Method and system for dynamically monitoring and evaluating surface environment after burst of lake in plateau frozen soil region

Abstract

The invention belongs to the technical field of surface environment monitoring, and discloses a surface environment dynamic monitoring evaluation method and system after outburst of a plateau permafrost region lake, and the evaluation method comprises the following steps: S1, obtaining long-time-sequence multi-source remote sensing data before and after outburst of a covering target drainage basin, and then carrying out the standardization preprocessing; s2, sampling points are selected in the drainage basin boundary, the environmental indexes of all seasons are calculated, and the environmental indexes comprise LST, NDVI and TVDI; s3, center-of-gravity migration and direction distribution of the LST, the NDVI and the TVDI are obtained; s4, determining frozen soil response key factors, calculating a ground temperature response index, an active layer response index, a ground temperature actual measurement response amount and an active layer actual measurement response amount of the frozen soil response key factors, and judging the frozen soil response type of the area; and step S5, performing visual output. According to the method, the environmental difference caused by surface water change after the lake outburst in the permafrost region can be accurately identified, and scientific support is provided for ecological restoration and frozen soil protection in the plateau frozen soil region.

Description

Technical Field

[0001] This invention relates to the field of surface environment monitoring technology, and in particular to a method and system for dynamic monitoring and assessment of the surface environment after a lake breach in a plateau permafrost region. Background Technology

[0002] Against the backdrop of global warming and increased humidity, lakes in plateau regions are showing an expansion trend. As an important distribution area for lakes, plateaus play a crucial role in the interaction of multiple spheres, including the hydrosphere, atmosphere, and cryosphere, and are sensitive to climate change. Lake expansion can trigger a series of natural disasters, including lake outburst floods. While lake outburst floods in plateau permafrost regions are less frequent than those in glacial regions, once they occur, they significantly alter the distribution of surface water bodies within the watershed, thereby affecting the thermal state of permafrost and the surface ecological environment.

[0003] Existing research on lake outburst floods in plateau regions largely focuses on macroscopic analyses of water distribution changes and permafrost conditions, lacking methods for long-term dynamic monitoring and systematic assessment of multiple surface environmental indicators (such as surface temperature, vegetation cover, and wet / dry conditions) after outburst events. Furthermore, traditional monitoring methods struggle to achieve large-area, long-term continuous observations and lack sufficient analysis of the spatial distribution characteristics and migration patterns of environmental indicators, failing to provide accurate and comprehensive technical support for ecological restoration and environmental management following lake outburst floods in plateau permafrost regions. Therefore, a method capable of integrating multi-source remote sensing data is urgently needed to achieve dynamic monitoring and comprehensive assessment of the surface environment after lake outburst floods in plateau permafrost regions. Summary of the Invention

[0004] The purpose of this invention is to address the technical deficiencies in the existing technology by providing a method for dynamic monitoring and assessment of the surface environment after lake outbursts in plateau permafrost regions.

[0005] Another objective of this invention is to provide a dynamic monitoring and assessment system for the surface environment after a lake breach in a plateau permafrost region.

[0006] The technical solution adopted to achieve the purpose of this invention is: A method for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions includes the following steps: Step S1: Obtain long-term multi-source remote sensing data covering the target watershed before and after the lake outflow, and then perform standardization preprocessing to obtain the basic dataset for each quarter of the long-term year. At the same time, obtain SRTM DEM data and surface water dataset for lake outflow identification. Step S2: Extract the watershed boundary based on the SRTM DEM data from step S1. Based on the basic dataset obtained in step S1, select sampling points within the watershed boundary and calculate the environmental indices for each quarter. The environmental indices include LST, NDVI, and the Temperature Vegetation Dryness-Wetness Index (TVDI). Step S3: Based on the environmental indices obtained in Step S2, perform spatiotemporal analysis. Combine the surface water dataset from Step S1 to systematically quantify and analyze the impact of lake outburst events in the plateau permafrost region on the surface environment. In the time dimension, perform time trend analysis on the environmental indices. In the spatial dimension, perform geographic centroid migration analysis and standard deviation ellipse analysis on the environmental indices to obtain the centroid migration and directional distribution of LST, NDVI, and TVDI. Step S4: Obtain permafrost temperature monitoring profiles based on SRTM DEM data from Step S1, set up monitoring points, monitor permafrost temperature and active layer thickness, and extract quarterly averages of LST, NDVI, and TVDI for the baseline period and assessment period at each monitoring point, using the lake outburst event as the time anchor point. The baseline period is N years before the outburst, and the assessment period is N years after the outburst, where N is 5 to 10. Construct a correlation analysis sample set that includes monitoring point number, monitoring year, monitoring period, changes in environmental indices, changes in ground temperature, and changes in active layer thickness. Pearson correlation analysis was performed on the changes in environmental indices with changes in geothermal temperature and active layer thickness at different depths, respectively, to obtain the correlation coefficient r and significance level p. A correlation coefficient |r| ≥ 0.5 and p were considered significant. An environmental index of 0.05 was identified as a key factor in the permafrost response. The geothermal response index of this key factor was calculated for each monitoring point. Activity layer response index and measured ground temperature response and measured response of the active layer ,in accordance with , , and Determine the type of permafrost response in this area; Step S5: Visualize and export the environmental index obtained in step S2, the spatiotemporal analysis results obtained in step S3, and the permafrost correlation results obtained in step S4.

[0007] In the above technical solution, in step S1, the long-term multi-source remote sensing data comes from the MODIS series dataset. The MODIS series dataset includes the MOD11A2 dataset for extracting land surface temperature (LST) and the MOD13A2 dataset for calculating the normalized vegetation index (NDVI). The span requirement of the MODIS series dataset is no less than 10 years before the flood and no less than 10 years after the flood. The surface water dataset is derived from the JRC Monthly Water History dataset.

[0008] In the above technical solution, the standardization preprocessing in step S1 relies on the GEE platform and includes radiometric calibration and atmospheric correction, target watershed boundary extraction, data clipping and time aggregation.

[0009] In the above technical solution, in step S2, LST is used. ℃ =LST K -273.15 Extracted surface temperature LST, LST ℃ LST is the surface temperature in °C. K LST obtained from the MOD11A2 dataset. K The surface temperature is expressed in Kelvin (K).

[0010] In the above technical solution, the formula for calculating the Normalized Difference Vegetation Index (NDVI) in step S2 is as follows: ; in, Red light reflectance, Near-infrared reflectance, , Obtained from the MOD13A2 dataset; In the above technical solution, the method for calculating the Temperature Vegetation Dryness Index (TVDI) in step S2 is as follows: In ArcGIS, import the Normalized Differential Vegetation Index (NDVI) and Land Surface Temperature (LST) for the same quarter. Construct an NDVI-LST feature space scatter plot with NDVI as the x-axis and LST as the y-axis, and then filter the corresponding values ​​for each NDVI interval. and The wet / dry boundary equation was fitted using linear regression. The wet boundary equation is: The dry boundary equation is ,in, For the maximum LST, To be the minimum LST, a , b For wet boundary coefficient, c , d As the dry boundary coefficient, finally, through the formula The TVDI for each quarter is calculated.

[0011] In the above technical solution, in step S3, the average values ​​of LST, NDVI, and TVDI for each year and quarter before and after the watershed collapse are grouped according to the overall watershed, upstream region, and downstream region. Then, Origin software is used to perform linear regression analysis on each indicator, and the trend is judged by calculating the trend slope.

[0012] In the above technical solution, in step S3, geographical centroid migration analysis and standard deviation ellipse analysis are performed in the spatial dimension. The steps for geographic center of gravity migration analysis are as follows: In ArcGIS, convert LST, NDVI, and TVDI raster data into vector datasets, where each pixel corresponds to a point containing latitude and longitude coordinates and an index value. Then, apply the formula... , Calculate the centroid of environmental indices for different seasons in each year. These are the spatial coordinates of valid LST, NDVI, or TVDI pixels. x , y The latitude and longitude of the center of gravity As the weight of the indicator value, n To determine the number of points, merge centroids and point sets to transform into lines, and analyze the migration patterns; The steps for standard deviation ellipse analysis are as follows: Ellipse center point coordinates , The calculation formula is: ; ; The average center coordinates of the effective pixels in LST, NDVI, or TVDI; Long axis short axis The formula for calculating the length is: ; ; Where: θ is the azimuth angle of the ellipse. , , , They are respectively in , The difference between the spatial coordinates of LST, NDVI, or TVDI in the direction and the average center coordinates; Elliptic flattening The calculation formula is: .

[0013] In the above technical solution, in step S4, when and When it is determined to be a strong ground temperature response, and When it is determined to be a weak ground temperature response, and There is only one item in the middle. The time was determined to be a response in the ground temperature; when and When it is determined to be a strong response of the active layer, and When it is determined to be a weak response of the active layer, and There is only one item in the middle. The response is determined to be a response in the active layer. When both the local temperate and active layers show strong responses, the monitoring location is determined to be a strong response zone of permafrost; when at least one of the local temperate and active layers shows a moderate response, and none of them show a strong response simultaneously, it is determined to be a moderate response zone of permafrost; when both the local temperate and active layers show weak responses, it is determined to be a weak response zone of permafrost.

[0014] In the above technical solutions, , ; in, Let be the ground temperature response index at the i-th monitoring location. Let represent the change in environmental index of the key factor for permafrost response at the i-th monitoring location and the j-th location. The standard deviation of the key factor is the standard deviation of the sample in the same quarter of the baseline period. Let be the Pearson correlation coefficient between the j-th key factor of permafrost response and the change in ground temperature. For the corresponding weighting coefficients, m is the number of key factors that are significantly correlated with the change in geothermal temperature; ; ; in, Let i be the activity layer response index at the i-th monitoring location. Let represent the change in environmental index of the key factor for permafrost response at the i-th monitoring location and the k-th location. The standard deviation of the key factor is the standard deviation of the sample in the same quarter of the baseline period. Let Pearson's correlation coefficient be the k-th key factor in the permafrost response and the change in the thickness of the active layer. For the corresponding weighting coefficients, n is the number of key factors that are significantly related to the change in the thickness of the active layer.

[0015] In the above technical solutions, ; ; Let be the change in ground temperature at the i-th monitoring location. The standard deviation of geothermal samples at the same depth during the baseline period. Let represent the change in the active layer thickness at the i-th monitoring location. This represents the standard deviation of the active layer thickness sample during the baseline period.

[0016] In the above technical solution, the environmental index visualization method in step S5 is as follows: using the hierarchical color setting function of ArcGIS software, spatial distribution maps of surface temperature, normalized vegetation index and temperature-vegetation dryness-wetness index before and after the lake outburst are generated respectively. The visualization method for spatiotemporal analysis results is as follows: By overlaying the centroid points of indicators for each year on a geographic analysis platform, arrows are used to connect them to form a centroid migration trajectory map, and the length of the line segment quantitatively represents the migration distance; at the same time, ellipses of the standard deviation of the years before and after the collapse are drawn in the same coordinate system, and the major axis, minor axis and azimuth parameters are marked in detail to comprehensively show the changes in the spatiotemporal distribution pattern of environmental elements. The visualization method for permafrost correlation is as follows: a bar chart is used to compare and analyze the difference in the thickness of the active layer before and after the collapse of the area. The year is used as the horizontal axis and the thickness is used as the vertical axis to intuitively present the evolution of the permafrost state and effectively evaluate the fitting effect of the permafrost environment correlation model.

[0017] Another aspect of the present invention includes a system for running the dynamic monitoring and assessment method for the surface environment after lake outburst floods in the plateau permafrost region, comprising a data acquisition and preprocessing module, an environmental index calculation module, a spatiotemporal analysis module, a permafrost correlation analysis module, and a result visualization module. The data acquisition and preprocessing module is used to perform step S1, the environmental index calculation module is used to perform step S2, the spatiotemporal analysis module is used to perform step S3, the permafrost correlation analysis module is used to perform step S4, and the result visualization module is used to perform step S5.

[0018] Compared with the prior art, the beneficial effects of the present invention are: This invention integrates multi-source remote sensing data and geographic information technology to achieve long-term monitoring of land surface temperature (LST), vegetation cover (NDVI), and total wet / dry condition (TVDI) and its correlation assessment with permafrost response. This method overcomes the limitations of traditional monitoring and can accurately identify environmental differences caused by changes in surface water after lake outbursts in permafrost regions, providing scientific support for ecological restoration and permafrost protection in plateau permafrost areas. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the method for dynamic monitoring and assessment of the surface environment after a lake outburst in a plateau permafrost region, as described in this embodiment of the invention.

[0020] Figure 2 This is a module composition of the surface environment dynamic monitoring and assessment system after lake outburst in the plateau permafrost region, as described in this embodiment of the invention.

[0021] Figure 3 This is a schematic diagram of the distribution of elements (TVDI) in the spatiotemporal analysis module in this embodiment of the invention, and a spatiotemporal analysis result diagram (centroid migration trajectory + standard deviation ellipse). Detailed Implementation

[0022] The present invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0023] Example 1 like Figure 1 As shown, a method for dynamic monitoring and assessment of the surface environment after a lake outburst flood in a plateau permafrost region includes the following steps: Step S1: Acquire long-term multi-source remote sensing data (MODIS series datasets), SRTM DEM data, and surface water datasets for lake outflow identification covering the target watershed before and after the outflow. Perform standardization preprocessing, data cropping, and temporal aggregation on the MODIS series datasets to obtain the basic datasets for land surface temperature (LST) and normalized difference vegetation index (NDVI) for each quarter of the long-term year. 1.1 Data Acquisition In the data selection phase, the MODIS series datasets were chosen as the source of multi-source remote sensing data. The MODIS series datasets include the MOD11A2 dataset and the MOD13A2 dataset; the MOD11A2 dataset was used for land surface temperature (LST) extraction, and the MOD13A2 dataset was used for normalized difference vegetation index (NDVI) calculation. The time span requirement for the MODIS series datasets is at least 10 years before the flood and at least 10 years after the flood.

[0024] In addition, regarding auxiliary datasets, SRTM DEM data (spatial resolution 30m) was acquired for watershed boundary extraction in step S2, and distribution maps of permafrost types in the target watershed were collected to clarify the boundary between permanent and seasonal permafrost, in order to support the correlation analysis in the subsequent step S4.

[0025] The surface water dataset used is the JRC Monthly Water History dataset, with a resolution of 30m, which can capture abrupt changes in the surface water area.

[0026] 1.2 Standardized Preprocessing Data preprocessing was conducted in batches using the Google Earth Engine (GEE) platform, fully leveraging its cloud computing capabilities to reduce local computing resource consumption. Radiometric calibration and atmospheric correction: The built-in MODIS preprocessing tools in GEE were used to process the MODIS series datasets to obtain accurate surface reflectance data.

[0027] 1.3 Data pruning and time aggregation: Using the generated target watershed boundary vector file, the preprocessed LST and NDVI data were spatially cropped, retaining only valid pixels with cloud coverage of less than 10% within the target watershed. The 8-day LST data and 16-day NDVI data were synthesized into monthly-scale data using the maximum value synthesis method to reduce cloud interference. Subsequently, the data were aggregated by quarter (Q1: January-March; Q2: April-June; Q3: July-September; Q4: October-December) to finally construct the basic datasets of LST and NDVI for each quarter of the long-term year.

[0028] Step S2: Extract the watershed boundary based on the SRTM DEM data obtained in Step S1. Based on the basic dataset of LST and NDVI for each quarter of the long-term time series, select sampling points within the watershed boundary and calculate the three core indices: surface temperature (LST), normalized vegetation index (NDVI), and temperature-vegetation dryness-wetness index (TVDI).

[0029] 2.1 Target watershed boundary extraction: The process was conducted in the ArcGIS environment. After importing the SRTM DEM data obtained in step S1, the "Hydrological Analysis" toolchain was used to perform depression filling, flow direction calculation, runoff accumulation calculation, river network extraction, and watershed segmentation operations in sequence. Combined with the specific location of the outflow lake, the vector file of the target watershed boundary was finally determined.

[0030] 2.2 Extraction of Land Surface Temperature (LST) Using daytime LST products from the MOD11A2 dataset (default unit is K), the LST formula is applied. ℃ =LST K -273.15 is converted to degrees Celsius. Based on this, LST spatial distribution data for each quarter from 2000 to 2023 are obtained, with the value range set at -15℃ to 33℃. Outliers greater than 35℃ or less than -20℃ are removed to ensure the accuracy and reliability of the data.

[0031] 2.3 Calculation of Normalized Difference Vegetation Index (NDVI) Based on red light reflectance from the MOD13A2 dataset ( ) and near-infrared reflectance ( ), using formula NDVI was calculated, with a theoretical range of -1 to 1. To obtain accurate vegetation cover information, the study used the Google Earth Engine (GEE) platform to perform water body masking based on the Normalized Difference Water Index (NDWI, where NDWI > 0 indicates a water body); and further removed NDVI values. Data for non-vegetated areas with NDVI ≥ 0 were used, and only data for vegetation-covered areas with NDVI ≥ 0 were retained to ultimately form the spatial distribution map of NDVI for each quarter.

[0032] 2.4 Construction of Temperature-Vegetation Moisture-Wet Index (TVDI) First, import NDVI and LST data for the same quarter into ArcGIS. Construct an NDVI-LST feature space scatter plot with NDVI (0-1) as the x-axis and LST (258K-306K) as the y-axis. Then, filter the maximum LST corresponding to each NDVI interval (0.05 intervals). ) and minimum LST ( The dry and wet boundary equations were fitted using linear regression. The wet boundary equation is: The dry boundary equation is , a , b For wet boundary coefficient, c , d This is the boundary coefficient. Finally, through the formula... Calculate TVDI, with values ​​ranging from 0 to 1 (values ​​closer to 1 indicate drier conditions, while values ​​closer to 0 indicate wetter conditions), and for TVDI > 1 or ... Outliers of 0 are removed to generate spatial distribution maps of TVDI for each quarter.

[0033] Step S3 involves systematically quantifying and analyzing the impact of lake outburst events on the surface environment in the plateau permafrost region across both time and space dimensions. Specifically, by quantifying the interannual variation trends of environmental indices, the temporal patterns of environmental evolution are revealed. Simultaneously, the migration patterns of environmental index distribution are analyzed to explore their spatial variation characteristics, ultimately achieving accurate identification of the differences in the temporal and spatial impacts of outburst events on the surface environment.

[0034] 3.1 Time Trend Analysis First, data statistics were conducted. Average surface temperature (LST), normalized difference vegetation index (NDVI), and temperature-vegetation dryness-wetness index (TVDI) for each year and quarter before and after the flood were imported into Excel and organized separately for the entire watershed, upstream region, and downstream region. Then, linear regression analysis was performed on each indicator using data analysis software, and the trend was determined by calculating the trend slope.

[0035] 3.2 Spatial distribution characteristic analysis, including geographic centroid migration analysis and standard deviation ellipse analysis. Geographic center of gravity shift analysis: First, convert the 1km resolution LST, NDVI, and TVDI raster data into a vector dataset in ArcGIS, with each pixel corresponding to a point containing latitude and longitude coordinates and an index value. Then, use the formula... , (in x , yThe latitude and longitude of the center of gravity (where n is the number of points) Calculate the index center of gravity for different seasons in each year, and perform operations such as merging and changing the center of gravity to analyze the migration pattern.

[0036] Standard deviation ellipse analysis: In standard deviation ellipse analysis, the standard deviation ellipse parameters of each indicator before and after the collapse are calculated, including the central coordinate, major axis, minor axis, azimuth, and flattening. Among them, the major axis reflects the dominant direction of indicator distribution, the minor axis reflects the distribution range, and the flattening characterizes the directionality (the smaller the flattening, the weaker the directionality).

[0037] First, calculate the center point of the ellipse, then calculate the major and minor axes, and finally calculate the flattening. In calculating the center of the ellipse, based on the spatial coordinates of the geographic elements corresponding to each environmental indicator, first calculate the average center of the elements (…). Then, calculate the standard deviations in the x-axis and y-axis directions using the following formula ( , Determine the center point of the ellipse: ; ; In the formula, The spatial coordinates of geographic features (such as LST, NDVI, or TVDI effective pixels). is the average center coordinate of the geographic feature, and n is the number of geographic features (i.e., the total number of valid pixels).

[0038] Introducing the azimuth angle θ of the ellipse (representing the direction of the major axis), the lengths of the major axis (σx) and minor axis (σy) are calculated using formulas, where... , , representing the difference between the geographic coordinates and the mean center: ; ; In the formula, The standard deviation is the major axis of the ellipse (representing the dominant direction of the data distribution). The minor axis represents the range of data distribution; the greater the difference between the major and minor axes, the greater the extension of the ellipse and the stronger the directionality of the geographic data; the larger the value of the minor axis, the greater the dispersion of the geographic data.

[0039] In the calculation of flattening, the flattening of an ellipse is calculated using a formula. To further quantify the directionality of data distribution, the smaller the flattening, the weaker the data directionality and the more uniform the distribution. ; In addition, index values ​​can be introduced as weights during the calculation process (such as LST value, NDVI value). By weighting the coordinate difference term in the above formula, a weighted standard deviation ellipse is obtained, which highlights the influence of environmental index values ​​and more accurately reflects the spatial distribution characteristics of high and low value areas.

[0040] The analysis results are output from both temporal and spatial dimensions. In the temporal dimension, the interannual variation trend of each indicator is output, including a linear fitting line and marking abrupt change points. In the spatial dimension, the centroid migration trajectory of each indicator and the standard deviation ellipse comparison diagram are output to intuitively present the spatiotemporal variation characteristics of the surface environment after lake outburst floods in the plateau permafrost region.

[0041] Step S4 involves establishing a quantitative correlation model between environmental index changes, permafrost temperature, and the active layer to analyze the permafrost response mechanisms in different regions after a breach, providing a scientific basis for permafrost protection. By integrating multi-source data, a precise characterization and dynamic assessment of permafrost environmental changes can be achieved.

[0042] Multiple monitoring points were set up in the aforementioned key areas, and systematic monitoring was conducted using borehole drilling (10cm diameter, 12m-20m depth). Soil temperature and soil moisture content at different depths were measured simultaneously, and the dynamic temperature changes during freeze-thaw cycles were recorded. The active layer thickness was determined by the maximum annual thaw depth, and the correlation between various environmental indices, permafrost temperature, and active layer thickness was systematically explored using Pearson correlation analysis.

[0043] Furthermore, using the lake outburst event as the time anchor, the quarterly average values ​​of LST, NDVI, and TVDI were extracted at each monitoring point for the baseline period (N years before the outburst, where N is 5-10) and the assessment period (N years after the outburst). Simultaneously, permafrost temperatures at different depths and the active layer thickness for the corresponding years were obtained. A correlation analysis sample set was constructed, including monitoring point number, monitoring year, monitoring period, environmental index changes, ground temperature changes, and active layer thickness changes. Among them, the environmental index changes were defined as the difference between the quarterly average value of each year in the assessment period and the quarterly average value of the baseline period; the ground temperature changes were defined as the difference between the average ground temperature at the corresponding depth in each year in the assessment period and the average ground temperature in the baseline period; and the active layer thickness changes were defined as the offset of the maximum melting depth of the year relative to the average maximum melting depth of the baseline period.

[0044] Based on the aforementioned correlation analysis sample set, Pearson correlation analysis was performed on the changes in LST, NDVI, and TVDI with changes in geothermal temperature at different depths and changes in active layer thickness, respectively, to obtain the correlation coefficient r and significance level p; |r|≥0.5 and p An environmental index of 0.05 was identified as a key factor in the response of permafrost.

[0045] For the key factors of permafrost response that are significantly correlated with the amount of ground temperature change, a ground temperature response index is constructed, and its calculation formula is as follows: ;

[0046] The weighting coefficients are determined by the following formula: ;

[0047] in, Let be the ground temperature response index at the i-th monitoring location. Let represent the change in environmental index of the key factor for permafrost response at the i-th monitoring location and the j-th location. The standard deviation of the key factor is the standard deviation of the sample in the same quarter of the baseline period. Let be the Pearson correlation coefficient between the j-th key factor of permafrost response and the change in ground temperature. For the corresponding weighting coefficients, m represents the number of key factors that are significantly correlated with the change in geothermal temperature.

[0048] For the key factors of frozen soil response that are significantly correlated with the change in active layer thickness, an active layer response index is constructed, and its calculation formula is as follows: ; The weighting coefficients are determined by the following formula: ; in, Let i be the activity layer response index at the i-th monitoring location. Let represent the change in environmental index of the key factor for permafrost response at the i-th monitoring location and the k-th location. The standard deviation of the key factor is the standard deviation of the sample in the same quarter of the baseline period. Let Pearson's correlation coefficient be the k-th key factor in the permafrost response and the change in the thickness of the active layer. For the corresponding weighting coefficients, n is the number of key factors that are significantly related to the change in the thickness of the active layer.

[0049] Furthermore, calculate the measured ground temperature response and the measured active layer response at the i-th monitoring location: ; in, Let be the change in ground temperature at the i-th monitoring location. The standard deviation of geothermal samples at the same depth during the baseline period. Let represent the change in the active layer thickness at the i-th monitoring location. This represents the standard deviation of the active layer thickness sample during the baseline period.

[0050] when and When it is determined to be a strong ground temperature response, and When it is determined to be a weak ground temperature response, and There is only one item in the middle. The time was determined to be a response in the ground temperature; when and When it is determined to be a strong response of the active layer, and When it is determined to be a weak response of the active layer, and There is only one item in the middle. The response is determined to be a response in the active layer. When both the temperature response and the active layer response are strong, the monitoring location is determined to be a strong response zone of permafrost; when at least one of the temperature response and the active layer response is a medium response, and neither is a strong response at the same time, it is determined to be a medium response zone of permafrost; when both the temperature response and the active layer response are weak, it is determined to be a weak response zone of permafrost.

[0051] The permafrost response mechanism refers to the relationship between changes in environmental indices caused by lake outburst disturbances, which further lead to changes in geothermal temperature and active layer thickness at different depths, and the different directions and magnitudes of these changes at different spatial locations.

[0052] Step S5, the environmental index visualization part, uses ArcGIS software's hierarchical color setting function to generate spatial distribution maps of surface temperature, normalized vegetation index, and temperature-vegetation wetness index before and after the lake outburst, respectively, to intuitively reveal the spatial differences of environmental elements at different times; using drawing analysis software, curves of the average changes of environmental indicators in each quarter before and after the lake outburst are drawn, with different colors distinguishing the quarters and vertical lines accurately marking the year of the lake outburst event, clearly showing the temporal evolution characteristics of the environmental index.

[0053] In terms of visualization of spatiotemporal analysis results, the centroids of indicators for each year are overlaid on the geographic analysis platform and connected by arrows to form a centroid migration trajectory map. The length of the line segment quantitatively represents the migration distance. At the same time, ellipses of the standard deviation of the years before and after the collapse are drawn in the same coordinate system, and the major axis, minor axis and azimuth parameters are marked in detail to comprehensively show the changes in the spatiotemporal distribution pattern of environmental elements.

[0054] In the visualization of permafrost correlation results, a bar chart is used to compare and analyze the difference in the thickness of the active layer before and after the collapse of the area. With the year as the horizontal axis and the thickness as the vertical axis, the evolution of the permafrost state is presented intuitively, and the fitting effect of the permafrost environment correlation model is effectively evaluated.

[0055] Example 2 like Figure 2As shown, this embodiment provides a dynamic monitoring and assessment system for the surface environment after lake outburst floods in plateau permafrost regions. It includes a data acquisition and preprocessing module, an environmental index calculation module, a spatiotemporal analysis module, a permafrost correlation analysis module, and a results visualization module. These five collaborative technical modules form a complete technology chain of "data-computation-analysis-correlation-visualization".

[0056] Through the coordinated operation of five modules, the system enables accurate identification of lake outburst events and dynamic monitoring of the surface environment. Each module has different functions and technical approaches.

[0057] The data acquisition and preprocessing module aims to acquire long-term, multi-source remote sensing data covering the period before and after the watershed overflow, and to perform standardized preprocessing. By eliminating various interference factors, a high-quality, directly usable basic dataset is constructed, laying the data foundation for subsequent monitoring and assessment work.

[0058] Based on preprocessed data, the environmental index calculation module constructs a calculation system for three core environmental indices: land surface temperature (LST), normalized difference vegetation index (NDVI), and temperature-vegetation moisture index (TVDI). By quantifying key environmental elements such as land surface temperature, vegetation cover, and moisture conditions, it provides scientific and reliable indicator support for the dynamic analysis of the surface environment after lake outburst floods in the plateau permafrost region.

[0059] The spatiotemporal analysis module systematically quantifies and analyzes the impacts of lake outburst events on the surface environment in the plateau permafrost region across both temporal and spatial dimensions. Specifically, it reveals the temporal patterns of environmental evolution by quantifying the interannual variation trends of environmental indices; simultaneously, it analyzes the migration patterns of environmental indicator distributions to explore their spatial variation characteristics, ultimately achieving accurate identification of the spatiotemporal differences in the impact of outburst events on the surface environment.

[0060] The permafrost correlation analysis module establishes a quantitative correlation model between environmental index changes, permafrost temperature, and the active layer, analyzing the permafrost response mechanisms in different regions after a breach, and providing a scientific basis for permafrost protection. By integrating multi-source data, it achieves accurate characterization and dynamic assessment of permafrost environmental changes.

[0061] The results visualization module, as the output of the technical process, transforms the analysis results of each module into various forms of visualization products, clearly presenting the lake outburst identification results and the characteristics of surface environmental changes.

[0062] Example 3 This embodiment provides a comprehensive and efficient scheme for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions. Based on the system's modular architecture and actual surface environmental characteristics, it enables long-term time-series analysis from 2000 to 2024.

[0063] The data acquisition and preprocessing module, as the core data source of the entire technical process, mainly relies on platforms such as Google Earth Engine (GEE) and USGS Earth Explorer to complete multi-source data collection and standardization processing, laying the foundation for subsequent lake outburst flood identification and environmental index calculation. Specifically, long-term remote sensing data from 2000 to 2024 were acquired through the GEE platform, including the MOD11A2 dataset for extracting land surface temperature (LST), the MOD13A2 dataset for calculating the normalized difference vegetation index (NDVI), and a surface water dataset for lake outburst flood identification. Simultaneously, 30m resolution SRTM DEM data was acquired through the USGS platform to extract topographic information (elevation, slope) of the study area, providing support for topographic correlation analysis of outburst locations. The acquired data were preprocessed according to the steps in Example 1.

[0064] The environmental index calculation module takes over the data preprocessed by the data acquisition and preprocessing module. On the one hand, it uses an improved water index algorithm to identify key parameters of lake outburst events. On the other hand, it calculates environmental indicators such as LST, NDVI, and TVDI to construct a quantitative analysis of environmental changes before and after the outburst event.

[0065] In terms of environmental index calculation, TVDI was constructed based on the GEE platform: LST and NDVI data corrected by the data acquisition and preprocessing modules were called to construct the LST-NDVI feature space of the study area; LST and NDVI values ​​of each sampling point were extracted by distributing uniform sampling points, and the sampling point data were fitted to the boundary using the linear regression method to obtain the dry boundary equation and the wet boundary equation; based on the fitted dry and wet boundaries, the difference between the actual LST of the pixel and the LST of the wet boundary was calculated, and then divided by the difference between the LST of the dry boundary and the LST of the wet boundary to obtain the pixel-by-pixel TVDI value of the study area (range 0-1, the larger the value, the more severe the drought in the area), while retaining the original calculation results of LST and NDVI to form three types of core environmental index datasets.

[0066] The spatiotemporal analysis module, as a core component of dynamic monitoring of the land surface environment, primarily relies on the ArcGIS platform. Through spatiotemporal dual-dimensional analysis, it reveals the changing characteristics of LST, NDVI, and TVDI before and after the lake outburst flood, specifically including two parts: temporal trend analysis and spatial distribution characteristic analysis. In the temporal trend analysis, based on the environmental index dataset generated by the environmental index calculation module, the average values ​​of LST, NDVI, and TVDI for each year before and after the outburst flood are extracted, and linear regression analysis is performed to calculate the slope of change and its variability. Regarding spatial distribution characteristic analysis, two key analyses are implemented using ArcGIS tools: first, geographic centroid migration analysis. The TVDI, NDVI, and LST data obtained from the environmental index calculation module are first vectorized, and then, based on the index values ​​of each point as weights, the geographic centroid of each summer is calculated, intuitively revealing the evolutionary patterns of the spatial distribution of environmental elements.

[0067] The permafrost correlation analysis module integrates the spatiotemporal analysis results from the spatiotemporal analysis module with field-collected permafrost data to establish a quantitative correlation between environmental index changes and permafrost status after lake outburst floods, thus deepening the ecological impact assessment of outburst events. Specifically, based on the core outburst impact area identified by the spatiotemporal analysis module, and combined with permafrost monitoring point data (0-20m soil temperature and active layer thickness) deployed by the data acquisition and preprocessing module, Pearson correlation analysis is used to calculate the correlation between permafrost active layer thickness and LST, NDVI, and TVDI, thereby accurately assessing the degree of permafrost degradation after an outburst and providing a scientific basis for engineering protection in cold regions.

[0068] The results visualization module, as the output of the technical workflow, transforms the analysis results of the first four modules into various forms of visualization products, clearly presenting the lake outburst flood identification results and surface environment change characteristics. Specifically, this includes: first, time-series charts, generating line graphs showing the changes in the mean values ​​of LST, NDVI, and TVDI before and after the outburst (with outburst nodes marked); second, spatial thematic maps, creating outburst location distribution maps and TVDI drought distribution maps in ArcGIS (with the center migration trajectory marked) (e.g., Figure 3 The system supports the export of preprocessed data (GeoTIFF format), index calculation results (CSV format), and vector boundaries (Shapefile format) for subsequent engineering design and ecological restoration applications.

[0069] Through the progressive and synergistic interaction of five modules, this invention achieves full-process coverage from lake outburst flood identification to environmental impact assessment. The technical solution is both accurate and practical, and can be widely applied to the dynamic monitoring of the surface environment of lake outburst flood events in high-altitude cold regions.

[0070] The above description is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions, characterized in that, Includes the following steps: Step S1: Obtain long-term multi-source remote sensing data covering the target watershed before and after the lake outflow, and then perform standardization preprocessing to obtain the basic dataset for each quarter of the long-term year. At the same time, obtain SRTM DEM data and surface water dataset for lake outflow identification. Step S2: Extract the watershed boundary based on the SRTM DEM data from Step S1. Based on the basic dataset obtained in Step S1, select sampling points within the watershed boundary and calculate the environmental indices for each quarter. The environmental indices include land surface temperature (LST), normalized difference vegetation index (NDVI), and temperature vegetation dryness index (TVDI). Step S3: Based on the environmental indices obtained in Step S2, perform spatiotemporal analysis. Combine the surface water dataset from Step S1 to systematically quantify and analyze the impact of lake outburst events in the plateau permafrost region on the surface environment. In the time dimension, perform time trend analysis on the environmental indices. In the spatial dimension, perform geographic centroid migration analysis and standard deviation ellipse analysis on the environmental indices to obtain the centroid migration and directional distribution of LST, NDVI, and TVDI. Step S4: Obtain permafrost temperature monitoring profiles based on SRTM DEM data from Step S1, set up monitoring points, monitor permafrost temperature and active layer thickness, and extract quarterly averages of LST, NDVI, and TVDI for the baseline period and assessment period at each monitoring point, using the lake outburst event as the time anchor point. The baseline period is N years before the outburst, and the assessment period is N years after the outburst, where N is 5 to 10. Construct a correlation analysis sample set that includes monitoring point number, monitoring year, monitoring period, changes in environmental indices, changes in ground temperature, and changes in active layer thickness. Pearson correlation analysis was performed on the changes in environmental indices with changes in geothermal temperature and active layer thickness at different depths, respectively, to obtain the correlation coefficient r and significance level p. A correlation coefficient |r| ≥ 0.5 and p were considered significant. An environmental index of 0.05 was identified as a key factor in the permafrost response. The geothermal response index of this key factor was calculated for each monitoring point. Activity layer response index and measured ground temperature response and measured response of the active layer ,in accordance with , , and Determine the type of permafrost response in this area; Step S5: Visualize and export the environmental index obtained in step S2, the spatiotemporal analysis results obtained in step S3, and the permafrost correlation results obtained in step S4.

2. The method for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions as described in claim 1, characterized in that, In step S1, the long-term multi-source remote sensing data comes from the MODIS series datasets. The MODIS series datasets include the MOD11A2 dataset for extracting land surface temperature (LST) and the MOD13A2 dataset for calculating the normalized vegetation index (NDVI). The MODIS series datasets are required to span at least 10 years before the landslide and at least 10 years after the landslide. The surface water dataset is derived from the JRC Monthly Water History dataset. In step S1, the standardization preprocessing relies on the GEE platform and includes radiometric calibration and atmospheric correction, data cropping and time aggregation.

3. The method for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions as described in claim 1, characterized in that, In step S2, through LST ℃ =LST K -273.15 Extracted surface temperature LST, LST ℃ LST is the surface temperature in °C. K LST obtained from the MOD11A2 dataset. K Surface temperature is expressed in Kelvin (K). The formula for calculating the Normalized Difference Vegetation Index (NDVI) is: ； in, Red light reflectance, Near-infrared reflectance, , Obtained from the MOD13A2 dataset; The Temperature-Vegetation Moisture Index (TVDI) is calculated as follows: In ArcGIS, import the Normalized Differential Vegetation Index (NDVI) and Land Surface Temperature (LST) for the same quarter. Construct an NDVI-LST feature space scatter plot with NDVI as the x-axis and LST as the y-axis, and then filter the corresponding values ​​for each NDVI interval. and The wet / dry boundary equation was fitted using linear regression. The wet boundary equation is: The dry boundary equation is ,in, For the maximum LST, To be the minimum LST, a , b For wet boundary coefficient, c , d As the dry boundary coefficient, finally, through the formula The TVDI for each quarter is calculated.

4. The method for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions as described in claim 1, characterized in that, In step S3, the average values ​​of LST, NDVI, and TVDI for each year and quarter before and after the watershed collapse are grouped according to the overall watershed, upstream region, and downstream region. Then, Origin software is used to perform linear regression analysis on each indicator, and the trend is judged by calculating the trend slope. In step S3, geographical centroid migration analysis and standard deviation ellipse analysis are performed in the spatial dimension. The steps for geographic center of gravity migration analysis are as follows: In ArcGIS, convert LST, NDVI, and TVDI raster data into vector datasets, where each pixel corresponds to a point containing latitude and longitude coordinates and an index value. Then, apply the formula... , Calculate the centroid of environmental indices for different seasons in each year. These are the spatial coordinates of valid LST, NDVI, or TVDI pixels. x , y The latitude and longitude of the center of gravity As the weight of the indicator value, n To determine the number of points, merge centroids and point sets to transform into lines, and analyze the migration patterns; The steps for standard deviation ellipse analysis are as follows: Ellipse center point coordinates , The calculation formula is: ； ； The average center coordinates of the effective pixels in LST, NDVI, or TVDI; Long axis short axis The formula for calculating the length is: ； ； Where: θ is the azimuth angle of the ellipse. , , , They are respectively in , The difference between the spatial coordinates of LST, NDVI, or TVDI in the direction and the average center coordinates; Elliptic flattening The calculation formula is: 。 5. The method for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions as described in claim 1, characterized in that, In step S4, the change in environmental index is the difference between the quarterly average of each year during the assessment period and the quarterly average of the baseline period; the change in geothermal temperature is the difference between the average geothermal temperature at the corresponding depth each year during the assessment period and the average geothermal temperature of the baseline period; and the change in active layer thickness is the offset of the maximum melting depth of the year relative to the average maximum melting depth of the baseline period.

6. The method for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions as described in claim 1, characterized in that, In step S4, when and When it is determined to be a strong ground temperature response, and When it is determined to be a weak ground temperature response, and There is only one item in the middle. The time was determined to be a response in the ground temperature; when and When it is determined to be a strong response of the active layer, and When it is determined to be a weak response of the active layer, and There is only one item in the middle. The response is determined to be a response in the active layer. When both the local temperate and active layers show strong responses, the monitoring location is determined to be a strong response zone of permafrost; when at least one of the local temperate and active layers shows a moderate response, and none of them show a strong response simultaneously, it is determined to be a moderate response zone of permafrost; when both the local temperate and active layers show weak responses, it is determined to be a weak response zone of permafrost.

7. The method for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions as described in claim 1, characterized in that, In step S4 , ; in, Let be the ground temperature response index at the i-th monitoring location. Let represent the change in environmental index of the key factor for permafrost response at the i-th monitoring location and the j-th location. The standard deviation of the key factor is the standard deviation of the sample in the same quarter of the baseline period. Let be the Pearson correlation coefficient between the j-th key factor of permafrost response and the change in ground temperature. For the corresponding weighting coefficients, m is the number of key factors that are significantly correlated with the change in geothermal temperature; ； ； in, Let i be the activity layer response index at the i-th monitoring location. Let represent the change in environmental index of the key factor for permafrost response at the i-th monitoring location and the k-th location. The standard deviation of the key factor is the standard deviation of the sample in the same quarter of the baseline period. Let Pearson's correlation coefficient be the k-th key factor in the permafrost response and the change in the thickness of the active layer. For the corresponding weighting coefficients, n is the number of key factors that are significantly related to the change in the thickness of the active layer.

8. The method for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions as described in claim 1, characterized in that, In step S4 ; ; Let be the change in ground temperature at the i-th monitoring location. The standard deviation of geothermal samples at the same depth during the baseline period. Let represent the change in the active layer thickness at the i-th monitoring location. This represents the standard deviation of the active layer thickness sample during the baseline period.

9. The method for dynamic monitoring and assessment of the surface environment after lake outburst floods in plateau permafrost regions as described in claim 1, characterized in that, In step S5, the environmental index visualization method is as follows: using the hierarchical color setting function of ArcGIS software, spatial distribution maps of surface temperature, normalized vegetation index and temperature-vegetation dryness-wetness index are generated before and after the lake breach. The visualization method for spatiotemporal analysis results is as follows: By overlaying the centroid points of indicators for each year on a geographic analysis platform, arrows are used to connect them to form a centroid migration trajectory map, and the length of the line segment quantitatively represents the migration distance; at the same time, ellipses of the standard deviation of the years before and after the collapse are drawn in the same coordinate system, and the major axis, minor axis and azimuth parameters are marked in detail to comprehensively show the changes in the spatiotemporal distribution pattern of environmental elements. The visualization method for permafrost correlation is as follows: a bar chart is used to compare and analyze the difference in the thickness of the active layer before and after the collapse of the area. The year is used as the horizontal axis and the thickness is used as the vertical axis to intuitively present the evolution of the permafrost state and effectively evaluate the fitting effect of the permafrost environment correlation model.

10. A system for operating the dynamic monitoring and assessment method for the surface environment after lake outburst floods in plateau permafrost regions as described in any one of claims 1 to 9, characterized in that, It includes a data acquisition and preprocessing module, an environmental index calculation module, a spatiotemporal analysis module, a permafrost correlation analysis module, and a result visualization module. The data acquisition and preprocessing module is used to perform step S1, the environmental index calculation module is used to perform step S2, the spatiotemporal analysis module is used to perform step S3, the permafrost correlation analysis module is used to perform step S4, and the result visualization module is used to perform step S5.

Images