A method, system and medium for identifying biodiversity conservation hotspots

Abstract

The present invention discloses a method, system and medium for identifying biodiversity conservation hotspots, relating to the field. The method includes: obtaining remote sensing image data within a preset year segment in the area to be identified, calculating the inter-annual variation index and spatial variation index of the enhanced vegetation index, as well as the inter-annual variation index and spatial variation index of the land surface temperature, and then determining the spatio-temporal remote sensing index and the hotspots identified based on the spatio-temporal remote sensing index; determining the human activity factors within the area to be identified, calculating the human footprint index and the hotspots identified based on the human footprint index; and determining the biodiversity conservation hotspots within the area to be identified according to the hotspots identified based on the spatio-temporal remote sensing index and the hotspots identified based on the human footprint index. By integrating the spatio-temporal remote sensing index and the human footprint index, the present invention respectively identifies the areas protected due to climate change and human pressure, and accurately identifies the areas that need to be protected preferentially.

Description

Technical Field

[0001] The present invention relates to the field of ecological reserve identification, and particularly to a method, system and medium for identifying hotspots of biodiversity conservation. Background Art

[0002] Human-induced global change has increased the extinction risk of a large number of species on Earth (Ceballos et al., 2015). Establishing protected areas is the primary conservation strategy to prevent biodiversity loss (Lewis et al., 2019), which can minimize habitat loss and more effectively maintain threatened populations (Geldmann et al., 2019). To address the rapid loss of global biodiversity, the number of protected areas has increased significantly in the past two decades. Currently, there are more than 2 million protected areas, accounting for approximately 15% of the global land area (Larsen et al., 2015), but the coverage of existing protected areas is still insufficient to address the current biodiversity crisis. The draft Global Biodiversity Framework Report 2020 proposes to expand protected areas by 30% of the land area by 2030. However, one of the greatest challenges in expanding protected areas is selecting the right ones. Conservation biologists believe that selecting the right protected areas is more important than simply pursuing the coverage rate of protected areas (Pimm et al., 2018). Instead of only focusing on achieving the target of quantitative coverage rate, the efficiency of designated protected areas should be evaluated to determine whether the protected areas also have the highest conservation needs (Barnes et al., 2018). Concentrating efforts on prioritizing the protection of some important areas is a more realistic and efficient approach to current biodiversity conservation. Summary of the Invention

[0003] The purpose of the present invention is to provide a method, system and medium for identifying hotspots of biodiversity conservation, which can accurately identify the areas to be prioritized for protection, thereby precisely protecting the biodiversity within the areas.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A method for identifying hotspots of biodiversity conservation, the method comprising:

[0006] Obtaining remote sensing image data within a preset year segment within the range of the area to be identified;

[0007] Calculating the inter-annual variation index and spatial variation index of the Enhanced Vegetation Index (EVI) and the inter-annual variation index and spatial variation index of the Land Surface Temperature (LST) based on the remote sensing image data;

[0008] Determine the spatio-temporal remote sensing index and the hot spot area identified based on the spatio-temporal remote sensing index according to the EVI inter-annual variation index, the EVI spatial variation index, the LST inter-annual variation index, and the LST spatial variation index;

[0009] Determine the human activity factors within the area to be identified, calculate the human footprint index and the hot spot area identified based on the human footprint index according to the human activity factors; the human activity factors include population density, livestock density, land use, night light index, and traffic impact index;

[0010] Determine the biodiversity conservation hot spot area within the area to be identified according to the hot spot area identified based on the spatio-temporal remote sensing index and the hot spot area identified based on the human footprint index.

[0011] Optionally, calculating the EVI inter-annual variation index of the enhanced vegetation index according to the remote sensing image data specifically includes:

[0012] Determine the enhanced vegetation index value of each pixel within the preset year segment according to the remote sensing image to obtain the enhanced vegetation index time series curve;

[0013] Determine the enhanced vegetation index phenological indicators for each year according to the enhanced vegetation index time series curve to obtain the enhanced vegetation index phenological indicator raster map for each year; the enhanced vegetation index phenological indicator raster map includes the start time raster map of the EVI growing season, the end time raster map of the EVI growing season, and the number of days raster map between the start and end of the EVI growing season; one raster corresponds to one pixel;

[0014] Calculate the inter-annual change index raster map of the start time of the EVI growing season according to the start time raster map of the EVI growing season for each year;

[0015] Calculate the inter-annual change index raster map of the end time of the EVI growing season according to the end time raster map of the EVI growing season for each year;

[0016] Calculate the inter-annual change index raster map of the number of days between the start and end of the EVI growing season according to the number of days raster map between the start and end of the EVI growing season for each year;

[0017] Calculate the correlation between the three inter-annual change indexes according to the inter-annual change index raster map of the start time of the EVI growing season, the inter-annual change index raster map of the end time of the EVI growing season, and the inter-annual change index raster map of the number of days between the start and end of the EVI growing season, and select the most relevant inter-annual change index according to the principal component analysis method to obtain the EVI inter-annual variation index raster map.

[0018] Optionally, calculate the EVI spatial variation index of the enhanced vegetation index based on the remote sensing image data, specifically including:

[0019] Preprocess the remote sensing image data, and generate a composite image by determining the 90th percentile of the enhanced vegetation index values for each year in the preprocessed remote sensing image data;

[0020] Apply an m×m pixel moving window to calculate the standard deviation of the enhanced vegetation index values of each pixel in the composite image to obtain an EVI spatial variation index raster map.

[0021] Optionally, calculate the LST spatial variation index of the land surface temperature based on the remote sensing image data, specifically including:

[0022] Select the remote sensing image data for at least two time periods in each year;

[0023] For the remote sensing image data of each time period, determine the median of the land surface temperature of each pixel to obtain at least two median composite images;

[0024] For each median composite image, apply an m×m pixel moving window to calculate the standard deviation of the land surface temperature of each pixel in the median composite image to obtain at least two LST spatial variation index raster maps.

[0025] Optionally, determine the spatio-temporal remote sensing index and the hotspot area identified based on the spatio-temporal remote sensing index according to the EVI interannual variation index, the EVI spatial variation index, the LST interannual variation index, and the LST spatial variation index, specifically including:

[0026] Integrate the EVI interannual variation index raster map and the EVI spatial variation index raster map to obtain an EVI interannual-spatial variation integrated raster map, and integrate the LST interannual variation index raster map and the LST spatial variation index raster map to obtain an LST interannual-spatial variation integrated raster map; the acquisition process of the LST interannual variation index raster map is the same as that of the EVI interannual variation index raster map;

[0027] Use the Getis-Ord G i * Statistical method is used to perform cluster analysis on the EVI interannual-spatial variation integrated raster map and the LST interannual-spatial variation integrated raster map respectively to obtain a hotspot area map identified based on EVI and a hotspot area map identified based on LST;

[0028] Determine the spatio-temporal remote sensing index and the hotspot area identified based on the spatio-temporal remote sensing index according to the hotspot area map identified based on EVI and the hotspot area map identified based on LST by applying the spatial consistency principle.

[0029] Optionally, integrate the EVI interannual variability index raster map and the EVI spatial variability index raster map to obtain an EVI interannual-spatial variability integrated raster map, and integrate the LST interannual variability index raster map and the LST spatial variability index raster map to obtain an LST interannual-spatial variability integrated raster map, which specifically includes:

[0030] Compare the interannual variability index of each grid in the interannual variability index raster map with a preset value; the interannual variability index raster map includes the EVI interannual variability index raster map and the LST interannual variability index raster map;

[0031] Determine that the grids with the interannual variability index higher than the preset value have high interannual variability, and determine that the grids with the interannual variability index lower than the preset value have low interannual variability;

[0032] Divide the high and low interannual variability regions in the EVI interannual variability index raster map and the LST interannual variability index raster map according to the high and low interannual variability;

[0033] Divide the high and low spatial variability regions in the EVI spatial variability index raster map by using the quantile method according to the standard deviation value of each grid in the EVI spatial variability index raster map;

[0034] Use the quantile method to divide different grade ranges according to the standard deviation value in each LST spatial variability index raster map;

[0035] Combine and sum the grade range values of each grid in each LST spatial variability index raster map, and compare the summation result with a preset sum threshold to determine the high and low spatial variability regions in the LST spatial variability index raster map;

[0036] Integrate the high and low interannual variability regions and the high and low spatial variability regions of the EVI interannual variability index raster map by applying the spatial consistency principle to obtain the EVI interannual-spatial variability integrated raster map;

[0037] Integrate the high and low interannual variability regions and the high and low spatial variability regions of the LST interannual variability index raster map by applying the spatial consistency principle to obtain the LST interannual-spatial variability integrated raster map.

[0038] Optionally, calculate the human footprint index according to the human activity factor and identify the hotspot regions based on the human footprint index, which specifically includes:

[0039] Determine the population density data within the area to be recognized, and assign the population density data to each grid within the area to be recognized according to the population density data to obtain a population density re-assigned grid map;

[0040] Determine the grazing density data within the area to be recognized, and assign the grazing density data to each grid within the area to be recognized according to the grazing density data to obtain a grazing density re-assigned grid map;

[0041] Determine the night-time light index within the area to be recognized, and scale the night-time light index within each grid using quantiles according to the night-time light index to obtain a night-time light index re-scaled grid map;

[0042] Determine the road distribution within the area to be recognized; evaluate and assign values according to the road distribution and the degree of influence on the surrounding area to obtain a traffic impact index assigned grid map;

[0043] Score the land use types within the area to be recognized to obtain a land use score grid map; among the population density re-assigned grid map, the grazing density re-assigned grid map, the night-time light index re-scaled grid map, the traffic impact index assigned grid map, and the land use score grid map, the assigned value range of each grid is the same;

[0044] Sum the assigned values of each grid in the population density re-assigned grid map, the grazing density re-assigned grid map, the night-time light index re-scaled grid map, the traffic impact index assigned grid map, and the land use score grid map to obtain the human impact index corresponding to each grid;

[0045] Calculate the human footprint index according to the human impact index and the maximum and minimum values of the human impact index to obtain a human footprint index grid map;

[0046] Use the natural breaks method to classify the scores of each grid in the human footprint index grid map to obtain a classified grid map;

[0047] Use the Getis-Ord G i * Use the statistical method to perform cluster analysis on the classified grid map to obtain a hot spot area map identified based on the human footprint index.

[0048] The present invention also provides a system for identifying hot spot areas for biodiversity conservation, the system comprising:

[0049] A remote sensing image data acquisition module for acquiring remote sensing image data within a preset year segment within the area to be recognized;

[0050] An interannual variation and spatial variation analysis module, configured to calculate an EVI interannual variation index, an EVI spatial variation index of an enhanced vegetation index, an LST interannual variation index, and an LST spatial variation index of land surface temperature according to the remote sensing image data;

[0051] A first type of hotspot area determination module, configured to determine a spatio-temporal remote sensing index and a hotspot area identified based on the spatio-temporal remote sensing index according to the EVI interannual variation index, the EVI spatial variation index, the LST interannual variation index, and the LST spatial variation index;

[0052] A second type of hotspot area determination module, configured to determine human activity factors within the area to be identified, calculate a human footprint index according to the human activity factors, and a hotspot area identified based on the human footprint index; the human activity factors include population density, livestock density, land use, night light index, and traffic accessibility;

[0053] A biodiversity conservation hotspot area determination module, configured to determine a biodiversity conservation hotspot area within the area to be identified according to the hotspot area identified based on the spatio-temporal remote sensing index and the hotspot area identified based on the human footprint index.

[0054] The present invention further provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, a method for identifying a biodiversity conservation hotspot area is implemented.

[0055] According to the specific embodiments provided by the present invention, the following technical effects are disclosed by the present invention:

[0056] The present invention provides a method, a system, and a medium for identifying a biodiversity conservation hotspot area. The research constructs an identification framework for biodiversity conservation hotspot areas based on two levels of climate change threat and human activity threat. First, according to the interannual and spatial change patterns of the enhanced vegetation index (EVI) and land surface temperature (LST), a spatio-temporal remote sensing index is generated to quantify the threat of climate change to biodiversity. In addition, the impact of human activities on biodiversity is quantified by generating the Inner Mongolia human footprint index. By integrating the spatio-temporal remote sensing index and the human footprint index, areas protected due to climate change and human pressure are identified, and finally, a biodiversity conservation hotspot area is obtained. It can be used for targeted management actions to minimize biodiversity loss. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0058] Figure 1 Flowchart of a method for identifying biodiversity conservation hotspots provided in the first embodiment of the present invention;

[0059] Figure 2 Coefficient of variation distribution of three EVI phenological indicators provided in the first embodiment of the present invention;

[0060] Figure 3 Coefficient of variation distribution of three LST phenological indicators provided in the first embodiment of the present invention;

[0061] Figure 4 Median composite image of LST in summer and winter provided in the first embodiment of the present invention;

[0062] Figure 5 Generating a land surface temperature variability index by combining summer and winter data provided in the first embodiment of the present invention;

[0063] Figure 6 Schematic diagram of the process for generating hotspots identified based on spatio-temporal remote sensing indices provided in the first embodiment of the present invention;

[0064] Figure 7 Distribution of various human footprint influencing factors in Inner Mongolia provided in the first embodiment of the present invention. Detailed implementation manners

[0065] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0066] The object of the present invention is to provide a method, system and medium for identifying biodiversity conservation hotspots, which integrate spatio-temporal remote sensing indices and human footprint indices to respectively identify areas protected due to climate change and human pressure, accurately identify areas for priority protection, and thus precisely protect the biodiversity within the areas.

[0067] To make the above objects, features and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation manners.

[0068] Example 1

[0069] As Figure 1 shown, this embodiment provides a method for identifying hotspots for biodiversity conservation, and the method includes:

[0070] S1: Obtain remote sensing image data within a preset year range in the area to be identified.

[0071] The unique geographical location and topography in Inner Mongolia have created a complex and diverse ecosystem. It is one of the regions with the most complete types of natural ecosystems in China and also one of the key regions for biodiversity conservation in China. In order to effectively address the impacts of climate change and human activities on biodiversity and determine the areas that need to be prioritized for protection. Therefore, this invention takes Inner Mongolia as the area to be identified to illustrate the identification method of this invention, and any area can also be selected according to needs to apply the identification method of this invention.

[0072] S2: Calculate the inter-annual variation index and spatial variation index of the Enhanced Vegetation Index (EVI) and the inter-annual variation index and spatial variation index of the Land Surface Temperature (LST) based on the remote sensing image data.

[0073] Climate change can lead to changes in vegetation phenology and temperature seasonality, which may cause phenological mismatches between species and the resources on which they depend for survival, reproduction, and habitat characteristics. This study uses MODIS multi-temporal remote sensing image data to obtain the annual phenological and seasonal indicators of EVI and LST through the method of fitting curves, generate the inter-annual variation index, and at the same time calculate the Landsat 8 standard deviation image to generate the spatial variation index. Integrate the inter-annual and spatial variation indexes to obtain the spatio-temporal distribution pattern of EVI and LST.

[0074] This invention uses MODIS multi-temporal remote sensing image data to obtain the annual phenological and seasonal indicators of EVI and LST through the method of fitting curves, generate the inter-annual variation index, and at the same time calculate the Landsat 8 standard deviation image to generate the spatial variation index.

[0075] The calculation methods of the EVI inter-annual variation index and the LST inter-annual variation index are the same. The following takes the EVI inter-annual variation index as an example to illustrate the specific calculation process.

[0076] In step S2, calculating the inter-annual variation index of the enhanced vegetation index based on the remote sensing image data specifically includes:

[0077] (1) Determine the enhanced vegetation index value of each pixel within the preset year segment based on the remote sensing image, and obtain the enhanced vegetation index time series curve.

[0078] First, obtain the image from Google Earth Engine and select cloud-free data with the best quality according to the MODIS quality control (QA) band. Use TIMESAT 3.3 software ( & Eklundh, 1964), and use the adaptive Savitzky-Golay filter to filter the time series of EVI and LST from 2002 to 2022 to obtain the enhanced vegetation index time series curve.

[0079] (2) Determine the enhanced vegetation index phenological indicators for each year based on the enhanced vegetation index time series curve, and obtain the enhanced vegetation index phenological indicator raster map for each year; the enhanced vegetation index phenological indicator raster map includes the start time raster map of the EVI growing season, the end time raster map of the EVI growing season, and the number of days raster map between the start and end of the EVI growing season; one raster corresponds to one pixel.

[0080] (3) Determine the start date (Start of growing season, SOS), end date (End of growing season, EOS), and length (Length of growing season, LOS) of the growing season using the dynamic threshold method based on the EVI time series curve. Table 1 shows the three phenological indices of EVI and the three phenological indicators of LST, as well as their respective interannual change indices.

[0081] Table 1 Phenology and seasonal indicators, and interannual change indices

[0082]

[0083] To determine the accuracy of the extracted phenological indicators, this study uses the phenological observation data released by the Data Sharing Network of the Chinese National Ecosystem Observation and Research Network to test the accuracy of SOS and EOS identified by different phenological models. According to the longitude and latitude of the observation points and the year of the phenological data, the average value of 3×3 pixels is extracted from the simulated phenological raster data respectively, and compared with the measured phenological data to calculate the correlation coefficient and root mean square error between the two.

[0084] (4) Calculate the inter - annual variation index raster map of the EVI growing season start time based on the start - time raster map of the EVI growing season for each year. Calculate the inter - annual variation index raster map of the EVI growing season end time based on the end - time raster map of the EVI growing season for each year. Calculate the inter - annual variation index raster map of the number of days between the start and end of the EVI growing season based on the raster map of the number of days between the start and end of the EVI growing season for each year.

[0085] To quantify the inter - annual variation, calculate the coefficient of variation (CV) for each of the six indicators. CV is a statistic that measures the degree of dispersion of sequence observations. This index can reflect the phenological time differences at the pixel level for any land - cover type. The calculation formula is:

[0086]

[0087] In the formula, T CV represents the inter - annual variation index, STD represents the standard deviation of the indicator, and μ represents the arithmetic mean of the indicator.

[0088] (5) Calculate the correlation between the three inter - annual variation indexes based on the inter - annual variation index raster map of the EVI growing season start time, the inter - annual variation index raster map of the EVI growing season end time, and the inter - annual variation index raster map of the number of days between the start and end of the EVI growing season. Select the most relevant inter - annual variation index according to the principal component analysis method to obtain the EVI inter - annual variation index raster map. As Figure 6 (a1) is the distribution of EVI inter - annual variability, Figure 6 (a2) is the distribution of LST inter - annual variability.

[0089] To test whether the phenological and seasonal indicators are correlated, calculate the Pearson correlation coefficient (r) between the three EVI and three LST inter - annual variation indexes respectively to determine whether it is meaningful to retain all indexes, and select the most closely related indicators according to the principal component analysis for further analysis. For both EVI and LST, one most relevant index is selected from the three different inter - annual variation index raster maps according to the correlation analysis for integration with the subsequent spatial variation index raster map.

[0090] Evaluate the inter - annual variation of EVI based on the coefficient of variation of the three phenological indicators of SOS, EOS, and LOS (T CV _E s ; T CV _E e ; T CV _E l)。These three indicators have a medium to high degree of correlation. Taking the Inner Mongolia region as a research example, the Pearson correlation index is between 0.72 and 0.84. The most relevant T is selected through principal component analysis CV _E s Further analyze the interannual variation of EVI, such as Figure 2 shown Figure 2 (a) represents the distribution of the interannual variation index T CV _E s ; Figure 2 (b) represents the distribution of the interannual variation index T CV _E e ; Figure 2 (c) represents the distribution of the interannual variation index T CV _E l . Based on the coefficient of variation of the three indicators of SOS, EOS, and LOS, the interannual variation of LST (T CV _L s ; T CV _L e ; T CV _L l ) is evaluated. The three land surface temperature indices have a medium to high degree of correlation. Taking the Inner Mongolia region as a research example, the Pearson correlation index is between 0.53 and 0.82, and T CV _L s is selected as the interannual variation index of LST. As Figure 3 shown Figure 3 (a) represents the distribution of the interannual variation index T CV _L s ; Figure 3 (b) represents the distribution of the interannual variation index T CV _L e ; Figure 3 (c) represents the distribution of the interannual variation index T CV _L l .

[0091] Among them, in step S2, calculating the EVI spatial variation index of the enhanced vegetation index according to the remote sensing image data includes:

[0092] (1) Preprocess the remote sensing image data, and generate a composite image by determining the 90th percentile of the enhanced vegetation index values for each year of the preprocessed remote sensing image data.

[0093] (2) Apply an m×m pixel moving window to calculate the standard deviation of the enhanced vegetation index values of each pixel in the composite image to obtain an EVI spatial variation index raster map. As Figure 6 (b1) is the EVI spatial variability distribution.

[0094] By processing the standard deviation images of Landsat 8 EVI and LST images on Google Earth Engine and analyzing their image textures, the spatial variability of vegetation greenness and surface temperature is characterized.

[0095] EVI is calculated from Landsat 8 surface reflectance bands 2, 3, and 4. First, according to the quality assurance (QA) band provided by Landsat 8, the images are masked for clouds, shadows, and water bodies to obtain the best-quality images. Second, a composite image is generated by calculating the 90th percentile of EVI values for all years from 2013 to 2022. The 90th percentile is the value that, when the data is sorted by size, divides the data into the lower 90% and the upper 10%. The exceeding 10% of the data will be discarded, which can exclude incorrect high EVI values. Then, an 11×11 pixel moving window is applied to calculate the standard deviation of the EVI 90% image to obtain the spatial variability index of EVI. The central pixel within the moving window obtains the standard deviation value based on the 90th percentile of the EVI of adjacent pixels. The 11×11 window size is selected because from an ecological perspective, a 330-meter window is a refuge area that most species can reach within the area. The formula for calculating the spatial variability index of EVI is as follows:

[0096]

[0097] In the formula, S STD_EVI represents the spatial variability index of EVI, μ is the arithmetic mean of EVI values, n is the sample size, that is, the number of grids (pixels), and xi is the EVI value of the i-th grid.

[0098] Among them, in step S2, calculating the LST spatial variability index of the surface temperature according to the remote sensing image data specifically includes:

[0099] (1) Select the remote sensing image data for at least two time periods in each year. The two selected time periods can be the summer period and the winter period.

[0100] (2) For the remote sensing image data of each time period, determine the median surface temperature of each pixel to obtain at least two median composite images.

[0101] (3) For each of the median composite images, apply an m×m pixel moving window to calculate the standard deviation of the surface temperature of each pixel in the median composite image to obtain at least two LST spatial variability index raster maps.

[0102] The LST is calculated from Band 10 of Landsat 8. First, images of Inner Mongolia region from 2013 to 2022 in each summer (June - August) and winter (December - February) were collected to obtain median composite images of all years in summer and winter, as Figure 4 shown, Figure 4 (a) is the median surface temperature in summer; Figure 4 (b) is the median surface temperature in winter. The median was analyzed because it can minimize the influence of extreme values in the data (Elsen et al., 2020). Based on the median composite images of summer and winter, the standard deviation within an 11×11 pixel moving window in summer and winter was calculated to represent the thermal variation of seasonal extreme weather, as extreme high and low temperatures can both affect biodiversity patterns (Clarke & Gaston, 2006; Elsen et al., 2020). To determine the spatial variability of surface temperature under hot and low temperatures, a surface temperature variability index was generated by combining summer and winter data, as Figure 5 shown, Figure 5 (a1) Spatial variability of summer land surface temperature: defined as the standard deviation (STD) of LST captured in an 11×11 pixel moving window in summer, Figure 5 (b1) Spatial variability of winter land surface temperature: defined as the standard deviation (STD) of LST captured in an 11×11 pixel moving window in winter; Figure 5 (a2) Classification of spatial variability of summer land surface temperature; Figure 5 (b2) Classification of spatial variability of winter land surface temperature; Figure 5 (c) Combining the spatial variability of summer and winter surface temperatures. The summer and winter standard deviation images were divided into five quantile classes, ranging from 1 (low standard deviation value, low variability) to 5 (high standard deviation value, high variability). All possible combinations of the two standard deviation classification images were generated, and the combined values ranged from 2 to 10, as Figure 5 (c). The formula for calculating the spatial variation index of LST is as follows:

[0103]

[0104] In the formula, S STD_LST represents the spatial variation index of LST, μ is the arithmetic mean of the LST value corresponding to the i - th grid, and n is the sample size.

[0105] S3: Determine the spatio - temporal remote sensing index and the hotspots identified based on the spatio - temporal remote sensing index according to the inter - annual variation index of EVI, the spatial variation index of EVI, the inter - annual variation index of LST, and the spatial variation index of LST.

[0106] Among them, step S3 specifically includes:

[0107] S31: Integrate the EVI interannual variability index raster map and the EVI spatial variability index raster map to obtain an EVI interannual-spatial variability integrated raster map, as shown in Figure 6 (c1), and integrate the LST interannual variability index raster map and the LST spatial variability index raster map to obtain an LST interannual-spatial variability integrated raster map, as shown in Figure 6 (c2); The acquisition process of the LST interannual variability index raster map is the same as that of the EVI interannual variability index raster map.

[0108] Among them, step S31 specifically includes:

[0109] S311: Compare the interannual variability index of each grid in the interannual variability index raster map with a preset value; The interannual variability index raster map includes the EVI interannual variability index raster map and the LST interannual variability index raster map;

[0110] S312: Determine the grids with interannual variability index higher than the preset value as having high interannual variability, and determine the grids with interannual variability index lower than the preset value as having low interannual variability.

[0111] S313: Divide the high and low interannual variability regions in the EVI interannual variability index raster map and the LST interannual variability index raster map according to the high and low interannual variability situations.

[0112] S314: Divide the high and low spatial variability regions in the EVI spatial variability index raster map by applying the quantile method according to the standard deviation value of each grid in the EVI spatial variability index raster map.

[0113] S315: Apply the quantile method to divide different grade ranges according to the standard deviation value in each LST spatial variability index raster map.

[0114] S316: Combine and sum the grade range values of each grid in each LST spatial variability index raster map, and compare the summation result with a preset sum threshold to determine the high and low spatial variability regions in the LST spatial variability index raster map.

[0115] Finally, classify the regions where the LST spatial variability in summer and the LST spatial variability in winter are both very high (combined sum ≥ 7) as high spatial variability, and classify the regions where the LST spatial variability in summer and the LST spatial variability in winter are relatively low (combined sum range 2 - 6) as low spatial variability. As shown in Figure 6 (b2) is the distribution of LST spatial variability.

[0116] S317: Apply the spatial consistency principle to integrate the high and low inter-annual variability regions and high and low spatial variability regions of the EVI inter-annual variability index raster map to obtain the EVI inter-annual-spatial variability integrated raster map.

[0117] S318: Apply the spatial consistency principle to integrate the high and low inter-annual variability regions and high and low spatial variability regions of the LST inter-annual variability index raster map to obtain the LST inter-annual-spatial variability integrated raster map, as Figure 6 (c2).

[0118] In steps S317 and S318, before integration, use nearest neighbor resampling to resample the input datasets to match the resolution of the coarsest dataset and create two spatio-temporal change maps.

[0119] By combining the inter-annual variability maps of EVI and LST with their respective spatial variability maps based on high and low variability combinations, as shown in Table 2, and according to the principle of spatial consistency, identify the protected areas caused by phenological and seasonal variability by combining the inter-annual variability index and the spatial variability index.

[0120] Table 1 Division of regional protection levels after inter-annual-spatial variability integration

[0121]

[0122] S32: Use the Getis-Ord statistical method to perform clustering analysis on the EVI inter-annual-spatial variability integrated raster map and the LST inter-annual-spatial variability integrated raster map respectively, and obtain the hot spot area map identified based on EVI (such as Figure 6 (d1)) and the hot spot area map identified based on LST (such as Figure 6 (d2)).

[0123] S33: Determine the hot spot areas based on the spatio-temporal remote sensing index and the hot spot areas identified based on the spatio-temporal remote sensing index according to the principle of spatial consistency based on the hot spot area map identified based on EVI and the hot spot area map identified based on LST, such as Figure 6 (d3).

[0124] S4: Determine the human activity factors within the range of the area to be identified, calculate the human footprint index and the hot spot areas identified based on the human footprint index; the human activity factors include population density, livestock density, land use, night light index, and traffic accessibility.

[0125] The Human Footprint Index (HF) is used to indirectly evaluate the relevant impacts of human activities. By incorporating a series of human pressures into a framework, the relative levels of the pressures are cumulatively aggregated together. Considering the actual situation in Inner Mongolia, five human activity factors are selected, namely population density, livestock density, land use, night-time light index, and traffic accessibility. A human footprint map is drawn at a spatial resolution of 1 km to quantify the interference of humans on the terrestrial ecosystem and determine the HF of Inner Mongolia. The higher the HF value, the greater the impact of human activities. Data related to population density, livestock density, land use, night-time light index, and traffic accessibility are assigned values from 1 - 10 and overlaid to obtain the Human Influence Index (HII), and then the Human Footprint Index is obtained.

[0126] Among them, step S4 specifically includes:

[0127] S41: Determine the population density data within the area to be identified, and assign values to the population density data of each grid within the area to be identified according to the population density data, obtaining a re-assigned population density grid map, as shown in Figure 7 (a).

[0128] Population density is an important indicator of the spatial distribution of the population. This method uses the latest released 2021 population spatial distribution grid data of landscan (https: / / landscan.ornl.gov). The impact of population density on the ecosystem follows a logarithmic variation law. The population density data is calculated and assigned according to the logarithmic equation, and the data is divided into 10 grades using the natural breaks method, with a value range of 0 - 10.

[0129] population score =2.21398×log(population density +1)

[0130] In the formula, population score represents the re-assigned score of this grid, and population density represents the population density value of this grid.

[0131] S42: Determine the grazing density data within the area to be identified, and assign values to the grazing density data of each grid within the area to be identified according to the grazing density data, obtaining a re-assigned grazing density grid map, as shown in Figure 7 (b).

[0132] The livestock density data comes from the World Gridded Livestock Dataset of the Food and Agriculture Organization of the United Nations (https: / / data.apps.fao.org).

[0133] As one of the important industries in Inner Mongolia, grazing is an important type of human activity in this region. In this paper, the sum of cattle and sheep densities is used to represent the grazing density. Since the world grid livestock dataset only obtained the cattle and sheep densities in 2016, before re - assignment, it is necessary to first perform a trend extrapolation analysis on the cattle and sheep densities in 2016 (Duan & Luo, 2020) to obtain the grazing density in 2021. First, according to the statistical yearbook, obtain the total output of beef and mutton in Inner Mongolia Autonomous Region in 2015, 2016, and 2021. Calculate the change rates of beef and mutton production in 2021 relative to the production in 2016 respectively. Multiply the cattle and sheep densities in 2016 by the change rates to obtain the cattle and sheep densities in 2021, and get the final grazing density. Since the impact of grazing density also shows a logarithmic change pattern, the grazing density data is also calculated and assigned according to the logarithmic equation, with a value range of 0 - 10 points. The specific calculation equation is as follows:

[0134] grazing score = 2.51531×log(grazing density + 1)

[0135] In the formula, grazing score represents the score of the re - assignment of this grid, and grazing density represents the livestock density value of this grid.

[0136] S43: Determine the night - time light index within the range of the area to be recognized, and use quantiles to scale the night - time light index in each grid to obtain a rescaled grid map of the night - time light index, as Figure 7 (c).

[0137] The night - time light index is used to represent the degree of night - time human activities and reflects the degree of economic development and the construction of power facilities. The data comes from the China DMSP - OLS time - series data from 1992 to 2019 (Wu et al., 2021). This data calibrates the DMSP - OLS data and considers the consistency of the time resolution between the DMSP - OLS data and the SNPP - VIIRS data, with a spatial resolution of 1 km. The data is scaled to 0 - 10 points using the quantile method.

[0138] S44: Determine the road distribution within the range of the area to be recognized; evaluate and assign values according to the road distribution and the degree of influence on the surrounding area to obtain a grid map of the traffic impact index assignment, as Figure 7 (d).

[0139] The traffic impact index is determined using road and railway data in Inner Mongolia. The road network data is from OpenStreetMap (https: / / www.openstreetmap.org / ). Highways and railways are scored based on research on the environmental impact of roads (Chen et al., 2007; Li et al., 2010; Li et al., 2017b). Since trains are enclosed when running along railways and have a relatively small impact on the areas on both sides, when evaluating and assigning values, only the areas within 2500 m on both sides of the railway are assigned 9 points, the areas within 2500 - 3000 m are assigned 5 points, and the areas within 3000 - 5000 m are assigned 3 points; the areas within 500 m on both sides of expressways are assigned 10 points; the areas within 500 - 2500 m are assigned 8 points; the areas within 2500 - 3000 m are assigned 7 points, and the areas within 3000 - 5000 m are assigned 5 points. The highway and railway buffers are mosaicked into a new grid, and the maximum value of the overlapping part of the railway and highway is obtained to evaluate the accessibility of transportation lines.

[0140] S45: Score the land use types within the range of the area to be identified to obtain a land use score raster map, as Figure 7 (e); in the population density re-assigned raster map, the grazing density re-assigned raster map, the night light index re-scaled raster map, the traffic impact index assigned raster map, and the land use score raster map, the assignment range of each raster is the same.

[0141] Land use is an important indicator to measure the intensity of human activities. Each land use type is scored based on the relative interference degree calculated in relevant studies at the regional scale. The highest score of 10 points is assigned to built-up land (the land use type most affected by humans), and 0 values are assigned to all other land use types with little impact from human activities. Finally, built-up land, cultivated land, grassland, water area, forest land, and unused land are assigned 10, 2, 0, 0, 0, 0 respectively.

[0142] This data is the 2021 global land cover map with a 10-meter resolution (https: / / www.esri.com / ), which is made using the Sentinnel-2 satellite as the data source and combined with an artificial intelligence land classification model.

[0143] S46: Sum the assignments of each raster in the population density re-assigned raster map, the grazing density re-assigned raster map, the night light index re-scaled raster map, the traffic impact index assigned raster map, and the land use score raster map to obtain the human impact index corresponding to each raster. The calculation formula of the human impact index HII is as follows:

[0144] HII = population + grazing + nightlight + landuse + transportation

[0145] Where: population is population density; grazing is grazing density; nightlight is night-time lighting; transportation is the traffic impact index; landuse is land use.

[0146] S47: Calculate the human footprint index based on the human impact index and the maximum and minimum values of the human impact index to obtain a human footprint index raster map.

[0147] Generate an Inner Mongolia human footprint map by calculating the human footprint index to quantify the impact of human activities on the ecosystem. The calculation formula for the human footprint index HF is as follows:

[0148]

[0149] Where: HF is the human footprint; HII is the human impact index. HII max and HII min respectively take the maximum and minimum values of the human impact index.

[0150] S48: Use the natural breaks method to classify the scores of each grid in the human footprint index raster map to obtain a raster map after classification.

[0151] S49: Use the Getis-Ord statistical method to perform cluster analysis on the raster map after classification to obtain a hot spot area map identified based on the human footprint index.

[0152] By using the Getis-Ord statistic (Getis & Ord, 1992) to identify pixel clusters in high and low conservation concern categories caused by climate change and human activities respectively to draw hot spots, which is a local spatial autocorrelation index based on a distance full matrix and can identify areas in the study area where elements are highly concentrated. The formula is as follows:

[0153]

[0154]

[0155] Where, x jTS and x jHF are the attribute values of element j TS and element j HF respectively, and is element i TS and element j TS 、element i HF and j HF is the spatial weight between them, and n is the total number of elements. The statistical data returns a z-score for each element in the dataset. A larger positive z-score value indicates a hot spot, and a larger negative z-score value indicates a cold spot. The hot and cold spot areas are identified with a 99% confidence interval.

[0156] S5: Determine the biodiversity conservation hot spot area within the area to be identified based on the hot spot area identified by the spatio-temporal remote sensing index and the hot spot area identified by the human footprint index.

[0157] Overlay the hot spot spatial maps of the two indices (spatio-temporal remote sensing index and human footprint index), and calculate the biodiversity conservation hot spot area using the formula as follows:

[0158]

[0159] In the formula, is the biodiversity conservation hot spot area identified by combining the two indices. is the hot spot area identified by the spatio-temporal remote sensing index TS, is the hot spot area identified by the human footprint index HF. The calculation method of is as follows: The overlapping part of the two hot spot areas, that is, the area where both indices have significantly high values, is classified as a "Level 1" protected area; the non-overlapping areas, that is, the areas where any one of the two indices has a significantly high value, are classified as "Level 2" protected areas; the areas where neither of the two indices has a high value are classified as "Level 3" protected areas. Finally, the biodiversity conservation hot spot area is divided into three levels of priority protected areas, as shown in Table 3.

[0160] Table 3 Basis for the division of priority protected areas

[0161]

[0162] Integrate the spatio-temporal remote sensing index (Spatial Temporal Remote Sensing Index, ST) and the human footprint index (Human Footprint Index, HF) to identify the areas protected due to climate change and human pressure, and finally obtain the biodiversity conservation hot spot area in Inner Mongolia.

[0163] In this embodiment, based on the ecological theory that climate and environmental factors affect large-scale biodiversity patterns, combined with remote sensing monitoring technology, a system for identifying hotspots for biodiversity conservation is constructed at two levels: the threat level of climate change and the threat level of human activities. Priority areas for biodiversity conservation are divided according to the identification results to protect as much biodiversity as possible. By combining spatio-temporal remote sensing indices and the human footprint index, the threat levels of biodiversity in Inner Mongolia are measured from two major aspects: climate change and human pressure, and priority areas for conservation in Inner Mongolia are demarcated.

[0164] Embodiment 2

[0165] This embodiment provides a system for identifying hotspots for biodiversity conservation, which includes:

[0166] A remote sensing image data acquisition module, configured to acquire remote sensing image data within a preset year range within the area to be identified.

[0167] An inter-annual and spatial variability analysis module, configured to calculate the EVI inter-annual variability index and the EVI spatial variability index of the enhanced vegetation index, and the LST inter-annual variability index and the LST spatial variability index of the land surface temperature according to the remote sensing image data.

[0168] A first type of hotspot determination module, configured to determine a spatio-temporal remote sensing index and hotspots identified based on the spatio-temporal remote sensing index according to the EVI inter-annual variability index, the EVI spatial variability index, the LST inter-annual variability index, and the LST spatial variability index.

[0169] A second type of hotspot determination module, configured to determine human activity factors within the area to be identified, calculate the human footprint index according to the human activity factors, and hotspots identified based on the human footprint index; the human activity factors include population density, livestock density, land use, night-time light index, and traffic accessibility.

[0170] A biodiversity conservation hotspot determination module, configured to determine hotspots for biodiversity conservation within the area to be identified according to the hotspots identified based on the spatio-temporal remote sensing index and the hotspots identified based on the human footprint index.

[0171] Embodiment 3

[0172] This embodiment of the present invention provides a computer-readable storage medium, which stores a computer program that, when executed by a processor, implements the method for identifying hotspots for biodiversity conservation in Embodiment 1.

[0173] In addition, this embodiment also provides an electronic device, including a memory and a processor. The memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the method for identifying biodiversity conservation hotspots in Embodiment 1.

[0174] Optionally, the above-mentioned electronic device may be a server.

[0175] Embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code.

[0176] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0177] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implement the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0178] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0179] In the present specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the various embodiments, reference can be made to each other. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple. For the relevant parts, reference can be made to the description in the method section.

[0180] Specific examples are used in this article to elaborate on the principles and implementation manners of the present invention. The descriptions of the above embodiments are only used to help understand the method of the present invention and its core idea. At the same time, for those of ordinary skill in the art, based on the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation on the present invention.

Claims

1. A method for identifying biodiversity conservation hotspots, characterized in that The method includes: Obtaining remote sensing image data within a preset year segment within the area to be identified; Calculating the inter-annual variation index and spatial variation index of the Enhanced Vegetation Index (EVI) and the inter-annual variation index and spatial variation index of the Land Surface Temperature (LST) based on the remote sensing image data; Determining the spatio-temporal remote sensing index and the hotspot area identified based on the spatio-temporal remote sensing index according to the EVI inter-annual variation index, the EVI spatial variation index, the LST inter-annual variation index, and the LST spatial variation index; Determining the human activity factors within the area to be identified, calculating the human footprint index and the hotspot area identified based on the human footprint index according to the human activity factors; the human activity factors include population density, livestock density, land use, night-time light index, and traffic impact index; Determining the biodiversity conservation hotspot area within the area to be identified according to the hotspot area identified based on the spatio-temporal remote sensing index and the hotspot area identified based on the human footprint index; Calculating the human footprint index and the hotspot area identified based on the human footprint index according to the human activity factors, specifically including: Determining the population density data within the area to be identified, and assigning values to the population density data of each grid within the area to be identified according to the population density data to obtain a population density re-assigned grid map; Determining the grazing density data within the area to be identified, and assigning values to the grazing density data of each grid within the area to be identified according to the grazing density data to obtain a grazing density re-assigned grid map; Determining the night-time light index within the area to be identified, and scaling the night-time light index within each grid using quantiles according to the night-time light index to obtain a night-time light index re-scaled grid map; Determining the road distribution within the area to be identified; evaluating and assigning values according to the road distribution and the influence degree on the surrounding area to obtain a traffic impact index assigned grid map; Scoring the land use types within the area to be identified to obtain a land use scoring grid map; in the population density re-assigned grid map, the grazing density re-assigned grid map, the night-time light index re-scaled grid map, the traffic impact index assigned grid map, and the land use scoring grid map, the assignment range of each grid is the same; Summing up the assignments of each grid in the population density re-assigned grid map, the grazing density re-assigned grid map, the night-time light index re-scaled grid map, the traffic impact index assigned grid map, and the land use scoring grid map to obtain the human impact index corresponding to each grid; Calculating the human footprint index according to the human impact index and the maximum and minimum values of the human impact index to obtain a human footprint index grid map; Using the natural breaks method to classify the scores of each grid in the human footprint index grid map to obtain a classified grid map; Using the Getis-Ord statistical method to perform clustering analysis on the raster map after the level division, and obtain a hotspot area map identified based on the human footprint index; Determine the biodiversity conservation hotspots within the area to be identified based on the hotspots identified by the spatio-temporal remote sensing index and the hotspots identified by the human footprint index, specifically including: Overlay the hotspot spatial maps of the spatio-temporal remote sensing index and the human footprint index, and calculate the biodiversity conservation hotspots using the formula as follows: In the formula, is the biodiversity conservation hotspots identified by combining two indices; is the hotspots identified by the spatio-temporal remote sensing index TS, is the hotspots identified by the human footprint index HF; The calculation method of is as follows: The overlapping part of the two hotspots, that is, the area where both indices have significantly high values, is classified as a "Level 1" protected area; the non-overlapping area, that is, the area where any one of the two indices has a significantly high value, is classified as a "Level 2" protected area; the area where any one value of the two indices is not high is classified as a "Level 3" protected area.

2. The method according to claim 1, characterized in that, Calculate the inter-annual variation index of the Enhanced Vegetation Index (EVI) based on the remote sensing image data, specifically including: Determine the Enhanced Vegetation Index values of each pixel within the preset year segment according to the remote sensing image to obtain the Enhanced Vegetation Index time series curve; Determine the phenological indicators of the Enhanced Vegetation Index for each year based on the Enhanced Vegetation Index time series curve to obtain the phenological indicator raster map of the Enhanced Vegetation Index for each year; the phenological indicator raster map of the Enhanced Vegetation Index includes the start time raster map of the EVI growing season, the end time raster map of the EVI growing season, and the number of days raster map between the start and end of the EVI growing season; one raster corresponds to one pixel; Calculate the inter-annual change index raster map of the start time of the EVI growing season based on the start time raster map of the EVI growing season for each year; Calculate the inter-annual change index raster map of the end time of the EVI growing season based on the end time raster map of the EVI growing season for each year; Calculate the inter-annual change index raster map of the number of days between the start and end of the EVI growing season based on the number of days raster map between the start and end of the EVI growing season for each year; Calculate the correlation between the three inter-annual change indexes based on the inter-annual change index raster map of the start time of the EVI growing season, the inter-annual change index raster map of the end time of the EVI growing season, and the inter-annual change index raster map of the number of days between the start and end of the EVI growing season, and select the most relevant inter-annual change index according to the principal component analysis method to obtain the EVI inter-annual variation index raster map.

3. The method according to claim 1, wherein Calculate the spatial variation index of the Enhanced Vegetation Index (EVI) based on the remote sensing image data, specifically including: Preprocess the remote sensing image data, and generate a composite image based on the 90th percentile of the Enhanced Vegetation Index values for each year determined from the preprocessed remote sensing image data; Apply an m×m pixel moving window to calculate the standard deviation of the Enhanced Vegetation Index values of each pixel in the composite image to obtain the EVI spatial variation index raster map.

4. The method according to claim 3, characterized in that, [[ID= ​ ​ ​ 5. The method according to claim 4, wherein ​ Integrate the EVI interannual variability index raster map and the EVI spatial variability index raster map to obtain an EVI interannual-spatial variability integrated raster map, and integrate the LST interannual variability index raster map and the LST spatial variability index raster map to obtain an LST interannual-spatial variability integrated raster map; the acquisition process of the LST interannual variability index raster map is the same as that of the EVI interannual variability index raster map; Using the Getis-Ord statistical method to perform clustering analysis on the EVI interannual-spatial variation integrated raster map and the LST interannual-spatial variation integrated raster map respectively, and obtain the hot spot area map identified based on EVI and the hot spot area map identified based on LST; Determine the spatio-temporal remote sensing index and the hotspot area identified based on the spatio-temporal remote sensing index according to the hotspot area map identified based on EVI and the hotspot area map identified based on LST by applying the spatial consistency principle.

6. The method according to claim 5, characterized in that, Integrate the EVI interannual variability index raster map and the EVI spatial variability index raster map to obtain an EVI interannual-spatial variability integrated raster map, and integrate the LST interannual variability index raster map and the LST spatial variability index raster map to obtain an LST interannual-spatial variability integrated raster map, specifically including: Compare the interannual variability index of each grid in the interannual variability index raster map with a preset value; the interannual variability index raster map includes the EVI interannual variability index raster map and the LST interannual variability index raster map; Determine that the grids with the interannual variability index higher than the preset value have high interannual variability, and determine that the grids with the interannual variability index lower than the preset value have low interannual variability; Divide the high and low interannual variability regions in the EVI interannual variability index raster map and the LST interannual variability index raster map according to the high and low interannual variability conditions; Divide the high and low spatial variability regions in the EVI spatial variability index raster map according to the standard deviation value of each grid in the EVI spatial variability index raster map by using the quantile method; Use the quantile method to divide the standard deviation values in each LST spatial variability index raster map into different grade ranges; Combine and sum the grade range values of each grid in each LST spatial variability index raster map, and compare the summation result with a preset sum threshold to determine the high and low spatial variability regions in the LST spatial variability index raster map; Integrate the high and low interannual variability regions and the high and low spatial variability regions of the EVI interannual variability index raster map by applying the spatial consistency principle to obtain the EVI interannual-spatial variability integrated raster map; Integrate the high and low interannual variability regions and the high and low spatial variability regions of the LST interannual variability index raster map by applying the spatial consistency principle to obtain the LST interannual-spatial variability integrated raster map.

7. A biodiversity conservation hotspot identification system for implementing the biodiversity conservation hotspot identification method according to any one of claims 1 to 6, characterized in that, The system includes: A remote sensing image data acquisition module for acquiring remote sensing image data within a preset year range in the area to be identified; An interannual variability and spatial variability analysis module for calculating the EVI interannual variability index and the EVI spatial variability index of the enhanced vegetation index and the LST interannual variability index and the LST spatial variability index of the land surface temperature according to the remote sensing image data; The first type of hotspot area determination module is used to determine the spatio-temporal remote sensing index and the hotspot area identified based on the spatio-temporal remote sensing index according to the EVI inter-annual variation index, the EVI spatial variation index, the LST inter-annual variation index, and the LST spatial variation index; The second type of hotspot area determination module is used to determine the human activity factors within the area to be identified, calculate the human footprint index according to the human activity factors, and the hotspot area identified based on the human footprint index; the human activity factors include population density, livestock density, land use, night light index, and traffic accessibility; The biodiversity conservation hotspot area determination module is used to determine the biodiversity conservation hotspot area within the area to be identified according to the hotspot area identified based on the spatio-temporal remote sensing index and the hotspot area identified based on the human footprint index.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method for identifying the biodiversity conservation hotspot area according to any one of claims 1 to 6.

Images