysis of imaging spectrometer data. Remote sensing of environment 44(2-3):145-63. https://doi.org/10.1016/0034-4257(93)90013-N Kumar R, Bahuguna IM, Ali SN, Singh R (2020) Lake inventory and evolution of glacial lakes in the Nubra-Shyok basin of Karakoram Range. Earth Systems and Environment 4:57-70. https://doi.org/10.1007/s41748-019-00129-6 Li W, Stamnes K, Chen B, Xiong X (2001) Snow grain size retrieved from near‐infrared radiances at multiple wavelengths. Geophysical Research Letters 28(9):1699-702. https://doi.org/10.1029/2000GL011641 Marks D, Dozier J (1992) Climate and energy exchange at the snow surface in the alpine region of the Sierra Nevada: 2.Snow cover energy balance. Water Resources Research 28(11):3043-54. https://doi.org/10.1029/92WR01483 Matthew MW, Adler-Golden SM, Berk A, Richtsmeier SC, Levine RY, Bernstein LS, Acharya PK, Anderson GP, Felde GW, Hoke ML, Ratkowski AJ (2000) Status of atmospheric correction using a MODTRAN4-based algorithm. In Algorithms for multispectral, hyperspectral, and ultraspectral imagery VI (Vol. 4049, pp. 199-207). SPIE. https://doi.org/10.1117/12.410341 Matthew MW, Adler-Golden SM, Berk A, Richtsmeier SC, Levine RY, Bernstein LS, Acharya PK, Anderson GP, Felde GW, Hoke ML, Ratkowski AJ (2000) Status of atmospheric correction using a MODTRAN4-based algorithm. In Algorithms for multispectral, hyperspectral, and ultraspectral imagery VI (Vol. 4049, pp. 199-207). SPIE. https://doi.org/10.1117/12.410341 Moughal, T.A (2013) Hyperspectral image classification using support vector machine. In journal of physics: conference series (Vol. 439, No. 1, p. 012042). IOP Publishing. Mudbhari, D., Kansal, M. L., & Kalura, P. (2022). Impact of climate change on water availability in Marsyangdi river basin, Nepal. Quarterly Journal of the Royal Meteorological Society, 148(744), 1407-1423. Negi HS, Jassar HS, Saravana G, Thakur NK, Snehmani, Ganju A (2013) Snow-cover characteristics using Hyperion data for the Himalayan region. International journal of remote sensing 34(6):2140-61. https://doi.org/10.1080/01431161.2012.742213 Nolin AW, Dozier J (2000) A hyperspectral method for remotely sensing the grain size of snow. Remote sensing of Environment 74(2):207-16. https://doi.org/10.1016/S0034-4257(00)00111-5 Negi HS, Kokhanovsky A (2011) Retrieval of snow grain size and albedo of western Himalayan snow cover using satellite data. The Cryosphere 5(4):831-47. https://doi.org/10.5194/tc-5-831-2011 Negi HS, Kulkarni AV, Semwal BS (2009) Study of contaminated and mixed objects snow reflectance in Indian Himalaya using spectroradiometer. International Journal of Remote Sensing 30(2):315-25. https://doi.org/10.1080/01431160802261197 Negi HS, Singh SK, Kulkarni AV, Semwal BS (2010) Field-based spectral reflectance measurements of seasonal snow cover in the Indian Himalaya. International Journal of Remote Sensing 31(9):2393-417. https://doi.org/10.1080/01431160903002417 Petropoulos GP, Kalaitzidis C, Vadrevu KP (2012) Support vector machines and object-based classification for obtaining land-use/cover cartography from Hyperion hyperspectral imagery. Computers & Geosciences 41:99-107. https://doi.org/10.1016/j.cageo.2011.08.019 Painter TH, Dozier J, Roberts DA, Davis RE, Green RO (2003) Retrieval of sub pixel snow-covered area and grain size from imaging spectrometer data. Remote Sensing of Environment 85(1):64-77. https://doi.org/10.1016/S0034-4257(02)00187-6 Qiu F, Abdelsalam M, Thakkar P (2006) Spectral analysis of ASTER data covering part of the Neoproterozoic Allaqi-Heiani suture, Southern Egypt. Journal of African Earth Sciences 44(2):169-80. https://doi.org/10.1016/j.jafrearsci.2005.10.009 Rogge DM, Rivard B, Zhang J, Sanchez A, Harris J, Feng J (2007) Integration of spatial–spectral information for the improved extraction of endmembers. Remote Sensing of Environment 110(3):287-303. https://doi.org/10.1016/j.rse.2007.02.019 Rowan LC, Mars JC (2003) Lithologic mapping in the Mountain Pass, California area using advanced space borne thermal emission and reflection radiometer (ASTER) data. Remote sensing of Environment 84(3):350-66. https://doi.org/10.1016/S0034-4257(02)00127-X Sarkar T, Kannaujiya S, Taloor AK, Ray PK, Chauhan P (2020) Integrated study of GRACE data derived interannual groundwater storage variability over water stressed Indian regions. Groundwater for sustainable development 10:100376. https://doi.org/10.1016/j.gsd.2020.100376 Scambos TA, Haran TM, Fahnestock MA, Painter TH, Bohlander J (2007) MODIS-based Mosaic of Antarctica (MOA) data sets: Continent-wide surface morphology and snow grain size. Remote sensing of environment 111(2-3):242-57. https://doi.org/10.1016/j.rse.2006.12.020 Singh SK, Kulkarni AV, Chaudhary BS (2010) Hyperspectral analysis of snow reflectance to understand the effects of contamination and grain size. Annals of Glaciology 51(54):83-8. https://doi.org/10.3189/172756410791386535 Warren SG (1982) Optical properties of snow. Reviews of Geophysics 20(1):67-89. https://doi.org/10.1029/RG020i001p00067 Wiscombe WJ, Warren SG (1980) A model for the spectral albedo of snow. I: Pure snow. Journal of Atmospheric Sciences 37(12):2712-33. https://doi.org/10.1175/1520-0469(1980)0372.0.CO;2 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 14 Apr, 2025 Read the published version in Applied Geomatics → Version 1 posted Editorial decision: Revision requested 12 Oct, 2024 Reviews received at journal 07 Oct, 2024 Reviews received at journal 04 Oct, 2024 Reviewers agreed at journal 27 Sep, 2024 Reviewers agreed at journal 27 Sep, 2024 Reviewers agreed at journal 10 Sep, 2024 Reviewers invited by journal 10 Sep, 2024 Editor assigned by journal 06 Sep, 2024 Submission checks completed at journal 06 Sep, 2024 First submitted to journal 04 Sep, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5031527","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":363088897,"identity":"3b375c6e-b39b-49cc-ba8b-6350b9b5c840","order_by":0,"name":"Manish Rawat","email":"data:image/png;base64,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","orcid":"","institution":"Indian Institute of Technology Roorkee","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Manish","middleName":"","lastName":"Rawat","suffix":""},{"id":363088898,"identity":"a57ca9b6-87d6-4bdd-a52b-27bd940504e5","order_by":1,"name":"Ashish Pandey","email":"","orcid":"","institution":"Indian Institute of Technology Roorkee","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ashish","middleName":"","lastName":"Pandey","suffix":""},{"id":363088899,"identity":"cdf929e1-09ff-4e09-871a-4bc1b8c19043","order_by":2,"name":"Dhananjay Paswan Das","email":"","orcid":"","institution":"Indian Institute of Technology Roorkee","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Dhananjay","middleName":"Paswan","lastName":"Das","suffix":""},{"id":363088900,"identity":"8f2b6216-8358-4deb-84db-40a97199dd4c","order_by":3,"name":"Praveen Kumar Gupta","email":"","orcid":"","institution":"Space Application Centre, Indian Space Research Organisation","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Praveen","middleName":"Kumar","lastName":"Gupta","suffix":""}],"badges":[],"createdAt":"2024-09-04 12:12:35","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5031527/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5031527/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s12518-025-00627-5","type":"published","date":"2025-04-14T15:57:19+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":66239593,"identity":"21fd0945-6be2-464c-b00a-fefb616e5bbd","added_by":"auto","created_at":"2024-10-09 06:18:32","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":316171,"visible":true,"origin":"","legend":"\u003cp\u003eStudy area- Bhinlgana basin, Uttarakhand in PRISMA satellite image\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/52038e27991fed0426f5f2b6.png"},{"id":66239591,"identity":"ad6bd1e9-ec5e-44d1-a353-1bb133f2476b","added_by":"auto","created_at":"2024-10-09 06:18:32","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":131512,"visible":true,"origin":"","legend":"\u003cp\u003eFlow chart of Methodology\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/c09ade2160e7dc1014db193a.png"},{"id":66239587,"identity":"607e8bbb-d037-4cdb-9088-85d8400ecd5f","added_by":"auto","created_at":"2024-10-09 06:18:32","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":191810,"visible":true,"origin":"","legend":"\u003cp\u003ePRISMA product (a) band with strip having low DN value (b) De-striped band with changed DN value after apply local filter\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/5eaa98493a72a87fa080bb17.png"},{"id":66240668,"identity":"5b517508-0678-40a9-a44b-350a31000f44","added_by":"auto","created_at":"2024-10-09 06:34:32","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":51793,"visible":true,"origin":"","legend":"\u003cp\u003ea) Spectral profile (Z-profile) of randomly selected snow pixels before atmospheric correction b) Spectral profile (Z-profile) of randomly selected snow pixels after atmospheric correction\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/2fb570955800c559f2449130.png"},{"id":66240458,"identity":"29f8b78b-2163-4fec-a321-ca9b86f49328","added_by":"auto","created_at":"2024-10-09 06:26:32","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":42479,"visible":true,"origin":"","legend":"\u003cp\u003eDiagnostic ice absorption near 1030nm wavelength for grain size estimation\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/04cae1e2ca0e0081aff83162.png"},{"id":66240461,"identity":"ef7678ae-fd97-485d-a6af-6ce71da49a55","added_by":"auto","created_at":"2024-10-09 06:26:49","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":41791,"visible":true,"origin":"","legend":"\u003cp\u003eDecision tree for grain size estimation\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/8112debb7116ceab31c54a9f.png"},{"id":66239598,"identity":"32f20477-5065-4042-83f0-80c80c0cc32c","added_by":"auto","created_at":"2024-10-09 06:18:32","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":207139,"visible":true,"origin":"","legend":"\u003cp\u003ea) Snow grain extraction by using G.I index, b) Snow grain size classified map based on grain index threshold values\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/aa0334741e67079a3a86f524.png"},{"id":66241446,"identity":"f2765dc1-62f6-494c-b02e-cdb3f801ffe6","added_by":"auto","created_at":"2024-10-09 06:42:32","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":46502,"visible":true,"origin":"","legend":"\u003cp\u003eHistogram of snow grain index with two bands\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/ce3837a92c30eb8020379ebc.png"},{"id":66240452,"identity":"ab710cde-9350-41da-bed6-c51a9f7684e5","added_by":"auto","created_at":"2024-10-09 06:26:32","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":184629,"visible":true,"origin":"","legend":"\u003cp\u003eSnow cover map based on NDSI threshold a) NDSI\u003cu\u003e\u0026gt;\u003c/u\u003e0.40, b) NDSI\u003cu\u003e\u0026gt;\u003c/u\u003e0.70 using bands 25(587.79nm), 127(1646.96nm)\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/410a6bcd5f235196c6ac8e19.png"},{"id":66239589,"identity":"57456226-c177-428d-838d-deba79f5ef0b","added_by":"auto","created_at":"2024-10-09 06:18:32","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":28482,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation between eigenvalue and Eigen number using MNF transformation\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/09ef2d27ed810d65daa39df1.png"},{"id":66239601,"identity":"de91e11d-4dbd-418e-b904-3e31f29bd164","added_by":"auto","created_at":"2024-10-09 06:18:32","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":29732,"visible":true,"origin":"","legend":"\u003cp\u003ePPI curve of dataset\u003c/p\u003e","description":"","filename":"11.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/c3d9bf52a6db22bb9a29ea69.png"},{"id":66240459,"identity":"69c7d45a-8fc7-43d6-8a7c-7be579dfc98b","added_by":"auto","created_at":"2024-10-09 06:26:32","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":161886,"visible":true,"origin":"","legend":"\u003cp\u003e3-D visualization for endmember selection of different classes in the PRISMA image\u003c/p\u003e","description":"","filename":"12.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/f063b7b11ae60575d1baa38c.png"},{"id":66639887,"identity":"901a08c7-45e0-4fd2-9cd4-f0c832290d0d","added_by":"auto","created_at":"2024-10-15 06:02:12","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":24392,"visible":true,"origin":"","legend":"\u003cp\u003eSpectral angle between unknown and reference spectra in SAM algorithm\u003c/p\u003e","description":"","filename":"13.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/0e0bb2bd238503598494db1a.png"},{"id":66241447,"identity":"4655cbf3-4f27-49be-ab45-262e44ba9af5","added_by":"auto","created_at":"2024-10-09 06:42:32","extension":"png","order_by":14,"title":"Figure 14","display":"","copyAsset":false,"role":"figure","size":535788,"visible":true,"origin":"","legend":"\u003cp\u003eSpectral reflectance of different snow grain size\u003c/p\u003e","description":"","filename":"14.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/a58b51b5241fda26d2ff411c.png"},{"id":66240453,"identity":"42f209a0-122b-4e00-bcf3-26a7815f8130","added_by":"auto","created_at":"2024-10-09 06:26:32","extension":"png","order_by":15,"title":"Figure 15","display":"","copyAsset":false,"role":"figure","size":424576,"visible":true,"origin":"","legend":"\u003cp\u003eSAM classified image for snow-glacier region\u003c/p\u003e","description":"","filename":"15.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/611bfad9cbb037154739a623.png"},{"id":66241558,"identity":"b5afce6f-8b40-40c6-baa9-4465f9d6b40d","added_by":"auto","created_at":"2024-10-09 06:50:32","extension":"png","order_by":16,"title":"Figure 16","display":"","copyAsset":false,"role":"figure","size":428628,"visible":true,"origin":"","legend":"\u003cp\u003eSpectral signature of distinct Glacier features\u003c/p\u003e","description":"","filename":"16.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/e42820a722c7b393f9a56f4d.png"},{"id":66239603,"identity":"97ebcdad-1ce4-47ff-bd76-b069a1d3d94b","added_by":"auto","created_at":"2024-10-09 06:18:33","extension":"png","order_by":17,"title":"Figure 17","display":"","copyAsset":false,"role":"figure","size":363888,"visible":true,"origin":"","legend":"\u003cp\u003eSnow grain size map generated by (a) GSI (b) SAM\u003c/p\u003e","description":"","filename":"17.png","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/61c408270f8122d179ed4f49.png"},{"id":81050957,"identity":"7577392e-920a-4d2c-92eb-5b4124f3d5ad","added_by":"auto","created_at":"2025-04-21 16:08:45","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4110280,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5031527/v1/6127e15d-6a7d-4d70-9314-867651a44acf.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eCharacterization and Retrieval of Snow Grain Size in the Upper Himalayan Region Using Hyperspectral Prisma Data\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eSnow and glaciers are valuable natural resources in mountainous regions, influencing major rivers, snowmelt runoff, regional climate, and snow avalanches. Snow plays a significant role in hydrological and climatic models, especially in a snowmelt runoff simulation. However, due to rugged terrain and harsh weather, snow mapping and its physical characteristics with field-based surveys is extremely challenging. In such an environment, remote sensing plays an important role because it gives high temporal and spatial information about the surface of the earth and its physical characteristics. Significant mass loss of glaciers has been caused by continuous climate change in high and rough terrains throughout the world, and several of them are currently at risk (Kumar et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Ahmed et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2021\u003c/span\u003ea; Sarkar et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The snow cover extent and its spatial extent, as well as albedo, are the most analytical snow parameters for energy-mass balance modeling (Bloschl, 1991; Dozier and Painter \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). During the winter season, more than 40% of the northern hemisphere region is covered by snow, therefore the pertaining snow cover information can be used for various applications i.e. snow melt run-off modeling, snow water equivalent estimation, and climatic modeling. The complex meshing of climate and geological processes is responsible for the degradation of natural resources in the Himalayan ecosystem (Mudbhari et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Due to the Himalayan challenging terrain, data collection by conventional methods has its limitations, therefore real-time satellite data is a useful tool for mapping the extent of snow cover. Seasonal snow cover variations are a key indicator of climate change in the Himalayan region. At the surface, the assessment of snow grain size holds significant importance in the modeling of snow albedo, which serves as a primary determinant of both snow energy balance and the timing of snowmelt (Marks and Dozier \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1992\u003c/span\u003e). However, due to various landscapes and heterogeneous snow and vegetation cover, the snowpack frequently gets poorly sampled. Therefore, it is necessary to develop an optimal framework for mapping snow grain size and snow cover spatially using advanced remote sensing techniques. One of the most valuable snow physical parameters is snow grain size which indicates the advancement in snow metamorphism. Snow grain size represents a fundamental characteristic of snow that influences its reflectivity (Wiscombe and Warren \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e1980\u003c/span\u003e), and it serves as a means to describe the processes of snow metamorphism and stratigraphy (Colbeck \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e1991\u003c/span\u003e). In the snow avalanches, snowpack phenomenon stability has been analyzed with the help of snow grain size condition, as fine-grain size snow cover has more strength than coarse-grain snow cover. Snow metamorphism is related to the transformation and structural changes that snow undergoes over time as a result of environmental conditions such as temperature, humidity, and wind.\u003c/p\u003e \u003cp\u003eHyperspectral remote sensing is one of the most advanced technology for surface features extraction and its mapping is based on the identification of the spectral signature of different materials. For example, hyperspectral imagery has been used for the identification of distinct snow properties such as snow cover extent, snow grain size, and different surface material present on the glacier region that can be collected by the hyperspectral sensor, and this strengthens the identification of different features among similar land cover class than the multispectral data. Assessment of hyperspectral data is a quite challenging process due to extensive spatial variability of the different spectral signatures of different land use land cover classes, atmospheric implications, and large data dimensionality (Moughal \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). To overcome this challenging process, semi-automatic techniques (SAM), bi-spectral methods, and machine learning techniques like Support Vector Machines (SVM) can be adopted. The hyperspectral dataset contains multiple narrow continuous spectral bands from the range of visible to the shortwave infrared region, acquiring a huge amount of spectral information at each wavelength region for feature identification (Petropoulos et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). The albedo of snow varies with the snow surface's physical and textural properties that influence the snowmelt phenomenon (Casacchia et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). The reflectance of fresh snow in the visible region is observed to be approximately 90% and will decrease continuously at longer wavelengths (Warren and Winscombe 1980). Fresh snow has a higher albedo in the visible range of the electromagnetic spectrum and gradually decreases with snow aging (Negi et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). The snow melting causes grains to grow clusters and act as a large single grain (Dozier et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e1981\u003c/span\u003e; Warren \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e1982\u003c/span\u003e). Water content presence between ice crystals leads snow to act as optically large grain size (Colbeck S.C 1979). In the Himalayan region, snow grain size is broadly classified by estimating the snow grain index method using the central band wavelength at 440 and 1030 nm based on ground-based Spectroradiometer data (Negi et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Multispectral and hyperspectral remote sensing, covering a range of wavelengths from 0.4 to 15 µm, enables the retrieval of various properties such as snow-covered area, albedo, grain size, the presence of liquid water near the surface, and temperature (Dozier and Painter \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2004\u003c/span\u003e) Finer the grain size, the higher is the reflectance observed at this wavelength (Painter et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). In (Negi et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) compared snow grain sizes in the Himalayan region using three different methods: the spectral angle method, the grain index, (SAM), and the ART theory method, based on Hyperion data.\u003c/p\u003e \u003cp\u003eSnow reflectance in the visible region is strongly influenced by absorbing impurities and is virtually independent of grain size. In contrast, snow reflectance in the NIR region primarily depends on snow grain size, with a decrease in reflectance occurring as grain size increases. Snow reflectance is highly sensitive to grain size within a wavelength range of 1.0- 1.3µm, extending the diagnostic ice absorption features at 1.03µm and 1.26µm (Negi and Kokhanovsky \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). The maximum deviation in reflectance or sensitive wavelength for snow grain size was examined in the NIR wavelength region as 1030, 1050, and 1240 nm, as the wavelength is more absorptive due to ice presence. According to (Negi and Kokhanovsky \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), these wavelengths have been used to estimate the size of the snow grains. The snow grain index was proposed by (Kokhanovsky et al. 2013) and is a bi-spectral technique, with one region in the visible and the other in the NIR region (440 and 1030/1240). Using AVIRIS data for the US region, a method was developed for quantitatively retrieving snow characteristics (Green et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). Snow Grain size mapping was done using NDSGI techniques between MODIS band 1 (620–670 nm) and MODIS Band 2 (841– 876) radiances of pure snow pixels (Scambos et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). The snow grain size was retrieved from AVIRIS hyperspectral data using multiple wavelengths of 860, 1050, 1240, and 1730nm respectively (Le et al. 2001).\u003c/p\u003e \u003cp\u003eThe objective of this study is to utilize the PRISMA hyperspectral dataset to retrieve various snow grain sizes in the Bhilangana basin of the upper Himalayan region in Uttarakhand on February 18, 2020. This was done using the Spectral Angle Mapper (SAM) classification method and the Snow Grain Index (SGI) method. Additionally, the spectral reflectance of different glacier features was generated using the SAM classification technique on PRISMA hyperspectral imagery and validated against spectra from the USGS spectral library for accurate mapping. The estimated parameters derived from this data can be instrumental in advancing our understanding of regional climatology by providing insights into snowpack characteristics and their influence on local and global climate patterns.\u003c/p\u003e \u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e "},{"header":"2. Study area","content":"\u003cp\u003eThe study area is located in the upper Himalayan range. It is located between the latitude of 31°1'39.31\"N to 30°42' 8.30\"N and longitude of 78°43'29.61\"E to 79°50'45.55\"E, with an elevation ranging from 4000 to 5800 meters. This study focuses on the PRISMA data scene within the Bhilangana basin in the upper Himalayan region, which falls under the Uttarkashi district of Uttarakhand, India (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), covering an area of 886.56 km\u003csup\u003e2\u003c/sup\u003e to retrieve different snow grain size characteristics. This region experiences a cold and arid climate and experiences heavy snowfall during the winter season while most of the terrain in this area lacks significant vegetation cover, which can have implications for the study of snow grain size characteristics. Additionally, the terrain in this area is largely devoid of substantial vegetation, which influences snow dynamics. The absence of vegetation affects snow accumulation patterns, snow cover duration, and the physical properties of the snowpack. These factors are crucial for understanding snow grain size characteristics and their implications for hydrological and climatic processes in this high-altitude environment.\u003c/p\u003e"},{"header":"3. Data used and sensor description","content":"\u003cp\u003eThe study has been carried out by using PRISMA (PRecursore IperSpettrale Della Missione Applicativa), the latest Hyperspectral satellite sensor of the Italian Space Agency (ASI), which was launched into orbital on March 22, 2019, and is positioned in a sun-synchronous Low Earth Orbit at an altitude of 620 km. It operates on a repeat cycle of approximately 29 days. The PRISMA hyperspectral payload consists of a push broom sensor having 230 contiguous spectral bands (400 to 2500 nm wavelength region) of which 66 bands are in the Visible Near Infrared (VNIR) range and 164 bands in the Short Wave Infrared (SWIR) range with coherent, and having a spectral resolution is about 12 nm narrow bandwidth and a 30 m spatial resolution with a swath width of 30 km. The PRISMA L1 data products distributed with HDF5 file format were downloaded from the PRISMA portal (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://prisma.asi.it\u003c/span\u003e\u003cspan address=\"https://prisma.asi.it\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and re-projects with a geographic look-up table (GLT) for the correction of Bowtie artifacts associated with missing data from the ENVI software.\u003c/p\u003e"},{"header":"4. Methodology","content":"\u003cp\u003eThe first step in processing the PRISMA Hyperspectral dataset involved converting it from HDF5 to HDR format using the R package within ENVI software. This dataset contains a continuous spectrum with a high-dimensional volume of spectral data. Due to the large numbers of the band available in the PRISMA spectral dataset, there are radiometric interferences, low Signal-to-Noise Ratio (SNR), and heavy water absorption influences in several spectral bands, as a result, a total of 24 bands were dropped from 230 original bands, Finally, 206 bands are calibrated in this dataset, which requires accurate pre-processing for operating the noise control in several optimized bands. During the pre-processing phase, the visual interpretation of the remaining 206 spectral bands involved removing uncalibrated bands, de-striping a few SWIR bands, and applying atmospheric corrections using ENVI\u0026apos;s FLAASH model. Before performing atmospheric correction on the PRISMA data, we eliminated uncalibrated images or bands in the SWIR region that were either noisy or contained no data. We selected radiometrically calibrated bands with band numbers ranging from 9 to 57 in the VNIR range and from 78 to 220 in the SWIR wavelength range, resulting in a total of 158 spectral channels used in the PRISMA Hyperspectral dataset. The primary objective of this study is to analyze the spectral signatures of snow-glacier features within the PRISMA scene of the Bhilangana basin in the upper Himalayan region and ascertain in mapping different snow features using hyperspectral remote sensing techniques.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eGiven the challenging conditions of the high rugged terrain and extremely cold climate during the winter season, ground truth data is lacking. Therefore, in this study, we compared the snow grain size acquired through the Snow Grain Size Index method with the results of the spectral angle classification method. The comprehensive methodology employed in this study is depicted in Fig 2.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.1. Elimination of Bad Band in Dataset\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDuring the pre-processing of PRISMA hyperspectral imagery, certain bands were identified as \u0026apos;bad bands. These bands exhibited either a lack of data or strong absorption due to water vapor. The noise management strategy for the bands involves identifying \u0026apos;bad\u0026apos; pixels, mitigating vertical striping effects, and compensating for atmospheric distortions in the data. It also aims to minimize the compounding effects through image processing (Datt et al. 2003).To identify and remove these undesirable bands, we performed manual visual inspection, carefully examining each spectral band for anomalies. Out of the initial 230 bands present in the PRISMA data, a total of 24 bands were identified as \u0026apos;bad bands\u0026apos; and only 206 bands were considered for further analysis. Table 1 shows the list of eliminated bands.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1.\u0026nbsp;\u003c/strong\u003eList of unusual bands of the PRISMA sensor\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"378\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 34.3915%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBands number\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65.6085%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCondition\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 34.3915%;\"\u003e\n \u003cp\u003e1, 48,84,85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65.6085%;\"\u003e\n \u003cp\u003eVertical stripping bands\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 34.3915%;\"\u003e\n \u003cp\u003e100-104\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65.6085%;\"\u003e\n \u003cp\u003eVertical stripping bands\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 34.3915%;\"\u003e\n \u003cp\u003e136\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65.6085%;\"\u003e\n \u003cp\u003eVertical stripping band\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 34.3915%;\"\u003e\n \u003cp\u003e149 - 156\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65.6085%;\"\u003e\n \u003cp\u003eVertical stripping bands\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 34.3915%;\"\u003e\n \u003cp\u003e159, 203\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65.6085%;\"\u003e\n \u003cp\u003eVertical stripping bands\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 34.3915%;\"\u003e\n \u003cp\u003e161,164, 229,230\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65.6085%;\"\u003e\n \u003cp\u003eWater vapour absorption bands\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e4.2 Destriping of PRISMA Bands\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThere are vertical stripes observed in a few bands of the PRISMA dataset consisting of white or dark striping pixels. The bright pixels exhibit a significantly higher sensor response, while the dark vertical strips show minimal or no sensor response. These vertical stripes accommodating abnormal pixels have lower DN values or no information when compared to neighbouring pixels (Fig 3). These distorted vertical strips can be replaced by their adjacent DN pixels value using the spatial editor tool available in ENVI software\u003cstrong\u003e.\u003c/strong\u003e This is a critical issue in remote sensing, as it can affect the accuracy of interpretations and analyses derived from the PRISMA dataset\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.3. Radiometric Correction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRescaling of dataset\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePRISMA data consists of 230 spectral bands ranging from 402.40 to 2496.86 nanometres (nm) with spectral resolution of 12 nm and 30m spatial resolution. The sensor consists of two array detectors, VNIR (Visible Near Infrared) and SWIR (Short Wave Infrared) bands. Among the 230 bands, only 206 bands are calibrated for the VNIR, which wavelength region from 402.40 nm to 998.36 nm, and a SWIR range from 919.13 to 2496.86 nm wavelength regions. To convert Digital Numbers (DN) to radiance values, the data needs to be rescaled. As a result, the VNIR and SWIR bands in the PRISMA image have considered two scaling factors. The scaling factors of 40 for the VNIR band and 80 for the SWIR bands are provided as input for the FLAASH Model. When converting DNs to radiance values for subsequent analysis and interpretation, calibrating the bands and incorporating scaling factors are essential to ensure the accuracy and reliability of the data. This procedure is an important part of the data processing workflow.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.4. FLAASH for Atmospheric Correction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFLAASH (Fast Line-of-Sight Atmospheric Analysis of Spectral Hypercubes) is an atmospheric correction model that uses a radiative transfer program to reduce atmospheric inaccuracies. The FLAASH model based on the Moderate Resolution Atmospheric Transmission version 4 (MODTRAN4) radiative transfer algorithm was used to perform atmospheric correction (Anderson et al. 1999; Cooley et al. 2002). The input radiance image for FLAASH should contain a radiometrically calibrated radiance image. This model is rigorous and requires various parameters for data scaling as well as to define other information such as scene centre location, sensor information, atmospheric, and aerosol model that are to be used in MODTRAN-4 radiative transfer code which is used in FLAASH, the various parameters used for the atmospheric correction are listed below in Table 2. FLAASH allows us to understand the different standard and derived atmospheric and aerosol models attempted for the Hyperion dataset (Matthew et al. 2000) and provides water vapor estimates for each atmospheric model as shown in Table 3. It has been observed that the water vapor data derived from the atmospheric correction model, particularly the water vapor absorption feature at 820 nm wavelength, was sufficient for estimating the water vapor quantity column for each pixel. As a result, the atmospheric model (Mid-latitude winter) was used to determine the water vapor quantity for each pixel in the image, and a 2-band KT aerosol model was employed for atmospheric correction (Kaufman et al. 1997). Consequently, we selected the atmospheric model \u0026apos;Mid-latitude summer\u0026apos; and the aerosol model \u0026apos;rural\u0026apos; for our specific analysis. The FLAASH atmospheric correction results for the PRISMA image are displayed in their respective spectral profiles in Fig. 4 with snow pixel Spectral profile (Z-profile), a) before Atmospheric corrections and b) after Atmospheric FLAASH corrections.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e. Various parameters were used in the FLAASH model\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eScene location\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e31\u0026deg;1\u0026rsquo;39.79 N and 78\u0026deg;43\u0026rsquo;29.13 E\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eSensor altitude\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e620.00 km\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003ePixel size\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e30.00 m\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eAcquisition date\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e18-Feb-2020\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eAtmospheric model\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eMid-latitude Winter\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eWater absorption feature\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e820 nanometers\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eAerosol model\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eRural\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eAerosol retrieval\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e2-Band(K-T)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eInitial visibility\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e40 km\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eWidth polishing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eAerosol scale height\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e1.50 km\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eModtran multiscatter model\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eScaled DISCORT\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003eNumber of Discort stream\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 50%;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cstrong\u003eTable 3\u003c/strong\u003e. Different atmospheric model parameters were attempted to correct the atmospheric effects in PRISMA image of the study area\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"643\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStudy area\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAtmospheric model\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 96px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAerosol model\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 228px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eObtained water vapors(gm/cm\u003csup\u003e2\u003c/sup\u003e)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"6\" valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003ePRISMA scene in\u003c/p\u003e\n \u003cp\u003eBhinlgana basin\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(Feb 2020)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003eMid-Latitude summer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 96px;\"\u003e\n \u003cp\u003e2- Band KT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 228px;\"\u003e\n \u003cp\u003e0.1286\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003eSub-Arctic summer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 96px;\"\u003e\n \u003cp\u003e2- Band KT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 228px;\"\u003e\n \u003cp\u003e0.0474\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003eSub-Arctic winter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 96px;\"\u003e\n \u003cp\u003e2- Band KT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 228px;\"\u003e\n \u003cp\u003e0.0483\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003eMid-Latitude winter (1135nm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 96px;\"\u003e\n \u003cp\u003e2- Band KT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 228px;\"\u003e\n \u003cp\u003e0.0511\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003eMid- Latitude winter(820nm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 96px;\"\u003e\n \u003cp\u003e2- Band KT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 228px;\"\u003e\n \u003cp\u003e0.0535\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 207px;\"\u003e\n \u003cp\u003eTropical 1135nm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 96px;\"\u003e\n \u003cp\u003e2- Band KT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 228px;\"\u003e\n \u003cp\u003e0.0831\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e4.5. Snow Grain Size Extraction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSnow grain size is one of the important physical parameters of snow characteristics that is accountable for snow metamorphic characteristics such as spectral reflectance, snow avalanches, snow melting rate, snow depth, etc. Snow grain size is a vital parameter whose understanding is required across different disciplines, from snow chemistry and gas interactions to Earth\u0026apos;s radiation balance. In this study, the Grain Size Index (GSI) and Spectral Angle Mapping (SAM) were employed to classify snow grain size. The snow grain size map generated by the GSI method, as proposed by (Negi et al. 2010) is based on ground truth data collected by a hyperspectral sensor and was further used for validation against the SAM-classified image. These two methods SAM and GSI are qualitative. This implies that they provide relative assessments of snow particle size, allowing researchers to classify snow grains as fine, medium, and coarse based on their spectral properties.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.5.1 Snow Grain size estimation using Grain Index method\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn studies related to snow avalanches, snow grain size is closely linked to the stability of the snowpack. Finer snow grain sizes tend to exhibit greater stability compared to coarser snow grains. It has been observed that the reflectance of snow in the near-infrared wavelength range decreases as the snow grain size increases. The spectral reflectance of snow in the wavelength spectrum range of 1030 nm to 1045 nm is found to be the most sensitive region for the estimation of snow grain size variation. In this range, finer snow particles result in higher reflectance measurements (Nolin and Dozier 2000; Painter et al. 2003). The interpreted snow reflectance characteristics to understand the effect of soil contamination and snow grain size. To analyze the shape and depth of the absorption peak between 1025 nm and 1040 nm as snow grain size increased. Reflectance in the visible region is significantly affected by absorbing impurities and shallow snow depth. However, in this wavelength range, snow grain size variability is nearly independent. Contamination reduces snow reflectance in the visible region, while surface snow grain size is more susceptible to changes in the near-infrared region (Singh et al. 2010). They measured different snow grain sizes using snow reflectance at 440\u0026ndash;590 nm in the visible range and 1040\u0026ndash;1050 nm in the NIR spectrum (Negi et al. 2010). They found that as grain index threshold values increased, there was a corresponding increase in snow particle size. Additionally, it was observed that due to snow aging and variations in snow grain size, a diagnostic ice absorption depth feature appeared near 1030 nm, as shown in Fig 5. In this wavelength region, snow grain size exhibited higher sensitivity for reflectance.\u003c/p\u003e\n\u003cp\u003eSnow grain size mapping in this study has been derived from the grain size index (GSI) method. The snow grain index model is based on the bi-spectral method, which consists of visible (dependence on contamination) and near-infrared (dependence on grain size) regions for retrieving snow grain size. This method is suitable for clean, dry snow in the upper Himalayan region, which is heterogeneous in terms of spatial-temporal changes in snow grain size. In this research, snow grain size formulation has been done by using the PRISMA bands 6 (visible channel wavelength 441.63 nm) and 69 (NIR channel wavelength 1028.79 nm) as shown in the following equation:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" width=\"451\" height=\"41\"\u003e\u003c/p\u003e\n\u003cp\u003eGrain size was classified into fine, medium, and coarse classes having snow grain index threshold values as follows: 0.000-0.170, 0.170-0.260, and 0.260-0.350, as recommended by (Negi et al., 2010) and given in Table 4. The extraction of snow grain size using the Grain Index and its classified map, based on threshold values of the Grain Index, are displayed in Figure 7 (a and b). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 4.\u003c/strong\u003e The snow grain size was categorised into three different classes\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"575\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8.15972%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eS. No\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 27.6042%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClass\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31.0764%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRange of snow grain size (mm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 33.1597%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eThreshold for snow grain size index\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8.15972%;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 27.6042%;\"\u003e\n \u003cp\u003eFine\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31.0764%;\"\u003e\n \u003cp\u003e\u0026lt; 0.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 33.1597%;\"\u003e\n \u003cp\u003e0.0 \u0026ndash; 0.17\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8.15972%;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 27.6042%;\"\u003e\n \u003cp\u003eMedium\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31.0764%;\"\u003e\n \u003cp\u003e0.5 \u0026ndash; 1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 33.1597%;\"\u003e\n \u003cp\u003e0.17 -0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8.15972%;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 27.6042%;\"\u003e\n \u003cp\u003eCoarse\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31.0764%;\"\u003e\n \u003cp\u003e1.0 -2.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 33.1597%;\"\u003e\n \u003cp\u003e0.26 -0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 8.15972%;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 27.6042%;\"\u003e\n \u003cp\u003eUnclassified\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 31.0764%;\"\u003e\n \u003cp\u003eNon-snow-covered area\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 33.1597%;\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eIn this study, a decision tree algorithm, a non-parametric classifier, and a machine learning technique were employed for snow grain size classification using the bi-spectral grain index method, as illustrated in Fig 6. Decision tree algorithms feature a tree-like structure comprising root nodes, internal nodes, and leaves. They evaluate one or more input features in various combinations to aid in the classification process. Using different grain size threshold values, as detailed in Table 4, the decision tree successfully classified snow grain sizes into fine, medium, and coarse classes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.5.2. Band Histogram statistic\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe wavelengths 1028 nm and 441 nm in hyperspectral images are associated with PRISMA band numbers 69 and 06, respectively. After performing an atmospheric correction on the PRISMA imagery, statistical analysis of the imagery data was performed in this study. This analysis brings valuable insight into the physical characteristics of the imagery that can aid in understanding the information contained in the images. The histogram statistics of the PRISMA imagery after atmosphere correction show the average value was 0.266, this statistic shows the mean value of the pixel in the image. It offers an indication of the overall brightness or reflectivity of the scene, and the highest value was 1.00 this value represents the highest pixel intensity within the image. It indicates the highest reflectance or radiance at the given wavelengths. The standard deviation was observed to be 0.225, the standard deviation quantifies the variability or dispersion of pixel intensities in the image. Fig 8 depicts a histogram of snow grain index. A histogram is a useful tool for evaluating an image\u0026apos;s general properties, such as contrast, brightness, and the presence of different materials or features. It is indispensable to obtain a better understanding of the nature of the PRISMA imagery after atmospheric changes by examining these data and visualizing the histogram.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.6. Estimation of Snow Cover Area\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSnow cover classification mapping is challenging to apply in the Himalayan region due to mountain shadows and cloud cover over the snow region (Kulkarni et al. 2002). To address these challenges, the normalized difference snow index (NDSI) can be used to overcome this problem. One more advantage of NDSI techniques is the ability to remove snow from cumulus clouds (Negi et al. 2009). In this study, we pre-processed PRISMA data and employed three wavelength channel combinations consisting of (587.79, 1646.96), (554.53, 1646.96), and (500, 1726.43) nm were used for the NDSI. Generally, NDSI threshold values are set at 0.4 for snow cover mapping. However, to mitigate heavy contamination in snow, the threshold value was raised to 0.6 in this study (Negi, Kulkarni, and Semwal Citation 2009; Negi et al. 2010). Normalized difference snow index (NDSI) was calculated using green and SWIR bands ratio to produce the snow cover map in mountain shadows (Kulkarni et al. 2002). To further refine our analysis, we used a higher threshold value of 0.7 to identify pure snow pixels, as illustrated in Fig 9. Any pixels with values below 0.7 were masked and considered as non-snow-covered regions. Fig 9 displays two distinct snow cover maps of the study area, each generated with different NDSI threshold values: 0.40 and 0.70. As shown in Table 5, the NDSI threshold increases from 0.40 to 0.70, the snow cover area decreases significantly across all band combinations. This is because higher thresholds filter out pixels with lower snow reflectance or mixed land cover, identifying only pure snow areas. The NDSI thresholds used to generate snow cover areas at different wavelength regions and their details are given in Table 5.\u003c/p\u003e\n\u003cp\u003eThe NDSI formula shown in given equation:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" width=\"350\" height=\"40\"\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 5.\u0026nbsp;\u003c/strong\u003eSnow cover area for different spectral ratios at different NDSI threshold values\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"642\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 27.1028%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBands 25(587.79nm), 127(1646.96nm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 37.3832%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBands 21(554.53nm), 127(1646.96nm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 35.514%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBands 14(500.10nm), 135(1726.43nm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.215%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNDSI threshold values\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.8879%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSnow cover area\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(sq. km)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.6916%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNDSI threshold values\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.6916%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSnow cover area\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(sq. km)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.757%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNDSI threshold values\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.757%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSnow cover area\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(sq. km)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.215%;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.8879%;\"\u003e\n \u003cp\u003e822.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.6916%;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.6916%;\"\u003e\n \u003cp\u003e815.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.757%;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.757%;\"\u003e\n \u003cp\u003e778.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.215%;\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.8879%;\"\u003e\n \u003cp\u003e785.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.6916%;\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.6916%;\"\u003e\n \u003cp\u003e777.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.757%;\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.757%;\"\u003e\n \u003cp\u003e676.67\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.215%;\"\u003e\n \u003cp\u003e0.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.8879%;\"\u003e\n \u003cp\u003e705.82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.6916%;\"\u003e\n \u003cp\u003e0.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.6916%;\"\u003e\n \u003cp\u003e696.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.757%;\"\u003e\n \u003cp\u003e0.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.757%;\"\u003e\n \u003cp\u003e452.24\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.215%;\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.8879%;\"\u003e\n \u003cp\u003e518.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.6916%;\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.6916%;\"\u003e\n \u003cp\u003e491.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.757%;\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 17.757%;\"\u003e\n \u003cp\u003e234.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eThe combination of Bands 25 (587.79 nm) and 127 (1646.96 nm) consistently shows the largest snow cover area across all thresholds, indicating that this band pairing is more inclusive of snow cover. The combination of Bands 14 (500.10 nm) and 135 (1726.43 nm) shows the smallest snow cover areas, especially at higher NDSI thresholds, suggesting that this pairing may be better for identifying only the cleanest or freshest snow.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.7. Land Cover Classification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHyperspectral Image Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe following hyperspectral analysis processes, including the Minimum Noise Fraction (MNF) for spectral data reduction, Pixel Purity Index (PPI) for determining spectrally pure pixel values, and subsequently n-dimensional Visualizer for classifying the endmembers directly from the image were applied to the PRISMA dataset. The approach of the Spectral angle mapper (SAM) method has been employed to determine the snow cover features in the glacier region.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.7.1. Minimum Noise Fraction Transformation (MNF)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe minimum Noise Fraction transform technique is an improved version of the Principal Components transform that produces output results by decreasing the signal-to-noise ratio\u0026nbsp;in hyperspectral data. Hyperspectral data consists of different spectral bands that are affected by the low signal-to-noise ratio,\u0026nbsp;especially in the short-wave infrared wavelength spectrum. Noise in hyperspectral images is one of the most common problems that has been observed, it causes interference with feature identification and abundance calculation, so it is necessary to eliminate this complication to acquire good matching spectra. For this determination, Minimum Noise Fraction is the most common processing technique.\u003c/p\u003e\n\u003cp\u003eMNF transformation disintegrates noise from data and determines the inherent dimensionality of the dataset, helping to identify which bands are suitable for further data analysis. There are two steps in the data transformation process. First, the computed covariance matrix is decorrelated to remove noise without band-to-band correlation. Subsequently, the Eigenvalues of each band are calculated, and bands with higher Eigenvalues are selected. Generally, more than one Eigenvalue contains coherent images having valuable information, while the coherent images associated with small eigenvalues contain erroneous information that is dominated by noise (Jensen 2005). The image pixels are presented by Eigenvalues. The MNF technique reduces the dimensionality of large data by considering the coherent images for forwarding data analysis. Fig 10 shows the relation between\u0026nbsp;Eigen number and eigenvalues i.e. MNF band images used to assess the dimensionality of the data and identify the noise-affected band (Qiu et al. 2006). In this study, the first 20 MNF bands containing the most coherent data have been chosen based on the Eigenvalues, while the remaining MNF bands would have noise-affected data that make them unsuitable for data analysis\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAfter visualization of each MNF transformed image, only 20 MNF bands have been selected for subsequent analysis as they consist of higher Eigen containing more feature information.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.7.2 Pixel Purity Index (Spatial\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;data reduction technique)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe processing of the Pixel Purity Index in ENVI has been used to identify the purest pixel from the mass majority of pixels representing mixed pixels in the hyperspectral image. These pure pixels are also known as spectral endmember and their pure spectral signatures. Pixel spectra are considered as an image endmember that lies in an n-dimensional space (Rogge et al. 2007). After eliminating noise and dimensionality reduction of data from MNF transformation techniques. The top 20 coherent MNF bands have been incorporated into the Pixel Purity Index (PPI) algorithm. Fig 11 shows the results of the PPI curve which indicates both the number and distribution of pure pixels in our study area. Fig 11 shows the PPI iteration process was 5000 which obtained 6000 number of pure pixels for endmember extraction in hyperspectral image analysis. \u0026nbsp;This comprehensive approach allows us to gain a deeper understanding of the spectral characteristics and composition of the studied area, contributing to more accurate and insightful analysis and interpretation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.7.3 N-Dimensional Visualizer Technique\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe N-Dimensional Visualizer is an interactive approach for enhancing the process of selecting spectrally pure pixels from a 2-dimensional PPI (Pure Pixel Index) image. In remote sensing and hyperspectral data analysis, identifying pure pixels are the spectral characteristics of individual elements or structures on Earth\u0026apos;s surface. In the present analysis, each end member has been assigned to a distinct glacier feature and then obtained using the 3-D scatter plot that is shown in Fig 12. This 3D scatter plot allows for a more tangible and interpretable visualization of the end members, which can help in understanding the spectral characteristics and variations of glacier features. The spectra of image end-members have been used as reference spectra for different snow surface characteristics. These reference spectra are employed in conjunction with the classified Spectral Angle Mapper (SAM) method. SAM is a spectral classification technique that compares the similarity of the spectra in the hyperspectral data to the reference spectra. This method helps classify different surface materials or features in the study area based on their spectral characteristics.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e4.7.4 Spectral Angle Mapper (SAM)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe classification of hyperspectral images by a spectral angle mapper is one of the finest techniques. It is a pixel-based automated classification algorithm. The SAM technique compares the spectral signature of each pixel in an image to reference spectra of various feature classes extracted from a spectral library or in the field with a spectrometer. The reference spectra of different features considered as endmembers can be obtained primarily through field measurements, the laboratory, or extracted directly from the hyperspectral imagery. The field spectral signature was used to reproduce the snow grain size map by SAM, as described in Negi et al. (2013). The selected spectra generated from the image were further used as reference spectra using the SAM method for grain size mapping (Rowan and Mars 2003) The SAM method determines the spectral similarity between image pixel spectra and reference spectra by computing the angle between two vectors containing these spectral signatures (Kruse et al. 1993) as shown in fig 13. The obtained image spectra from the SAM technique were used for identification and comparison with a spectral signature that was collected from the USGS spectral library built in ENVI software. The following formula calculates the spectral angle (\u0026alpha;).\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" width=\"494\" height=\"79\"\u003e\u003c/p\u003e"},{"header":"5. Results and discussion ","content":"\u003cp\u003e\u003cstrong\u003e5.1. Snow grain size mapping\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.1.1 Grain size mapping using the SAM method\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBased on their spectral signature, PRISMA data from the central Himalayan region was used to estimate different snow grain sizes. The spectral signature for fine, medium, and coarse snow grain has been obtained using Z-profiling from the selected endmember extract from the SAM-classified image, as shown in Fig 14. The generated spectral signature has been used for snow grain size mapping. These size classes represent the relative size and shape of individual snow grains, which can be indicative of snowpack properties such as density and temperature.\u003c/p\u003e\n\u003cp\u003eThe reflectance of various snow grain sizes was observed to decrease as contamination levels increased, as illustrated in Fig 14. Snow reflectance at 1030 nm has been found to drop by 15% when metamorphic characteristics of grain size change from fine to medium and a 10% decrease when transitioning from medium to coarse metamorphic grain characteristics. However, at visible wavelengths (ranging from 400 nm to 700 nm), reflectance remained relatively insensitive to changes in grain size. Reflectance in the near-infrared region, predominantly at 1030 nm, exhibited a high sensitivity to grain size, as previously reported by (Nolin and Dozier 2000, Dozier 1988). The presence of contamination on snow particles significantly influences the reflectance within the visible region, which can be examined using a hyperspectral remote sensing method. The spectral signatures of the glacier region were extracted from PRISMA data and are displayed in Fig 16. These figures illustrate how the high reflectance of snow changes in the visible region in response to varying types of contamination. It has been observed that fresh snow particles exhibit the highest reflectance in the visible spectrum, with reflectance decreasing almost to zero beyond the shortwave infrared (SWIR) region due to the pronounced ice absorption characteristics. It is also observed that in the near-infrared region, snow contamination has less effect on snow grain size. This classification process involves collecting spectra from the image and identifying each distinct class. Fig 15 shows the SAM-classified image for the snow-glacier region. However, in this classification approach, some areas were not classified because of low spectral signatures. This happened for high altitude areas, backside shadows of hilly regions. As shown in Table 6, the maximum covered area in the study region is observed in the class of less contamination snow, which spans 264.94 km\u0026sup2;, followed by Fresh snow with a coverage area of 190.33 km\u0026sup2;. In contrast, the minimum coverage areas are attributed to Barren Rocky and Vegetation, which encompass 7.83 km\u0026sup2; and 7.95 km\u0026sup2;, respectively. The High contamination snow class covers an area of approximately 72 km\u0026sup2;.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 6.\u003c/strong\u003e Retrieved of glacier features using SAM Classification\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"395\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47.3418%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClass\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 52.6582%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eArea (Km\u003csup\u003e2\u003c/sup\u003e)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47.3418%;\"\u003e\n \u003cp\u003eFresh snow\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 52.6582%;\"\u003e\n \u003cp\u003e190.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47.3418%;\"\u003e\n \u003cp\u003eWet snow\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 52.6582%;\"\u003e\n \u003cp\u003e51.84\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47.3418%;\"\u003e\n \u003cp\u003eLess contamination snow\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 52.6582%;\"\u003e\n \u003cp\u003e264.94\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47.3418%;\"\u003e\n \u003cp\u003eHigh contamination snow\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 52.6582%;\"\u003e\n \u003cp\u003e72.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47.3418%;\"\u003e\n \u003cp\u003eGlacier ice\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 52.6582%;\"\u003e\n \u003cp\u003e102.05\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47.3418%;\"\u003e\n \u003cp\u003eBarren Rocky\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 52.6582%;\"\u003e\n \u003cp\u003e7.83\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47.3418%;\"\u003e\n \u003cp\u003eVegetation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 52.6582%;\"\u003e\n \u003cp\u003e7.95\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47.3418%;\"\u003e\n \u003cp\u003eShadow region\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 52.6582%;\"\u003e\n \u003cp\u003e104.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eThe retrieval of snow grain size was performed using the GSI method and compared with the SAM algorithm for a specific region in the greater Himalayas range. Fig. 17(a and b) show the spatial distribution of snow grain sizes extracted from SAM and GSI respectively. In this study, it has been observed that a large portion snow cover area is covered with fine and medium grain size due to accumulation season in the month of mid-February when temperatures are low and rainfall is scarce. These conditions affect the snow properties in higher altitude regions. Table 7 shows the regional distribution of fine and medium snow particle sizes, indicating a remarkable uniformity between the results of the particle Size Index (GSI) and the SAM-classified imagery. The determination of the SAM angle threshold value helps avoid misclassification over steep slopes and dark or low-illuminated areas, which are marked as unclassified (black) areas. In a few places, a small deviation may be due to the selection of grain size threshold values used for the conversion of grain index to grain size classes. A comparison of the two datasets indicates that the SAM-classified image shows an increase of 13.69 km\u0026sup2; in fine grain size snow cover area, a decrease of 23.16 km\u0026sup2; in medium grain size snow, and also decrease of 71.48 km\u0026sup2; in coarse grain size snow.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 7.\u0026nbsp;\u003c/strong\u003eRetrieval of snow grain size using SAM, and Grain Index method\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"550\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 28%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSnow grain class\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGrain size (mm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eArea (Km\u003csup\u003e2\u003c/sup\u003e)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(GI method)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 26%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eArea (Km\u003csup\u003e2\u003c/sup\u003e)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(SAM method)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 28%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFine snow grain\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24%;\"\u003e\n \u003cp\u003e0.00 - 0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22%;\"\u003e\n \u003cp\u003e211.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 26%;\"\u003e\n \u003cp\u003e225.56\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 28%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMedium snow grain\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24%;\"\u003e\n \u003cp\u003e0.50 - 1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22%;\"\u003e\n \u003cp\u003e236.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 26%;\"\u003e\n \u003cp\u003e213.42\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 28%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCoarse snow grain\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24%;\"\u003e\n \u003cp\u003e1.00 - 2.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22%;\"\u003e\n \u003cp\u003e139.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 26%;\"\u003e\n \u003cp\u003e68.13\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eThe SAM and G.S.I qualitative grain size methods studied the classification of fine, medium, and coarse grain size classes in the snow glacier region. Such classifications are important in studying the avalanches, snow melting rate as well as snow depth. The two-grain size methods G.S.I and SAM have their advantage as well as limitations. The grain index method consists of two spectral bands that classify the snow grain based on the threshold values. This ratio technique reduces the topographic effects, which helps identify the snow grain under the mountainous shadow regions. The second classified Spectral angle mapper method uses selected maximum band information data to obtain qualitative snow grain sizes. The main advantage of incorporating all bands is the retrieval of snow grain and other features of the snow glacier region based on the identification of spectral signatures. The current methodology for mapping snow grain sizes using SAM is well-suited for the Himalayan region at varying altitudes, ranging from 4000 meters and above.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e5.1.2 Classification evaluation for PRISMA data\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe few random training samples were selected by using certain pixels of the PRISMA image that endmember spectra closely matched to the endmember spectra of the features found in the USGS spectral library and some reference spectra signatures have been taken from different literature. The confusion matrix analysis from the grain size classified map, generated using hyperspectral data, was utilized to assess classification accuracy in ENVI, as presented in Table 8. As a result, the overall accuracy and Kappa coefficient were computed. Classification accuracy assessment of Grain index map using SAM technique for glacier:\u003c/p\u003e\n\u003cp\u003eOverall accuracy = (1107166/ 1249122)/100 = 88.63%\u003c/p\u003e\n\u003cp\u003eKappa coefficient = 0.82 \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo enhance classification accuracy and ensure the reliability of the results, SAM classification outcomes were incorporated into a confusion matrix concerning user-defined ROI end members. The classification accuracy is presented in the Table 9. In this study area, the classification accuracy results reveal that the producer accuracy for coarse snow grain is 69.02%, while the user accuracy for coarse snow grain is 59.21%, suggesting a significant commission error in the coarse snow grain class. The user accuracy for fine snow grain is 85.07%, and for medium grain, it is 76.85%.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 8\u003c/strong\u003e. Theoretical error matrix of snow grain size classification\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"682\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 17.4487%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClass\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.5689%;\"\u003e\n \u003cp\u003eUnclassified\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.6891%;\"\u003e\n \u003cp\u003eFine snow grain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.9413%;\"\u003e\n \u003cp\u003eMedium snow grain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.1818%;\"\u003e\n \u003cp\u003eCoarse snow grain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.1701%;\"\u003e\n \u003cp\u003eTotal\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 17.4487%;\"\u003e\n \u003cp\u003eUnclassified\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.5689%;\"\u003e\n \u003cp\u003e614047\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.6891%;\"\u003e\n \u003cp\u003e2461\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.9413%;\"\u003e\n \u003cp\u003e5052\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.1818%;\"\u003e\n \u003cp\u003e84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.1701%;\"\u003e\n \u003cp\u003e621644\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 17.4487%;\"\u003e\n \u003cp\u003eFine snow grain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.5689%;\"\u003e\n \u003cp\u003e4022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.6891%;\"\u003e\n \u003cp\u003e254337\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.9413%;\"\u003e\n \u003cp\u003e40592\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.1818%;\"\u003e\n \u003cp\u003e21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.1701%;\"\u003e\n \u003cp\u003e298972\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 17.4487%;\"\u003e\n \u003cp\u003eMedium snow grain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.5689%;\"\u003e\n \u003cp\u003e34418\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.6891%;\"\u003e\n \u003cp\u003e3153\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.9413%;\"\u003e\n \u003cp\u003e192835\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.1818%;\"\u003e\n \u003cp\u003e20510\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.1701%;\"\u003e\n \u003cp\u003e250916\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 17.4487%;\"\u003e\n \u003cp\u003eCoarse snow grain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.5689%;\"\u003e\n \u003cp\u003e4192\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.6891%;\"\u003e\n \u003cp\u003e18387\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.9413%;\"\u003e\n \u003cp\u003e9064\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.1818%;\"\u003e\n \u003cp\u003e45947\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.1701%;\"\u003e\n \u003cp\u003e77590\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 17.4487%;\"\u003e\n \u003cp\u003eTotal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.5689%;\"\u003e\n \u003cp\u003e656679\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.6891%;\"\u003e\n \u003cp\u003e278338\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 19.9413%;\"\u003e\n \u003cp\u003e247543\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.1818%;\"\u003e\n \u003cp\u003e66562\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 12.1701%;\"\u003e\n \u003cp\u003e1249122\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eTable 9.\u0026nbsp;\u003c/strong\u003eAccuracy assessment of different classification\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"594\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.9696%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eClass\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 36.2563%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eProduction Accuracy (percent)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 37.774%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUser Accuracy (percent)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.9696%;\"\u003e\n \u003cp\u003eUnclassified\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 36.2563%;\"\u003e\n \u003cp\u003e93.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 37.774%;\"\u003e\n \u003cp\u003e98.77\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.9696%;\"\u003e\n \u003cp\u003eFine snow grain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 36.2563%;\"\u003e\n \u003cp\u003e91.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 37.774%;\"\u003e\n \u003cp\u003e85.07\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.9696%;\"\u003e\n \u003cp\u003eMedium snow grain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 36.2563%;\"\u003e\n \u003cp\u003e77.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 37.774%;\"\u003e\n \u003cp\u003e76.85\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.9696%;\"\u003e\n \u003cp\u003eCoarse snow grain\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 36.2563%;\"\u003e\n \u003cp\u003e69.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 37.774%;\"\u003e\n \u003cp\u003e59.21\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eThe snow grain size classified map created using the Spectral Angle Mapper (SAM) method demonstrated an overall matching area of approximately 88.63% when compared to the Grain Size Index (GSI) classified maps. This shows that SAM is a more effective technique for classifying snow grain size, providing valuable information for various applications, and contributing to the understanding of several important factors, such as snow contamination, fractional snow cover area, and albedo. However, it is essential to emphasize that a quantitative estimation of snow grain size features was not feasible in this study owing to the lack of ground truth data. These parameters are essential for accurately mapping snow cover, predicting the timing and magnitude of snowmelt, and better estimating future water resources and climate effects. Additionally, accurate snow cover mapping is crucial for applications related to glaciers, including distinguishing between clean and debris-covered glaciers, assessing glacial hazards, and conducting research in climatology, hydrology, and snow avalanche hazard analysis applications in the Himalayan region. In comparison to standard remote sensing data, hyperspectral data provides comprehensive spectral information that can considerably increase the accuracy of snow-related studies. By comparing the Spectral Angle Mapper method with bi-spectral satellite data, the study successfully captures changes in snow grain size dynamics, contributing to our understanding of snowpack characteristics in the upper Himalayan region. The application of this methodology is particularly beneficial in regions where snow conditions remain relatively consistent, as it allows for the accurate and timely assessment of snow properties. This, in turn, supports critical applications such as avalanche forecasting, water resource management, and climate research, all of which are essential for both scientific investigations and practical decision-making in the Himalayan region, where snow cover plays a vital role in various aspects of life and the environment.\u003c/p\u003e"},{"header":"6. Conclusions","content":"\u003cp\u003eIn this research, we examined the variations in spectral properties of different snow types within the upper Himalayan region using state-of-the-art hyperspectral PRISMA data. We employed two distinct classification methods: SAM and GI. These methods offer critical insights into snow attributes like grain size and snow cover, which are of utmost importance for hydrological modeling. These classification approaches serve as valuable tools for estimating snow grain size in scenarios where qualitative information about grain size is needed, such as in the context of melting snow, glacier forecasting, and climate studies, among others.\u003c/p\u003e\n\u003cp\u003eWhile both SAM and GI techniques were utilized to assess snow particle size, the SAM method demonstrated superior effectiveness in identifying unique snow characteristics within the glacier region. This advantage stems from the SAM classifier\u0026apos;s ability to utilize a select few spectral bands that yield more informative results with reduced noise. In contrast, the grain index method relies on just two spectral bands. Additionally, the ratio approach stands out due to its capacity to account for topographical influences. Spectral wavelengths around 441 nanometers and 1043 nanometers emerge as the most sensitive indicators of snow grain size. However, extracting endmembers from PRISMA data, characterized by a 30-meter resolution and a broad swath width, poses a challenging task when estimating snow grain size. Furthermore, the PRISMA hyperspectral data proves invaluable for studies in glacier regions. Detecting vertical variations in the snowpack through grain retrievals utilizing various ice absorption channels can significantly contribute to the understanding of snowpack stability in Himalayan snow avalanche research.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgment\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was carried out using PRISMA Products freely provided by the Italian Space Agency (ASI), and delivered under an ASI License. The Italian Space Agency is grateful for the PRISMA hyperspectral data.\u0026nbsp;The authors are also thankful to the Department of Water Resources Development and Management (WRD\u0026amp;M), IIT Roorkee, for providing all necessary facilities and constant encouragement for doing this research work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research received no funding\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors have read and approved the final manuscript and have no conflicts of interest related to this research work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eManish Rawat- conceptualized, formalized, and interpreted the results, and writing—original draft preparation.; Ashish Pandey- Supervision, results analysis, review, editing.; Dhananjay Paswan Das—review and Praveen Kumar Gupta- Supervision, \u0026nbsp;results analysis, review, editing. All authors have read and approved the final manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAhmed, R., Wani, G. F., Ahmad, S. T., Sahana, M., Singh, H., \u0026amp; Ahmed, P. (2021). A review of glacial lake expansion and associated glacial lake outburst floods in the Himalayan region. Earth Systems and Environment, 5(3), 695-708 . https://doi.org/10.1007/s41748-021-00230-9\u003c/li\u003e\n\u003cli\u003eAnderson GP, Pukall B, Allred CL, Jeong LS, Hoke MA, Chetwynd JH, Adler-Golden SM, Berk A, Bernstein LS, Richtsmeier SC, Acharya PK (1999) FLAASH and MODTRAN4: state-of-the-art atmospheric correction for hyperspectral data. In1999 IEEE Aerospace Conference. Proceedings (Cat. No. 99TH8403) Mar 7 (Vol. 4, pp. 177-181). IEEE. https://doi.org/10.1109/AERO.1999.792088\u003c/li\u003e\n\u003cli\u003eBloschl G, Kirnbauer R, Gutknecht D (1991) Distributed snowmelt simulations in an alpine catchment: 1. Model evaluation on the basis of snow cover patterns. Water resources research 27(12):3171-9. https://doi.org/10.1029/91WR02250\u003c/li\u003e\n\u003cli\u003eCooley T, Anderson GP, Felde GW, Hoke ML, Ratkowski AJ, Chetwynd JH, Gardner JA, Adler-Golden SM, Matthew MW, Berk A, Bernstein LS (2002) FLAASH, a MODTRAN4-based atmospheric correction algorithm, its application and validation. InIEEE international geoscience and remote sensing symposium. 1414-1418. IEEE. https://doi.org/10.1109/IGARSS.2002.1026134\u003c/li\u003e\n\u003cli\u003eColbeck SC (1991) The layered character of snow covers. Reviews of Geophysics 29(1):81-96. https://doi.org/10.1029/90RG02351\u003c/li\u003e\n\u003cli\u003eColbeck SC (1979) Water flow through heterogeneous snow. Cold Regions Science and Technology 1(1):37-45. https://doi.org/10.1016/0165-232X(79)90017-X\u003c/li\u003e\n\u003cli\u003eCasacchia R, Lauta F, Salvatori R, Cagnati A, Valt M, \u0026Oslash;rb\u0026aelig;k JB (2001) Radiometric investigation of different snow covers in Svalbard. Polar Research 20(1):13-22. https://doi.org/10.1111/j.1751-8369.2001.tb00035.x\u003c/li\u003e\n\u003cli\u003eDozier J, Painter TH (2004) Multispectral and hyperspectral remote sensing of alpine snow properties. Annu. Rev. Earth Planet 32:465-94.\u003c/li\u003e\n\u003cli\u003eDozier J, Schneider SR, McGinnis Jr DF (1981) Effect of grain size and snowpack water equivalence on visible and near‐infrared satellite observations of snow. Water Resources Research 17(4):1213-21. https://doi.org/10.1029/WR017i004p01213\u003c/li\u003e\n\u003cli\u003eDozier J, Davis RE, Chang AT, Brown K (1988) The spectral bidirectional reflectance of snow. In Spectral Signatures of Objects in Remote Sensing (Vol. 287, p. 87).\u003c/li\u003e\n\u003cli\u003eDozier J, Painter TH (2004) Multispectral and hyperspectral remote sensing of alpine snow properties. Annu. Rev. Earth Planet. Sci.32:465-94.\u003c/li\u003e\n\u003cli\u003eDatt B, McVicar TR, Van Niel TG, Jupp DL, Pearlman JS (2003) Preprocessing EO-1 Hyperion hyperspectral data to support the application of agricultural indexes. IEEE Transactions on Geoscience and Remote Sensing 41(6):1246-59. https://doi.org/10.1109/TGRS.2003.813206\u003c/li\u003e\n\u003cli\u003eGreen RO, Dozier J, Roberts D, Painter T (2002) Spectral snow-reflectance models for grain-size and liquid-water fraction in melting snow for the solar-reflected spectrum. Annals of Glaciology 34:71-3. https://doi.org/10.3189/172756402781817987\u003c/li\u003e\n\u003cli\u003eJenson, J.R., (2005) Introductory Digital Image Processing: a Remote Sensing Perspective.,(Pearson Prentice Hall: Upper Saddle River, NJ.\u003c/li\u003e\n\u003cli\u003eKokhanovsky A (2013) Spectral reflectance of solar light from dirty snow: a simple theoretical model and its validation. The Cryosphere 7(4):1325-31. https://doi.org/10.5194/tc-7-1325-2013\u003c/li\u003e\n\u003cli\u003eKaufman YJ, Wald AE, Remer LA, Gao BC, Li RR, Flynn L (1997) The MODIS 2.1-/spl mu/m channel-correlation with visible reflectance for use in remote sensing of aerosol. IEEE transactions on Geoscience and Remote Sensing 35(5):1286-98. https://doi.org/10.1109/36.628795\u003c/li\u003e\n\u003cli\u003eKulkarni AV, Srinivasulu J, Manjul SS, P M (2002) Field based spectral reflectance studies to develop NDSI method for snow cover monitoring. Journal of the Indian Society of Remote Sensing 30:73-80. https://doi.org/10.1007/BF02989978\u003c/li\u003e\n\u003cli\u003eKruse FA, Lefkoff AB, Boardman JW, Heidebrecht KB, Shapiro AT, Barloon PJ, Goetz AF (1993) The spectral image processing system (SIPS)\u0026mdash;interactive visualization and analysis of imaging spectrometer data. Remote sensing of environment 44(2-3):145-63. https://doi.org/10.1016/0034-4257(93)90013-N\u003c/li\u003e\n\u003cli\u003eKumar R, Bahuguna IM, Ali SN, Singh R (2020) Lake inventory and evolution of glacial lakes in the Nubra-Shyok basin of Karakoram Range. Earth Systems and Environment 4:57-70. https://doi.org/10.1007/s41748-019-00129-6\u003c/li\u003e\n\u003cli\u003eLi W, Stamnes K, Chen B, Xiong X (2001) Snow grain size retrieved from near‐infrared radiances at multiple wavelengths. Geophysical Research Letters 28(9):1699-702. https://doi.org/10.1029/2000GL011641 \u003c/li\u003e\n\u003cli\u003eMarks D, Dozier J (1992) Climate and energy exchange at the snow surface in the alpine region of the Sierra Nevada: 2.Snow cover energy balance. Water Resources Research 28(11):3043-54. https://doi.org/10.1029/92WR01483 \u003c/li\u003e\n\u003cli\u003eMatthew MW, Adler-Golden SM, Berk A, Richtsmeier SC, Levine RY, Bernstein LS, Acharya PK, Anderson GP, Felde GW, Hoke ML, Ratkowski AJ (2000) Status of atmospheric correction using a MODTRAN4-based algorithm. In Algorithms for multispectral, hyperspectral, and ultraspectral imagery VI (Vol. 4049, pp. 199-207). SPIE. https://doi.org/10.1117/12.410341\u003c/li\u003e\n\u003cli\u003eMatthew MW, Adler-Golden SM, Berk A, Richtsmeier SC, Levine RY, Bernstein LS, Acharya PK, Anderson GP, Felde GW, Hoke ML, Ratkowski AJ (2000) Status of atmospheric correction using a MODTRAN4-based algorithm. In Algorithms for multispectral, hyperspectral, and ultraspectral imagery VI (Vol. 4049, pp. 199-207). SPIE. https://doi.org/10.1117/12.410341\u003c/li\u003e\n\u003cli\u003eMoughal, T.A (2013) Hyperspectral image classification using support vector machine. In journal of physics: conference series (Vol. 439, No. 1, p. 012042). IOP Publishing.\u003c/li\u003e\n\u003cli\u003eMudbhari, D., Kansal, M. L., \u0026amp; Kalura, P. (2022). Impact of climate change on water availability in Marsyangdi river basin, Nepal. Quarterly Journal of the Royal Meteorological Society, 148(744), 1407-1423.\u003c/li\u003e\n\u003cli\u003eNegi HS, Jassar HS, Saravana G, Thakur NK, Snehmani, Ganju A (2013) Snow-cover characteristics using Hyperion data for the Himalayan region. International journal of remote sensing 34(6):2140-61. https://doi.org/10.1080/01431161.2012.742213\u003c/li\u003e\n\u003cli\u003eNolin AW, Dozier J (2000) A hyperspectral method for remotely sensing the grain size of snow. Remote sensing of Environment 74(2):207-16. https://doi.org/10.1016/S0034-4257(00)00111-5\u003c/li\u003e\n\u003cli\u003eNegi HS, Kokhanovsky A (2011) Retrieval of snow grain size and albedo of western Himalayan snow cover using satellite data. The Cryosphere 5(4):831-47. https://doi.org/10.5194/tc-5-831-2011\u003c/li\u003e\n\u003cli\u003eNegi HS, Kulkarni AV, Semwal BS (2009) Study of contaminated and mixed objects snow reflectance in Indian Himalaya using spectroradiometer. International Journal of Remote Sensing 30(2):315-25. https://doi.org/10.1080/01431160802261197\u003c/li\u003e\n\u003cli\u003eNegi HS, Singh SK, Kulkarni AV, Semwal BS (2010) Field-based spectral reflectance measurements of seasonal snow cover in the Indian Himalaya. International Journal of Remote Sensing 31(9):2393-417. https://doi.org/10.1080/01431160903002417\u003c/li\u003e\n\u003cli\u003ePetropoulos GP, Kalaitzidis C, Vadrevu KP (2012) Support vector machines and object-based classification for obtaining land-use/cover cartography from Hyperion hyperspectral imagery. Computers \u0026amp; Geosciences 41:99-107. https://doi.org/10.1016/j.cageo.2011.08.019\u003c/li\u003e\n\u003cli\u003ePainter TH, Dozier J, Roberts DA, Davis RE, Green RO (2003) Retrieval of sub pixel snow-covered area and grain size from imaging spectrometer data. Remote Sensing of Environment 85(1):64-77. https://doi.org/10.1016/S0034-4257(02)00187-6\u003c/li\u003e\n\u003cli\u003eQiu F, Abdelsalam M, Thakkar P (2006) Spectral analysis of ASTER data covering part of the Neoproterozoic Allaqi-Heiani suture, Southern Egypt. Journal of African Earth Sciences 44(2):169-80. https://doi.org/10.1016/j.jafrearsci.2005.10.009\u003c/li\u003e\n\u003cli\u003eRogge DM, Rivard B, Zhang J, Sanchez A, Harris J, Feng J (2007) Integration of spatial\u0026ndash;spectral information for the improved extraction of endmembers. Remote Sensing of Environment 110(3):287-303. https://doi.org/10.1016/j.rse.2007.02.019\u003c/li\u003e\n\u003cli\u003eRowan LC, Mars JC (2003) Lithologic mapping in the Mountain Pass, California area using advanced space borne thermal emission and reflection radiometer (ASTER) data. Remote sensing of Environment 84(3):350-66. https://doi.org/10.1016/S0034-4257(02)00127-X\u003c/li\u003e\n\u003cli\u003eSarkar T, Kannaujiya S, Taloor AK, Ray PK, Chauhan P (2020) Integrated study of GRACE data derived interannual groundwater storage variability over water stressed Indian regions. Groundwater for sustainable development 10:100376. https://doi.org/10.1016/j.gsd.2020.100376\u003c/li\u003e\n\u003cli\u003eScambos TA, Haran TM, Fahnestock MA, Painter TH, Bohlander J (2007) MODIS-based Mosaic of Antarctica (MOA) data sets: Continent-wide surface morphology and snow grain size. Remote sensing of environment 111(2-3):242-57. https://doi.org/10.1016/j.rse.2006.12.020\u003c/li\u003e\n\u003cli\u003eSingh SK, Kulkarni AV, Chaudhary BS (2010) Hyperspectral analysis of snow reflectance to understand the effects of contamination and grain size. Annals of Glaciology 51(54):83-8. https://doi.org/10.3189/172756410791386535\u003c/li\u003e\n\u003cli\u003eWarren SG (1982) Optical properties of snow. Reviews of Geophysics 20(1):67-89. https://doi.org/10.1029/RG020i001p00067\u003c/li\u003e\n\u003cli\u003eWiscombe WJ, Warren SG (1980) A model for the spectral albedo of snow. I: Pure snow. Journal of Atmospheric Sciences 37(12):2712-33. https://doi.org/10.1175/1520-0469(1980)037\u0026lt;2712:AMFTSA\u0026gt;2.0.CO;2\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"applied-geomatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"agmj","sideBox":"Learn more about [Applied Geomatics](http://link.springer.com/journal/12518)","snPcode":"12518","submissionUrl":"https://submission.nature.com/new-submission/12518/3","title":"Applied Geomatics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"hyperspectral imagery, Grain Size Index, Snow cover classification, endmembers, and Eigenvalues","lastPublishedDoi":"10.21203/rs.3.rs-5031527/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5031527/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRapid urbanization have significantly increased freshwater consumption, leading researchers to focus on accurately predicting snowmelt-derived streamflow using hydrological models. The glacierized basins of the Himalayan region are significantly vulnerable to climate change. The understanding of physical characterization of snow such as snow cover and snow grain size, remains challenging due to inaccessible terrains which creates hindrance for in-situ data collection. The hyperspectral remote sensing datasets are most promising for monitoring and retrieving the snow properties at the micro and macro levels. In this study, the PRISMA hyperspectral dataset was used for the retrieval of different snow grain sizes in the Bhilangana basin of the upper Himalayan region using the Spectral Angle Mapper (SAM) classification method and the Snow Grain Index (SGI) method. The spectral reflectance of different types of glacier features was generated using the SAM classification technique on PRISMA hyperspectral imagery and validated from the USGS spectral library for mapping. The results demonstrated that there is good qualitative agreement observed between the class-wise grain size classes using the grain index and SAM method. Additionally, the overall accuracy of the SAM and the grain size index classification methods for retrieving the grain size of different classes was approximately 88% .The outcomes of the study discloses the competency of PRISMA data for snow grain size mapping over the Mountainous region. The estimated parameters could be applied to climatology, hydrology, and mountain hazard mapping over the Himalayan region. Detailed snow grain size mapping can enhance mountain hazard assessments, including the prediction and management of snow avalanches, thereby contributing to the safety and sustainability of mountainous areas in the Himalayan region.\u003c/p\u003e","manuscriptTitle":"Characterization and Retrieval of Snow Grain Size in the Upper Himalayan Region Using Hyperspectral Prisma Data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-10-09 06:18:27","doi":"10.21203/rs.3.rs-5031527/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-10-12T13:55:32+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-10-07T10:51:36+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-10-04T06:35:49+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"210800032367594511529501598445452787459","date":"2024-09-27T16:07:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"202620127319227927419061784182565221594","date":"2024-09-27T15:43:54+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"323330010183948042925332911001812260168","date":"2024-09-10T15:54:16+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-09-10T10:40:36+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-09-06T09:31:43+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-09-06T09:30:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"Applied Geomatics","date":"2024-09-04T12:11:12+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"applied-geomatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"agmj","sideBox":"Learn more about [Applied Geomatics](http://link.springer.com/journal/12518)","snPcode":"12518","submissionUrl":"https://submission.nature.com/new-submission/12518/3","title":"Applied Geomatics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"5cff8344-4806-47b1-b196-23f303e0300c","owner":[],"postedDate":"October 9th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-04-21T16:03:02+00:00","versionOfRecord":{"articleIdentity":"rs-5031527","link":"https://doi.org/10.1007/s12518-025-00627-5","journal":{"identity":"applied-geomatics","isVorOnly":false,"title":"Applied Geomatics"},"publishedOn":"2025-04-14 15:57:19","publishedOnDateReadable":"April 14th, 2025"},"versionCreatedAt":"2024-10-09 06:18:27","video":"","vorDoi":"10.1007/s12518-025-00627-5","vorDoiUrl":"https://doi.org/10.1007/s12518-025-00627-5","workflowStages":[]},"version":"v1","identity":"rs-5031527","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5031527","identity":"rs-5031527","version":["v1"]},"buildId":"7rjqhiLT3MXkJMwkYKINL","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}