Evaluation method for monitoring elevation change performance of mountain glacier through GEDI

Abstract

The invention relates to the technical field of glacier change monitoring, and discloses an evaluation method for monitoring elevation change performance of mountain glacier through GEDI, which comprises the following steps of: preprocessing laser footprint data in GEDI L2A products and ICESat-2 ATL06 products, specifically, extracting ground elevation and latitude and longitude of lowest mode echoes from the GEDI L2A, and calculating the elevation change of the ground elevation and the latitude and longitude of the lowest mode echoes in the GEDI L2A products and ICESat-2 ATL06 products according to the extracted ground elevation and latitude and longitude of the lowest mode echoes; a quality control label and a degradation state label are adopted for screening, the latitude and longitude, the elevation value, the elevation error and the quality control index of the photon point cloud are extracted from ICESat-2 ATL06, and screening is carried out according to conditions; the TanDEM-X DEM data is used as a reference, and the elevation difference between each footprint point and the DEM is calculated; after the abnormal values are removed, the elevation change of the glacier area and the non-glacier area is counted; and carrying out interannual and seasonal analysis on the overall glacier region and the main aggregated glacier sub-regions based on elevation change, and carrying out GEDI monitoring performance evaluation in combination with topographic factors and glacier attributes. According to the method, the elevation change performance of the mountain glacier monitored by the GEDI can be effectively evaluated when field survey data is lacked.

Description

Technical Field

[0001] The present invention relates to the technical field of glacier change monitoring, and in particular to a method for evaluating the performance of GEDI in monitoring elevation changes of mountain glaciers. Background Art

[0002] Glaciers are a crucial component of Earth's hydrological and climate systems, widely distributed across high mountains and polar regions. They are not only a vital reservoir of global water resources but also crucial for climate change research. Glacier changes are closely linked to climate fluctuations, particularly their melting and accretion processes, which directly reflect long-term trends in global climate change. As global temperatures rise, many mountain glaciers are experiencing accelerated melting, which not only affects the distribution of water resources but also has potential impacts on downstream ecosystems and human societies. However, the process of glacier elevation change is complex, influenced by multiple factors, including climate change, precipitation, glacier surface cover, and topographic characteristics. Existing glacier monitoring technologies often use remote sensing to assess elevation changes. However, due to technical and data limitations, it is often difficult to fully and accurately reflect the true changes in glaciers, especially in complex terrain and variable climate conditions.

[0003] Existing technologies typically rely on a single remote sensing data source for monitoring glacier elevation changes, such as traditional satellite elevation data or lidar data. While these data can provide information on elevation changes in glacier areas to a certain extent, they are subject to significant errors and instability due to inherent data limitations. A single data source cannot fully reflect the multidimensional changes in glacier elevation, making it difficult to conduct effective glacier change assessments. Furthermore, most existing monitoring methods ignore the differences between glaciated and non-glaciated areas and lack comprehensive analysis, resulting in results that may be affected by local data errors.

[0004] On the other hand, traditional technical solutions also lack consideration for seasonal variations and topographic factors. Existing methods typically rely on fixed elevation measurement techniques, failing to effectively distinguish seasonal variations or variations in slope or altitude. This results in a lack of in-depth analysis of glacier changes in complex terrain or climatic conditions. Seasonal variations are particularly pronounced in mountainous areas, and these factors are often underrepresented in existing technologies, making it difficult to accurately capture the subtle nuances of glacier melt. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides an evaluation method for the GEDI performance of monitoring mountain glacier elevation changes, which solves the problem of accurately monitoring mountain glacier elevation changes under complex terrain and climatic conditions.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for evaluating the performance of GEDI in monitoring the elevation change of mountain glaciers, comprising the following steps:

[0007] Preprocessing of laser footprint data in GEDI L2A products and ICESat-2 ATL06 products;

[0008] Using TanDEM-X DEM data as a reference, the elevation difference between each footprint point in the laser footprint data and the DEM was calculated;

[0009] After removing outliers, the elevation changes between glaciated and non-glaciated areas were calculated;

[0010] Based on the results of elevation changes, interannual and seasonal analyses were conducted in the overall glacier area and the main glacier aggregation sub-areas, and the GEDI monitoring performance was evaluated in combination with topographic factors and glacier properties.

[0011] The pre-processing step comprises:

[0012] Extract the ground elevation, longitude, and latitude of the lowest mode echo from GEDI L2A. Filter the data using quality control tags and degraded status tags indicating problems with satellite pointing or positioning information. Filter high-quality data with a quality control tag value of 1 and a degraded status tag value of 0.

[0013] The latitude, longitude, elevation, elevation error, and quality control index of the photon point cloud were extracted from ICESat-2 ATL06, and the data with a quality control index of 0 and an elevation error of ≤25m were retained.

[0014] Preferably, the calculating elevation difference includes:

[0015] The grid values of TanDEM-X DEM were matched to each footprint point using bilinear interpolation;

[0016] Calculate the difference between the elevation value of each footprint point and the DEM value as the elevation difference;

[0017] The formula for the elevation change at a single footprint point is as follows:

[0018] Δh GEDI =H GEDI -H TanDEM-X (1)

[0019] Δh ICESat-2 =H ICESat-2 -H TanDEM-X (2)

[0020] Where: Δh GEDI and Δh ICESat-2Indicates the elevation change of each footprint; H GEDI and H ICESat-2 is the height of the GEDI L2A and ICESat-2 ATL06 footprints; H TanDEM-X Represents the corresponding TanDEM-X interpolated elevation at each GEDI or ICESat-2 footprint point;

[0021] A threshold of ±50m was set. If the elevation change exceeded this threshold, it would be considered an outlier and removed.

[0022] Then, the cumulative elevation change of the entire glacier area from the time of TanDEM-X DEM reference data collection to the time of GEDI and ICESat-2 data collection was estimated by calculating the mean of all valid laser footprints in the glacier area. The calculation expression of the overall glacier elevation change in the glacier area is:

[0023] Where: Δh represents the cumulative elevation change over many years; ∑Δh GEDI / Δh ICESat-2 is the sum of the elevation changes of all valid footprint points of GEDI L2A or ICESat-2; n is the number of valid footprint points of GEDI or ICESat-2;

[0024] The annual average change rate calculated based on the above-mentioned cumulative elevation changes over many years can reflect the speed of glacier elevation change. The expression for calculating the glacier elevation change rate is shown in equation (4):

[0025]

[0026] Where v represents the annual average elevation change rate; T GEDI / ICESat-2 is the data collection time of GEDI L2A or ICESat-2; T TanDEM-X The data product collection time of TanDEM-X DEM is set as 2015.

[0027] The statistical analysis includes:

[0028] Statistics were conducted in glaciated and non-glaciated areas respectively, and the mean, standard deviation, median, median absolute deviation (MAD) of the cumulative elevation change and the average annual elevation change rate were calculated.

[0029] The formula for median absolute deviation (MAD) is as follows: MAD = median(|Δh i -m(Δh)|);

[0030] Where: Δh i represents the elevation difference observation value of the i-th footprint point; m(Δh) represents all Δh iThe median of the medians; median(*) means taking the median of the absolute deviations of these medians;

[0031] The results from non-glaciated areas are used to verify the accuracy of GEDI products in representing surface elevation and the uncertainty of glacier elevation change estimates.

[0032] The interannual and seasonal analyses include:

[0033] Data from different years were collected separately to conduct interannual variation analysis; elevation changes in spring, summer, autumn and winter were calculated separately; and seasonal variation trends were fitted using the least squares method.

[0034] The terrain factors include altitude, slope and aspect. The specific analysis includes: segmenting the data by altitude and analyzing the elevation changes in different altitude segments; dividing the glaciers into multiple levels by slope and analyzing the relationship between elevation changes and slope; and classifying by aspect and analyzing the differences in elevation changes under different directions.

[0035] The glacier attributes include glacier area and surface moraine coverage. The specific analysis includes: dividing glaciers into multiple levels according to area and counting the elevation changes of each level; dividing glaciers into two categories, with and without surface moraine, and analyzing the differences in their elevation changes; calculating the inhibition rate of the impact of moraine cover on glacier melting, the formula is: inhibition rate = (elevation change of glaciers without moraine - elevation change of glaciers with moraine) ÷ elevation change of glaciers without moraine × 100%.

[0036] Preferably, during the slope analysis process, the slope is divided into five levels, including: 5-15°, 15-25°, 25-35°, 35-45°, and greater than or equal to 45°.

[0037] Preferably, the glacier area is divided based on a multi-level classification standard, which subdivides the glacier area into the following eight ranges: less than 0.1 km 2 , 0.1–0.2km 2 , 0.2–0.5km 2 , 0.5–1km 2 1–2 km 2 2–5 km 2 5–10 km 2 , and greater than or equal to 10km 2 .

[0038] Preferably, the ICESat-2 ATL06 product is used as the reference data. By comparing the elevation differences between GEDI and ICESat-2 in the same area and combining trend analysis methods, the characteristics of GEDI in revealing glacier elevation changes are evaluated.

[0039] The present invention provides a method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes. It has the following beneficial effects:

[0040] 1. This paper uses the more accurate ICESat-2 land ice ATL06 data product to accurately evaluate the GEDI's performance in monitoring elevation changes of mountain glaciers. This solves the difficulty of directly verifying the absolute elevation accuracy of GEDI's monitoring capabilities when there is no field monitoring data or insufficient monitoring data, and provides an effective method for evaluating GEDI's performance in monitoring elevation changes.

[0041] 2. This invention achieves a more detailed and comprehensive elevation change monitoring effect by combining seasonal analysis with topographic factors. Compared with previous solutions that do not consider topographic factors or seasonal changes, this invention can accurately characterize the changing characteristics of different seasons and different slopes and altitudes, overcoming the shortcomings of single seasonal analysis or ignoring topographic factors.

[0042] 3. This invention optimizes the method for assessing glacier elevation changes by analyzing different glacier attributes, such as moraine cover and glacier area. Compared to existing techniques that ignore glacier attributes, this invention accurately assesses elevation changes across different glacier types, providing more targeted data support. This approach offers unique advantages, particularly in assessing the impact of glacier melt.

[0043] 4. This invention utilizes a bilinear interpolation-based elevation difference calculation technique, enhancing accuracy and eliminating outliers for more robust results. Compared to the cruder elevation difference calculation methods used in existing technologies, this invention effectively reduces errors and improves data reliability, particularly in applications involving complex terrain and dynamic environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flow chart of the method of the present invention;

[0045] Figure 2 This is a comparison chart of the height difference between glacier areas monitored by GEDI and ICESat-2, taking 2019 as an example;

[0046] Figure 3 This is a comparison chart of the height difference between GEDI and ICESat-2 monitoring non-glacial areas in the present invention, taking 2019 as an example;

[0047] Figure 4 A seasonal comparison chart of glacier elevation changes monitored by GEDI and ICESat-2 of the present invention;

[0048] Figure 5 The glacier elevation change map of the four glacier sub-regions of the present invention;

[0049] Figure 6 The number and area distribution diagram of glaciers with different average slopes of the present invention;

[0050] Figure 7 A diagram showing changes in glacier elevation at different terrain altitudes according to the present invention;

[0051] Figure 8 This is a comparative analysis diagram of the difference in elevation monitored by GEDI and ICESat-2 at different altitudes of the present invention;

[0052] Figure 9 The distribution map of glacier area and quantity on different slopes of the present invention;

[0053] Figure 10 A comparative diagram of glacier elevation changes on different terrain slopes of the present invention;

[0054] Figure 11 This is a comparative analysis diagram of the elevation differences monitored by GEDI and ICESat-2 on different slopes of the present invention;

[0055] Figure 12 It is the elevation change standard deviation diagram under different slopes of the present invention;

[0056] Figure 13 The figure shows the number and area ratio of glaciers in different slope directions according to the present invention;

[0057] Figure 14 This is a graph of glacier elevation changes at different slopes according to the present invention;

[0058] Figure 15 The number and area distribution map of glaciers of different area levels of the present invention;

[0059] Figure 16 A diagram showing changes in glacier elevation within different area levels of the present invention;

[0060] Figure 17 This is a comparative analysis of the differences in glacier elevation changes between GEDI and ICESat-2 within different area levels of the present invention;

[0061] Figure 18 The elevation change diagram of glaciers with and without surface moraines of the present invention;

[0062] Figure 19 This is a flow chart of the GEDI monitoring of mountain glacier elevation changes and its performance evaluation. DETAILED DESCRIPTION

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0064] Please see the attached Figure 1 -Attached Figure 19 An embodiment of the present invention provides a method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes, comprising the following steps: preprocessing laser footprint data in the GEDI L2A product and the ICESat-2 ATL06 product; calculating the elevation difference between each footprint point and the DEM using TanDEM-XDEM data as a reference; statistically analyzing the elevation changes between glacier areas and non-glacier areas after removing outliers; conducting interannual and seasonal analysis of the overall glacier area and the main glacier aggregation sub-areas based on the elevation change results, and evaluating the GEDI monitoring performance in combination with terrain factors and glacier properties.

[0065] The method employed consists of three main components: preprocessing of GEDI and ICESat-2 footprint data, extraction of elevation data, comparative statistical analysis of changes, and analysis of factors influencing GEDI-monitored glacier elevation changes. This analysis includes an assessment of the impact of GEDI data parameter selection, terrain environment (slope, aspect, and elevation distribution), and glacier properties (area and moraine cover) on GEDI detection accuracy.

[0066] Data Preparation: According to the GEDI positioning principle, the position of the ground echo in the waveform is determined by the last detected peak position. Therefore, the parameter "elev_lowestmode" is extracted to represent the glacier surface elevation. For this study, only the elevation value "elev_lowestmode" in the GEDI default algorithm was selected for testing. The "lon_lowestmode" and "lat_lowestmode" parameters were also used to determine the latitude and longitude of each surface footprint. To eliminate erroneous or low-quality data, the quality assessment parameters "quality_flag" and "degrade_flag" were used to filter valid data. "quality_flag" takes a value of 0 or 1, with 1 indicating good quality and laser footprints with a value of 0 being discarded. "degrade_flag" also takes a value of 0 or 1, with 1 indicating degraded pointing or positioning information, which could lead to data issues. Therefore, footprints with a value of 0 are retained. Furthermore, different GEDI beam intensity modes (full power or coverage) were selected for terrain estimation.

[0067] In ATL06 preprocessing, extracted fields include geographic location (latitude, longitude), surface elevation (h_li), elevation uncertainty (sigma_geo_h), and a quality metric for each segment (atl06_quality_summary). Similar to GEDI processing, no distinction is made between strong and weak beam modes when estimating terrain. The "atl06_quality_summary" parameter takes a value of 0 or 1, with 0 indicating no issues found in the data quality test and 1 indicating potential issues. Therefore, only segments with a value of 0 are selected to obtain high-quality data. The parameter "sigma_geo_h" is also used. This parameter is a comprehensive indicator of the total vertical geolocation error caused by satellite pointing or positioning, including the contribution of horizontal geolocation error to vertical error. For the glacier catalog, values ≤ 25 m are selected to provide high-precision reference information for evaluating elevation changes acquired by GEDI.

[0068] Extracting Glacier Elevation Changes: To monitor glacier elevation changes, this application compares and analyzes GEDI and ICESat-2 elevation data with the TanDEM-X DEM reference data, which contains historical glacier elevations. Furthermore, statistical characteristics of each glacier (including area, maximum / minimum elevation, average height, average slope, and moraine cover) were extracted from the glacier catalog file for detailed evaluation of the GEDI data. The specific processing flow is as follows:

[0069] Based on glacier boundary vector data, we initially extracted laser footprints from GEDI L2A and ICESat-2 ATL06 within the SETP region of southeastern Tibet. We obtained the corresponding glacier elevation, geographic location, and related accuracy metrics for each footprint. Using bilinear interpolation, we interpolated the TanDEM-X 90m DEM grid values to the locations of each GEDI and ICESat-2 footprint point, extracting the DEM grid center elevation. Because all three data sources are based on the WGS84 ellipsoidal high datum, no elevation datum conversion was performed; instead, the elevation difference between the interpolated DEM and the corresponding GEDI L2A and ICESat-2 ATL06 footprint points was directly calculated.

[0070] A ±50m threshold was set to exclude elevation differences outside this range. Using valid footprints as input data, the final elevation change for each observation mission relative to the TanDEM-X 90mDEM was determined by calculating the mean and median. A statistical analysis of elevation change differences within a 50m range was also performed.

[0071] To evaluate the elevation change values estimated by GEDI, ICESat-2 data were used as a reference standard. Statistical analysis of interannual and seasonal variations was conducted across the entire southeastern Tibet region and four sub-regions with concentrated glacier distribution. The statistical results in non-glacier areas were used as a benchmark for accuracy verification.

[0072] Statistical analysis: To evaluate the surface elevation accuracy of GEDI observations, this application compared and analyzed the elevation changes corresponding to ICESat-2 both inside and outside the glacier boundary. Among them, the elevation differences in the glacier area mainly reflect the actual changes in the glacier elevation, while the differences in the non-glacier area represent the observation quality of each satellite mission itself. Although the surface elevation may change locally due to various factors, in the non-glacier area, it is assumed that the average elevation will not change significantly, and the elevation changes at most points are small. Therefore, the elevation difference between the spatial altimetry data and the reference DEM can effectively reflect the measurement accuracy of GEDI and ICESat-2 products.

[0073] To statistically evaluate the estimated results of glacier elevation changes, this application calculated the following indicators between the GEDI and ICESat-2 footprint points and the TanDEM-X DEM reference data: First, the mean and standard deviation (SD) of the elevation differences of all footprint points were calculated. Considering the presence of outliers in the GEDI dataset, these outliers will significantly affect the calculation results of the mean and standard deviation. To improve the robustness of the statistical analysis, this application also uses the median and median absolute deviation (MAD) as supplementary evaluation indicators. The formula of median absolute deviation MAD is as follows: MAD = median(|Δh i -m(Δh)|);

[0074] Where: Δh i represents the elevation difference observation value of the i-th footprint point; m(Δh) represents the total Δh i median(*) means taking the median of the absolute deviations from these medians.

[0075] The results from non-glaciated areas are used to verify the accuracy of GEDI products in representing surface elevation and the uncertainty of glacier elevation changes.

[0076] Example 1: GEDI monitors the overall glacier elevation change.

[0077] According to the above-mentioned glacier elevation change extraction method, the values in the GEDI data with elevation differences greater than 50m or less than -50m are regarded as outliers and eliminated. The data retention rates from 2019 to 2021 were 22.93%, 32.10% and 32.49% respectively, and the average effective monitoring rate was 29.17%. Figure 2 and Figure 3The results of the 2019 GEDI and ICESat-2 elevation change comparisons in glaciated and non-glaciated areas are shown. Figure 2 and Figure 3 It can be seen that the number of ICESat-2 observation points is greater than that of GEDI in both glacial and non-glacial areas. This denser observation distribution indicates that it is feasible to use ICESat-2 data to evaluate GEDI data. Figure 2 The results show the changes in glacier elevation within the range of ±50m. The results of GEDI and ICESat-2 show similar distribution characteristics and are both shifted to the left, indicating that the number of negative elevation changes (glacier melting) is greater than the positive elevation changes (glacier accumulation). This result confirms that the glaciers in the study area are in a state of melting. Figure 3 The elevation changes in non-glaciated areas are shown. While the elevation in non-glaciated areas should theoretically remain stable, seasonal factors such as snow cover may still cause fluctuations. Since the TanDEM-XDEM data were collected between 2010 and 2015, and the exact collection period is unknown, these seasonal effects have not been corrected, which may lead to an underestimation of elevation changes. Furthermore, differences in the quality of the multi-source data and errors introduced during processing can also affect the elevation change results. Compared with ICESat-2, the GEDI data exhibit a more pronounced asymmetric distribution and negative skewness. In addition to a greater degree of data dispersion, the GEDI data also exhibit more significant variations, as evidenced by the long tail of the histogram, indicating that significant above-average glacier melt may be occurring in some local areas.

[0078] Table 1 lists the elevation changes in glacier and non-glacier areas from 2019 to 2021. In terms of absolute values, the medians of GEDI and ICESat-2 are both greater than the average value of the glacier area, which is consistent with the left-biased characteristics of elevation changes shown in the aforementioned histogram. From 2019 to 2021, the average elevation difference (GEDI minus ICESat-2) was -0.53m, 0.30m, and -0.64m, respectively, and the corresponding three-year average was -0.29m; the median elevation difference was -0.86m, 0.27m, and -0.83m, respectively, and the three-year average was -0.47m. In terms of accuracy, GEDI has larger SD and MAD, with an average difference of 0.54m and 0.43m compared with ICESat-2, indicating that the overall quality of GEDI is relatively poor and the uncertainty in monitoring glacier elevation changes is large.

[0079] For non-glaciated areas, the average values of GEDI and ICESat-2 in 2019 were similar, but the results in 2020 (-0.94m vs. -0.56m) and 2021 (-1.07m vs. -0.69m) showed large differences, indicating that GEDI has a relatively large error and overestimates the terrain elevation changes to a certain extent. From the median point of view, the values of the three years are relatively close, indicating that using the GEDI median to represent elevation changes is more robust. At the same time, the SD and MAD indicators also show that the GEDI data is relatively scattered compared to ICESat-2. The reason for this may be that each GEDI footprint represents a larger area. The diameter of the GEDI return waveform is about 25m. In such a large alpine area, the terrain undulation may vary greatly. Therefore, taking the average value of all TanDEM-XDEM pixels within the footprint may be more appropriate than taking a single height.

[0080] Table 1 Comparison of glacier elevation changes inside and outside the glacier area monitored by GEDI and ICESat-2

[0081]

[0082]

[0083] This table shows the statistical results of elevation changes monitored by GEDI and ICESat-2 satellites in glacier areas and off-glacier areas from 2019 to 2021. Mean represents the average value, reflecting the overall trend of change (negative values indicate a decrease in elevation); Median is the median, reflecting the center position of the data distribution (less affected by outliers); SD is the standard deviation, which measures the degree of data dispersion (the larger the value, the more significant the fluctuation); MAD is the median absolute deviation, a dispersion indicator that is resistant to outliers.

[0084] Table 2 further analyzes the annual average elevation change rates for glaciers and non-glacier areas from 2019 to 2021. The mean and median values for GEDI and ICESat-2 are similar in glacier areas, with mean values of -0.60 ± 0.19 m / yr and -0.58 ± 0.15 m / yr, respectively, and medians of -0.70 ± 0.12 m / yr and -0.62 ± 0.08 m / yr, respectively. Because the elevation values collected by GEDI and ICESat-2 are averaged over a 25 m and 17 m diameter range, respectively, centered at their respective footprints, and considering factors such as natural terrain undulations, variations in snow depth, and the operating principles of the laser altimeters, some elevation variation also occurs in non-glacier areas. Judging from the mean values (-0.15±0.23m / yr and -0.11±0.17m / yr) and medians (-0.10±0.17m / yr and -0.09±0.09m / yr) of the elevation change rates, the differences between the two are within a reasonable range. These results indicate that GEDI data can be used as a reliable data source for monitoring glacier elevation changes in alpine areas.

[0085] Table 2. Comparison of the annual average elevation change rates of GEDI and ICESat-2 inside and outside glaciers

[0086]

[0087]

[0088] This table compares the average annual elevation change rates (meters per year, m / yr) of the GEDI and ICESat-2 satellites in the glacier outline and off-glacier outline areas, where Mean represents the arithmetic mean (with ± standard deviation) and Median represents the median (with ± absolute deviation of the median).

[0089] Example 2: GEDI monitors seasonal characteristics of glacier elevation changes.

[0090] Since SETP (Southeast Tibet) is significantly affected by the Indian Ocean monsoon, which carries a large amount of warm and humid air, resulting in abundant precipitation from June to September each year, we further evaluated and analyzed the monitoring capabilities of GEDI from a seasonal perspective. The four seasons are defined as follows: spring (March to May), summer (June to August), autumn (September to November), and winter (December to the following February), abbreviated as Spr. (spring), Sum. (summer), Aut. (autumn), and Win. (winter), respectively. Figure 4The average and difference of elevation changes from 2019 to 2021 obtained by GEDI and ICESat-2 are shown. Since the 2021 data is from January to July, the changes in autumn are not reflected in 2021.

[0091] Overall, the seasonal trends in glacier elevation changes captured by GEDI are similar to those captured by ICESat-2. From spring to winter, glacier loss first increases and then decreases, with the smallest change in spring, followed by winter; the largest change in autumn, followed by summer. The average loss for the two datasets over the four seasons (spring, summer, autumn, and winter) is -3.19 m versus -2.69 m, -3.71 m versus -3.85 m, -4.03 m versus -4.13 m, and -3.72 m versus -3.44 m, respectively. Furthermore, a seasonal regression model was used, fitted using the least squares method, to represent glacier change. The fitted lines are essentially parallel, indicating that the elevation changes and trends captured by GEDI and ICESat-2 are largely consistent. Overall, GEDI slightly overestimates glacier loss across the seasons. Specifically, the difference between GEDI and ICESat-2 is within ±0.8 m, with the largest difference occurring in autumn, with an average absolute value of 0.71 m, and the smallest difference occurring in winter, with an absolute value of 0.29 m. The difference between spring and summer is about 0.5m.

[0092] In summary, the seasonal ablation data show that seasonal variations actually affect the detection capabilities and accuracy of GEDI and ICESat-2, especially GEDI, and are generally consistent with the material replenishment pattern of marine glaciers in the SETP, such as spring snow accumulation and summer precipitation replenishment.

[0093] Example 3: GEDI monitors the change in glacier elevation within a sub-region.

[0094] In addition to temporal differences, spatial regional differences were also considered. To this end, the changes in the four sub-regions where glaciers are mainly concentrated (southern Yigong, eastern Yigong, Mount Namjagbarwa, and eastern Bomi) were analyzed. Figure 4 Shown are the changes in glacier elevation in these sub-regions monitored by GEDI and ICESat-2 from 2019 to 2021.

[0095] In general, the glacier melting in the southern Yigong and Mount Namjagbarwa sub-regions is relatively small, with annual average melting rates of -0.38±0.08m / yr and -0.30±0.05m / yr, and -0.42±0.10m / yr and -0.46±0.03m / yr, respectively. However, the other two sub-regions, the eastern Yigong and the eastern Bomi, have larger changes, with annual average melting rates of -0.58±0.08m / yr and -0.50±0.03m / yr, respectively. 04m / yr and -0.67±0.22m / yr versus -0.53±0.10m / yr, that is, the altitude retreat rate in the east of Bomi is the largest, and the retreat rate in the south of Yigong is the smallest. It is worth noting that the average altitude retreat rate in the east of Yigong is very close to the average altitude change rate in the entire SETP region (GEDI is -0.60±0.19m / yr, ICESat-2 is -0.58±0.15m / yr).

[0096] From the perspective of a single region, compared with ICESat-2, especially in 2021, the amount of altitude change and change trend reflected by GEDI are quite different. In addition to being related to the quality of the GEDI data itself and terrain factors, these differences may also be related to the uneven distribution of data from the two missions in various sub-regions.

[0097] Example 4: GEDI monitors glacier elevation changes at different terrain altitudes.

[0098] To evaluate the impact of topographic altitude factors on the altitude measurement capability of GEDI, the altitude was divided into six groups with an interval of 500 m. Figure 6 The distribution of glaciers at different altitudes and the footprint of GEDI and ICESat-2 are depicted. As can be seen from the figure, glaciers are primarily distributed between 3500 and 6500 meters, with similar trends in glacier number and area as altitude increases. Both missions are concentrated between 4500 and 6000 meters, accounting for 97.96% and 96.32% of the total glaciers, respectively. Glaciers in the 4500-5000, 5000-5500, and 5500-6000m ranges account for 20.12%, 57.89%, and 18.32%, respectively. Similarly, the footprints of GEDI and ICESat-2 are primarily distributed within these three altitude ranges. Both missions have virtually no effective footprint above 6000 meters, and their footprints between 3500 and 4500 meters are also relatively small, so changes in glaciers in these two altitude ranges are not discussed.

[0099] In order to reduce the error caused by the small number of footprints in the low-altitude and high-altitude groups, the altitude change pattern of glaciers mainly distributed between 4500-6000m above sea level and with an interval of 300m was further analyzed, such as Figure 7The results show that glacier retreat generally decreases with increasing altitude until accumulation occurs. Specifically, the greatest glacier retreat occurs between 4500 and 4800 meters, likely due to the smaller glacier area at this altitude. As glacier area increases, the extent of retreat decreases between 4800 and 5700 meters. Glaciers above 5700 meters show signs of accumulation.

[0100] Compared with ICESat-2, GEDI generally overestimates the elevation changes of glaciers at different altitudes, e.g. Figure 8 Specific data show that the annual average elevation changes of GEDI in the five altitude segments are -6.43m, -5.36m, -4.94m, -1.43m and 2.06m, respectively, while the corresponding changes of ICESat-2 are -5.50m, -3.97m, -4.45m, -0.64m and 4.12m, respectively. Within different altitude ranges, the elevation change difference between the two data sets shows a trend of first increasing, then decreasing, and then increasing again. In general, the extent of glacier retreat at different altitudes is mainly affected by factors such as the distribution range of glaciers, the number of glaciers, and the number of effective footprints.

[0101] Example 5: GEDI monitors glacier elevation changes across different terrain slopes.

[0102] Slope reflects the surface morphology and its changes. Slope influences the radiation balance of the glacier surface through the reception and reflection of solar radiation and is a key topographic factor determining the redistribution of material and energy across the glacier surface. The slope referred to here is the average slope of the ice surface, derived from glacier attribute data, and is divided into five levels (5-15°, 15-25°, 25-35°, 35-45°, and ≥45°) with intervals of 10°. Figure 9 Statistics show that as slope gradient increases, the area and number of glaciers increase and then decrease. Glaciers are primarily distributed within a slope range of 15-25°, accounting for 54.86% of the total area and 47.05% of the total number. The GEDI footprint dataset excludes glaciers with steeper slopes to ensure a comprehensive GEDI assessment and provide a reference for future research.

[0103] Figure 10The 2019-2021 GEDI and ICESat-2 monitoring results of glacier elevation changes with slope gradient show a downward trend in glacier elevation across all slopes. As slope gradient increases, the overall glacier retreat shows a trend of first decreasing, then increasing, and then decreasing again. Specifically, the glaciers with the greatest retreat are located in slopes less than 5-15°. As slope gradient increases, glacier retreat decreases between 15-35° and then increases between 35-45°. When slope gradients exceed 45°, the glacier retreat begins to decrease again. Compared with ICESat-2, GEDI generally overestimates the changes in glacier elevation for different slope distributions. The average annual reduction for the five slopes (5-15°, 15-25°, 25-35°, 35-45°, and ≥45°) is -4.82m versus -4.96m, -3.68m versus -3.47m, -3.17m versus -2.04m, -2.62m versus -2.14m, and -2.88m versus -2.25m, respectively.

[0104] Figure 11 The average differences at different slopes were further quantified, and the results showed that the absolute difference was largest between 25-35 degrees between 2019 and 2021. As the slope increased, the standard deviation of the elevation change gradually increased, with the average differences being 0.14m, -0.21m, -1.13m, -0.48m, and -0.63m, respectively.

[0105] Figure 12 The visualization results show that the standard deviation of GEDI's elevation difference ranges from 7.95 to 10.65 m, and the corresponding standard deviation of ICESat-2 ranges from 6.91 to 10.25 m, indicating that the accuracy of the monitoring results gradually decreases with increasing slope.

[0106] Example 6: GEDI monitors the elevation changes of glaciers distributed in different terrain slopes.

[0107] Slope aspect (the azimuth of the slope) is another important topographic factor, defined as the projection direction of the slope surface normal onto the horizontal plane. Slope aspect has a significant impact on the distribution and changes of glaciers, mainly because it affects the amount of solar radiation reaching the glacier surface, thereby affecting the energy and material exchange of the glacier. Figure 13As shown, the terrain is divided into eight levels: north (N), northeast (NE), east (E), southeast (SE), south (S), southwest (SW), west (W), and northwest (NW), based on the average slope direction of each glacier. The results show that glaciers in southeastern Tibet are unevenly distributed across the eight slope directions. Taking the northwest and southeast as the dividing lines, glaciers are significantly concentrated on the north, northeast, and east sides, accounting for 18.26%, 18.52%, and 19.30% of the total glacier area, respectively, and 16.50%, 18.81%, and 14.30% of the total glacier number, respectively. Glaciers on the south, southwest, and west sides are relatively less numerous. Overall, in terms of glacier area and number, glaciers on north-facing slopes in the study area are significantly more numerous than those on south-facing slopes. This may be because north-facing glaciers receive less solar radiation than south-facing glaciers, which facilitates glacier accumulation and ice formation.

[0108] Figure 14 This report depicts changes in glacier elevation from 2019 to 2021 for different slopes, as determined by GEDI and ICESat-2. Comparison of GEDI and ICESat-2 monitoring results reveals that the three slopes with the largest changes in glacier elevation in 2019 coincide with each other: the northwest slope, the north slope, and the west slope. The corresponding changes were -4.21±8.36m vs. 3.8±7.66m, -3.83±8.44m vs. -3.23±8.25m, and -3.67±8.73m vs. 3.82±8.25m, respectively. In 2020, the three slopes with significant changes were also consistent: the north-west slope (-4.37±8.74m vs. -3.84±7.64m), the west slope (-3.61±9.07m vs. -4.44±8.25m), and the north-east slope (-3.85±8.61m vs. -3.89±8.42m). Significant differences occurred on the southeast slope. In 2021, the northwest and west slopes showed significant changes, followed by the north and northeast slopes; significant differences were found on the east and south slopes. Overall, the three slopes with significant changes over the past three years were the northwest, west, and southwest slopes. These three slopes are consistent with the distribution of glaciers, possibly because sunny slopes receive greater solar radiation, which is not conducive to glacier accumulation.

[0109] Example 7: GEDI monitors glacier elevation changes at different glacier areas.

[0110] In order to further explore the impact of glacier area on glacier elevation changes, the glacier area is divided into 8 levels (<0.1km 2 , 0.1-0.2km 2 , 0.2-0.5km 2 , 0.5-1km 2 1-2km2 2-5km 2 5-10km 2 and >10km 2 ). Figure 15 The results show that as the glacier area grade increases, the number of glaciers decreases, and the area shows a trend of first decreasing, then increasing, and then decreasing again. The glacier area is concentrated in the range of 0.2-10km 2 The total area is 3807.42km 2 , accounting for about 59.11% of the total glacier area. Area ≥ 10km 2 The glaciers with the largest area account for 36.06% of the total glacier area. The glaciers with a smaller area, that is, the area <0.1km 2 0.1-0.2km 2 , 0.2-0.5km 2 The glaciers of the following categories account for less than 1% of the total area, which is significantly lower than the glaciers of other area levels. In terms of the number of glaciers, the main feature is that the area is less than 0.5 km 2 The number of glaciers is the largest. The area is less than 0.5km 2 There are 5,225 glaciers, accounting for 72.75% of the total number of glaciers. Among them, the area is less than 0.1 km 2 The largest number of glaciers is in the south, with approximately 2,284, followed by glaciers with an area of 0.2-0.5 km. 2 There are about 1,645 glaciers; but the area is less than 0.5 km 2 Compared with the number of glaciers with an area ≥ 0.5 km 2 The number of glaciers has decreased sharply, with only 1957 glaciers. Among them, those with an area of more than 10 km 2 and area ≥50km 2 There are only 84 and 9 glaciers in China and South Korea respectively.

[0111] Overall, although small glaciers account for a relatively small proportion in terms of area, they account for a large proportion in terms of number. To better evaluate GEDI's monitoring capabilities for glaciers of different sizes, this application does not ignore or exclude these small glaciers as other studies have done.

[0112] From 2019 to 2021, GEDI and ICESat-2 monitoring results of glacier elevation changes in the SETP region showed that glacier elevation showed a downward trend in all regions. Figure 16 As the area increases, the overall glacier retreat height shows a trend of first decreasing, then increasing, and then decreasing again. Specifically, the largest glacier retreat phenomenon is distributed in areas with an area of less than 0.2 km. 2 As the glacier area increases, the 0.2-5km 2 The glacier retreat height gradually decreases, while the 5-10km2 The glacier retreat height gradually increases. When the glacier area is greater than 10km 2 At this time, the glacier elevation retreat began to decrease again.

[0113] Compared with ICESat-2, GEDI overestimates the overall glacier elevation change across different regions, with average decreases of -4.17m and -3.58m in 2019, -4.26m and -4.40m in 2020, and -5.34m and -4.73m in 2021. The annual mean elevation changes for the eight regional layers were -1.24m, -1.26m, -1.10m, -1.04m, -0.90m, -0.69m, -0.70m, and -0.49m for GEDI, respectively, while the corresponding rates of change for ICESat-2 were -1.21m, -1.14m, -1.05m, -0.89m, -0.78m, -0.64m, -0.71m, and -0.43m. The monitoring results of both groups show that small glaciers with an area less than 0.1 km have undergone greater changes than large glaciers. 2 The thinning rate of glaciers with an area greater than 10 km is about 2 Three times the size of the glacier.

[0114] Figure 17 The results further quantified and demonstrated the differences in glacier elevation changes at different glacier area levels. The results showed that the area level was 0.5-1km 2 and 1-2km 2 The glacier elevation changes in the two regions are relatively large, with elevations of -0.77m and -0.54m respectively. In terms of absolute changes, the glacier area is greater than 10km 2 The difference is the largest, which may be caused by the uneven distribution of GEDI and ICESat-2 data.

[0115] Example 8: GEDI monitors glacier elevation changes of different glacier types (with or without surface moraines).

[0116] Glacier surface moraine cover is a key factor influencing glacier change. To investigate the impact of surface moraine cover on GEDI's ability to detect glacier elevation changes, glaciers in the SETP region were divided into two categories: those without surface moraine (no surface moraine) and those with surface moraine (completely covered by surface moraine). The distribution of these two types of glaciers was studied separately. Glacier attribute data revealed that 348 glaciers were covered with moraine, while 6,834 were not, accounting for 4.85% and 95.15%, respectively. Figure 18The elevation changes of two types of glaciers derived from the GEDI and ICESat-2 datasets are shown. Debrisflag1 indicates that these glaciers have moraines, while debrisflag0 indicates that they do not have moraines.

[0117] The results show that regardless of glacier type, their elevations show a common thinning trend, but the thinning is less pronounced for glaciers covered with debris than for those not covered with debris. This suggests that when debris reaches a certain thickness, it effectively reduces the glacier's surface heat absorption capacity, thereby slowing the rate of glacier melt. From 2019 to 2021, the average rates of change for the two types of glaciers were: -2.90 m for moraine-covered glaciers monitored by GEDI, and -4.35 m for moraine-free glaciers; and -2.19 m for moraine-covered glaciers monitored by ICESat-2, and -4.27 m for moraine-free glaciers. When comparing the melt of glaciers without moraines, the average suppression rates estimated by GEDI and ICESat-2 were 34.11% and 47.80%, respectively, indicating that differences in glacier surface properties significantly influence glacial melt and ice loss in southeastern Tibet.

[0118] Figure 2 This is a graph of "Glacier height difference", where "Glacier" refers to glacier, "height" refers to height or elevation, and "difference" refers to difference or change; "ICESat-2" is the Ice, Cloud and Land Elevation Satellite 2; "GEDI" is the Global Ecosystem Dynamics Investigation LiDAR; "2019" is the year, and in the coordinate axes and legend, "Δh" represents the elevation difference; "PointsNumber" refers to the number of data points or the frequency in a specific interval.

[0119] Figure 3 In the data, the title "Off-glacier height difference" means "off-glacier area elevation difference," "off-glacier" refers to areas not covered by glaciers, "height" refers to height or elevation, and "difference" refers to difference. The Y-axis label "Points Number" refers to the number or frequency of data points.

[0120] Figure 4In the figure, the title “Elevation changes and their differences (GEDI-ICESat-2)m” means “Elevation changes and their differences (GEDI vs. ICESat-2)”. The subtitle “Four seasons of each year” means “Four seasons of each year”. In the legend, “GEDI Mean” refers to the mean of GEDI satellite data, “ICESat-2 Mean” refers to the mean of ICESat-2 satellite data, “Mean difference” refers to the difference between the means, “Linear fitting GEDI mean” refers to the linear fit of the GEDI mean, and “Linear fitting ICESat-2 mean” refers to the linear fit of the ICESat-2 mean.

[0121] Figure 5 In the title "Elevation change (m)", it means "elevation change". "Southern_Yigong", "Eastern_Yigong", "Nancha_Barwa", and "Eastern_Bomi" are the names of geographical regions, specifically referring to the southern Yigong Zangbo Basin, the eastern Yigong Zangbo Basin, the Mount Namjagbarwa area, and the eastern part of Bomi County.

[0122] Figure 6 In the report, the main title "Proportion of glacier area and number in different altitude ranges" means "the proportion of glacier area and number in different altitude ranges". "Terrian altitude (m)" refers to "terrain altitude", "Proportion of glacier numbers (%)" refers to "proportion of glacier number". "Area" refers to area, "Number" refers to number, "Proportion of glacier area" refers to the proportion of glacier area, and "Mean values" refers to "mean values". "Mean GEDI" refers to the mean value of GEDI satellite data, and "Mean ICESat-2" refers to the mean value of ICESat-2 satellite data. Another subtitle "Distribution of GEDI and ICESat-2 footprints in different altitude ranges" means "the distribution of GEDI and ICESat-2 footprints in different altitude ranges", and "footprints" here refers to the observation footprints of satellite lidar.

[0123] Figure 7 In the figure, the main title "Glacier elevation distribution (m)" means "Glacier elevation distribution". Another label "Elevation change (m)" means "elevation change", and the numerical range in the figure represents different elevation intervals.

[0124] Figure 8 In the text, the title "Differences of elevation variation (GEDI minus ICESat-2) (m)" means "Differences in elevation variation (GEDI data minus ICESat-2 data)". "Terrian altitude (m)" means "Terrian altitude above sea level". "Average difference" in the caption refers to the average difference line.

[0125] Figure 9 In the table, the title "Glacier numbers" refers to the number of glaciers, and "glacier area" refers to the glacier area. "Terrain slope (°): 5-15" in the table indicates the terrain slope range in degrees. "Glacier numbers" refers to the number of glaciers, and "Glacier area" refers to the glacier area.

[0126] Figure 10 In the image, the main title "Elevation change(m)" means "elevation change"; "Terrain slope(°)" refers to "terrain slope".

[0127] Figure 11 In the image, the main title "Differences of elevation variation (GEDI minus ICESat-2) (m)" means "Differences in elevation variation (GEDI data minus ICESat-2 data)" and the unit is "m." "Terra inslope (°)" refers to "terrain slope," and "Average difference" refers to the average difference.

[0128] Figure 12 In the figure, the main title “SDs of glacier elevation change (m)” means “standard deviation of glacier elevation change”, where “SDs” is the abbreviation of “Standard Deviations” and “glacier elevation change” refers to the change of glacier elevation.

[0129] Figure 13Shows a radar chart that represents different data sets through two lines: "Number of Glaciers" (represented by the blue line) and "Glacier Area" (represented by the orange line).

[0130] Figure 14 This is a polar coordinate graph showing data from GEDI and ICESat-2 in different years, with direction points marked around the graph.

[0131] Figure 15 The main title "Glacier area grade (km 2 )vs.Number of glaciers and Glacierarea(km 2 )” means “the comparative relationship between glacier area grade, number of glaciers and glacier area”, where “Glacier areagrade” refers to glacier area grade, “number of glaciers” refers to the number of glaciers, and “Glacier area” refers to glacier area.

[0132] Figure 16 Main title Glacier area (km 2 ) vs. Elevation change (m)” means “Chart title: Relationship between glacier area and elevation change”.

[0133] Figure 17 The main title "Differences of elevation variation (GEDI minus ICESat-2) (m)" means "Differences of elevation variation (GEDI data minus ICESat-2 data)", with the unit in "m". The subtitle or table column name "Glacier area (km 2 )” refers to “glacier area”, the unit is “km 2 The first column of the table is "Glacier Area Range (km 2 )” refers to “Glacier Area”, and the last column “Average Difference” refers to “Average Difference”.

[0134] Figure 18 The main label or Y-axis label "Elevation change (m)" means "Elevation change".

[0135] Figure 19In the flowchart, in "Step 2: Outlier Removal," "SRTM DEM" refers to the Shuttle Radar Topography Mission Digital Elevation Model, and "3Sigma" refers to a statistical method based on triple standard deviations for identifying outliers. In "Step 3: Quality Constraints," "Quality_flag" refers to the quality flag; "Degrade_flag" indicates whether to degrade the data; "atl06_quality_summary" is a parameter in the ICESat-2 ATL06 product, representing the quality summary flag; and "sigma_geo_h" generally refers to the elevation uncertainty or standard deviation of geolocated heights in ICESat-2 data.

Claims

1. A method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes, characterized in that: The following steps are involved: Preprocessing of laser footprint data in GEDI L2A product and ICESat-2ATL06 product; Using TanDEM-X DEM data as a reference, the elevation difference between each footprint point in the laser footprint data and the DEM was calculated; After removing outliers, the elevation changes between glaciated and non-glaciated areas were calculated; Based on the results of elevation changes, interannual and seasonal analyses were conducted in the overall glacier area and the main glacier aggregation sub-areas, and the GEDI monitoring performance was evaluated in combination with topographic factors and glacier properties.

2. The method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes according to claim 1, characterized in that: The pre-processing step comprises: Extract the ground elevation, longitude, and latitude of the lowest mode echo from GEDI L2A. Filter the data using quality control tags and degraded status tags indicating problems with satellite pointing or positioning information. Filter high-quality data with a quality control tag value of 1 and a degraded status tag value of 0. The latitude and longitude, elevation values, elevation error, and quality control index of the photon point cloud were extracted from ICESat-2ATL06, and the data with a quality control index of 0 and an elevation error of ≤25m were retained.

3. The method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes according to claim 1, characterized in that: Calculating elevation difference includes: The grid values of TanDEM-X DEM were matched to each footprint point using bilinear interpolation; Calculate the difference between the elevation value of each footprint point and the DEM value as the elevation difference; The formula for the elevation change at a single footprint point is as follows: Δh GEDI =H GEDI -H TanDEM-X Δh ICESat-2 =H ICESat-2 -H TanDEM-X in: Δh GEDI and Δh ICESat-2 Indicates the elevation change of each footprint; H GEDI and H ICESat-2 is the height of the GEDI L2A and ICESat-2ATL06 footprints; H TanDEM-X Represents the corresponding TanDEM-X interpolated elevation at each GEDI or ICESat-2 footprint point; A threshold of ±50m was set. If the elevation change exceeded this threshold, it would be considered an outlier and removed. Then, the cumulative elevation change of the entire glacier area from the time of TanDEM-X DEM reference data collection to the time of GEDI and ICESat-2 data collection was estimated by calculating the mean of all valid laser footprints in the glacier area; The calculation expression of the overall glacier elevation change in the glacier area is: in: Δh represents the cumulative elevation change over many years; ∑Δh GEDI / Δh ICESat-2 is the sum of the elevation changes of all valid footprint points of GEDI L2A or ICESat-2; n is the number of valid footprint points of GEDI or ICESat-2; Based on the above-mentioned cumulative elevation changes over many years, the annual average change rate is calculated to reflect the speed of glacier elevation change. The expression for calculating the annual average change rate of glacier elevation is as follows: Where v represents the average annual change rate of elevation; T GEDI / ICESat-2 is the data collection time of GEDI or ICESat-2; T TanDEM-X It is the data product collection time of TanDEM-X DEM.

4. The method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes according to claim 1, wherein: The statistical analysis of elevation changes between glaciated and non-glaciated areas includes: Statistics were conducted in glaciated and non-glaciated areas respectively, and the mean, standard deviation, median, median absolute deviation (MAD) of the cumulative elevation change and the average annual elevation change rate were calculated. The formula for median absolute deviation (MAD) is as follows: MAD=median(|Δh i -m(Δh)|); in: Δh i represents the elevation difference observation value of the i-th footprint point; m(Δh) represents all Δh i the median; median(*) means taking the median of these median absolute deviations; The results from non-glaciated areas are used to verify the accuracy of GEDI products in representing surface elevation and the uncertainty of glacier elevation change estimates.

5. The method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes according to claim 1, characterized in that: The interannual and seasonal analyses include: Collect data from different years separately to conduct interannual variation analysis; Calculate elevation changes in spring, summer, autumn and winter respectively; The seasonal variation trend was fitted using the least squares method.

6. The method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes according to claim 1, characterized in that: The terrain factors include altitude, slope and aspect. The specific analysis includes: Divide the data into segments according to altitude and analyze the elevation changes in different altitude segments; Glaciers are divided into multiple levels according to slope, and the relationship between elevation change and slope is analyzed; Classify by slope direction and analyze the differences in elevation changes under different directions.

7. The method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes according to claim 1, characterized in that: The glacier attributes include glacier area and moraine coverage. Specific analysis includes: Glaciers are divided into multiple levels according to their area, and the elevation changes of each level are counted; The glaciers were divided into two categories: those with and those without surface moraines, and their elevation changes were analyzed; The inhibition rate of the effect of moraine cover on glacier melting was calculated using the following formula: inhibition rate = (change in elevation of glacier without moraine - change in elevation of glacier with moraine) ÷ change in elevation of glacier without moraine × 100%.

8. The method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes according to claim 6, characterized in that: During the slope analysis process, the slope is divided into five levels, including: 5–15°, 15–25°, 25–35°, 35–45°, and greater than or equal to 45°.

9. The method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes according to claim 7, characterized in that: The glacier area classification is based on a multi-level classification standard, which subdivides the glacier area into the following eight ranges: less than 0.1 km 2 , 0.1–0.2km 2 , 0.2–0.5km 2 , 0.5–1km 2 1–2 km 2 2–5 km 2 5–10 km 2 , and greater than or equal to 10km 2 .

10. The method for evaluating the performance of GEDI in monitoring mountain glacier elevation changes according to claim 1, wherein: Using the ICESat-2ATL06 product as reference data, the characteristics of glacier elevation changes revealed by GEDI were evaluated by comparing the elevation differences between GEDI and ICESat-2 in the same area and combining trend analysis methods.

Images