Regional climate shelter identification method based on climate change and species distribution model

Abstract

The invention relates to a regional climate shelter identification method based on climate change and a species distribution model. The method comprises the following steps: S1, species distribution modeling; step S2, biological velocity calculation; s3, calculating the richness of the species; s4, calculating the habitat quality; step S5, climate stability calculation; step S6, calculating a connectivity index and a reachability index; the method has the advantages that dimensions of three information including space, time and ecology can be effectively captured, species richness, habitat quality, climate stability and biological speed are integrated, the defect that a climate speed index neglects species specific diffusion rate and habitat suitability in shelter recognition is made up, and the method has the advantages of being high in adaptability and high in reliability. And the method is limited by a climate matching threshold value, so that the climate refuge can be identified more accurately.

Description

Technical Field

[0001] This invention relates to the field of biodiversity research and conservation, and in particular to a method for identifying regional climate refuges based on climate change and species distribution models. Background Technology

[0002] Climate change is causing significant alterations in the distribution and abundance of species worldwide and may become the most severe threat to biodiversity this century. Therefore, it is necessary to optimize and develop protected area networks to support species' range shifts in response to climate change. Climate refuges, which are buffer zones that protect species from the effects of climate change over time and provide stable climatic conditions, have enormous potential for protected area expansion and spatial optimization.

[0003] Currently, the identification of climate refuges mainly relies on climate velocity indicators. By identifying areas with similar climates under future conditions to the current climate, the minimum migration rate required for species to reach these areas is calculated, and low-velocity areas are designated as potential refuges. However, this method relies on a relatively narrow range of climate matching thresholds, implicitly assumes that local populations have almost no climate tolerance, and ignores the limitations of species distribution. Therefore, it is difficult to find climate refuges that can be targeted at specific populations.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention discloses a method for identifying regional climate refuges based on a climate change and species distribution model.

[0006] A method for identifying regional climate refuges based on climate change and species distribution models includes the following steps:

[0007] Step S1: Species distribution modeling. First, based on the species distribution information, land use data and elevation data of the target area, the habitat range (AOH) of the species is selected, and species distribution modeling is carried out in combination with environmental variables.

[0008] Step S2: Biovelocity Calculation. Based on the species binary habitat suitability map generated by the species distribution model for the current and future periods, the forward biovelocity and backward biovelocity of the species are calculated.

[0009] Step S3: Species richness calculation. Based on the species distribution model, the current and future binary habitat suitability maps of species are generated, and weights are assigned according to the endangerment level of each species. Then, in each grid cell with an area of ​​1 km2, the weighted sum of habitat suitability of all species of the same group is calculated to obtain the weighted species richness map.

[0010] Step S4: Habitat quality calculation, integrating land use data, biodiversity threat factors and sensitivity factors to generate a habitat quality map of the target area;

[0011] Step S5: Climate stability calculation. Principal component analysis (PCA) is performed on the current and future climate datasets to reduce dimensionality. The climate change intensity value (CCI) is calculated using Euclidean distance. A climate stability map is obtained based on the CCI.

[0012] Step S6: Calculation of connectivity index and accessibility index. The connectivity index CI is obtained based on the forward biovelocity and average dispersal rate of the species; the accessibility index AI is obtained based on the backward biovelocity and average dispersal rate of the species.

[0013] Step S7: Identification of potential climate refuges. The weighted species richness, habitat quality and climate stability data are normalized and multiplied to obtain the habitat sustainability index HPI. Based on the habitat sustainability index HPI and connectivity index CI, the stepping stone climate refuge potential index CRPInh is calculated. Based on the habitat sustainability index HPI and accessibility index AI, the new habitat type climate refuge potential index CRPInh is obtained. Based on the obtained CRPInh, the hotspot distribution areas of climate refuges are identified.

[0014] Furthermore, in step S1, after collecting habitat range AOH and converting it into point data, 10 variables with no strong correlation were selected from 25 initial environmental variables through variance inflation factor (VIF) test and Pearson correlation analysis as environmental factors for species distribution model. Speciation records were added as a supplement to the above species distribution points. Both were input into megaSDM for species distribution modeling.

[0015] Furthermore, in step S2, the forward and backward biological velocities required for each species are first calculated, and then the velocities of all species belonging to the same group are averaged to obtain the average forward and backward biological velocities for each group of species.

[0016] Furthermore, in step S3, the endangered status weights are assigned as follows: Vulnerable (Vu) = 2, Endangered (EN) = 4, Critically Endangered (CR) = 8; the species habitat suitability weighted sum is calculated using the following formula: Where Si represents whether species i has a suitable (1) or unsuitable (0) habitat in this grid cell, and Wi represents the threat level of each species.

[0017] Furthermore, in step S4, biodiversity threat factors include paddy fields, dry land, urban land, rural settlements, other construction land, and unused land.

[0018] Furthermore, in step S5, several current and future bioclimatic variables are transformed into independent principal component (PC) axes. The first four principal components are selected to calculate the CCI, using the following formula: Where xi and yi represent the current and future climate values ​​of the i-th principal component axis, respectively.

[0019] Furthermore, in step S6, Where Vfor represents the forward biovelocity, and Vmean represents the average diffusion rate of each species.

[0020] Furthermore, in step S6, Where Vback represents the backward biovelocity, and Vmean represents the average diffusion rate of each species.

[0021] Furthermore, in step S7, the stepping stone climate refuge potential index CRP1ss = HPIj × CIj, where HPIj represents the habitat sustainability index of species j and CIj represents the connectivity index CI of species j; the new habitat climate refuge potential index CRPInh = HPIj × AIj, where HPIj represents the habitat sustainability index of species j and AIj represents the accessibility index AI of species j.

[0022] Furthermore, in step S7, the top 25 percentile regions in the two types of climate refuge potential index maps are extracted as climate refuge hotspots and compared with protected areas to reveal protection gaps in the existing protected area network.

[0023] The advantages of this invention are:

[0024] 1. This invention can effectively capture the three dimensions of spatial, temporal and ecological information, and integrate species richness, habitat quality, climate stability and biological velocity. It makes up for the shortcomings of climate velocity indicators in identifying refuges, such as ignoring species-specific dispersal rates and habitat suitability, and being limited by climate matching thresholds. It can more accurately identify climate refuges and provide a forward-looking planning tool for biodiversity conservation under climate change.

[0025] 2. Define and identify two types of climate refuges: stepping stone climate refuges and new habitat climate refuges. This will enable researchers to spatially align these two types of refuges with existing nature reserves in order to develop climate change-adaptive reserve plans. Attached Figure Description

[0026] Figure 1 This diagram illustrates the steps involved in identifying regional climate refuges based on climate change and species distribution models.

[0027] Figure 2 Spatial distribution maps of forward and backward biovelocities of two taxa under the SSP2-4.5 climate scenario (unit: km / year).

[0028] Figure 3 Spatial distribution maps of forward and backward biovelocities of two taxa under the SSP5-8.5 climate scenario (unit: km / year).

[0029] Figure 4 This is a spatial distribution map of potential climate refuges for two taxa under the SSP2-4.5 climate scenario.

[0030] Figure 5 This is a spatial distribution map of potential climate refuges for two taxa under the SSP5-8.5 climate scenario. Detailed Implementation

[0031] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0032] Example 1:

[0033] A method for identifying regional climate refuges based on climate change and species distribution models, such as Figure 1 As shown, it includes the following steps:

[0034] Step S1: Species Distribution Modeling. First, based on the species distribution information, land use data, and elevation data of the target area, the habitat range (AOH) of the species is obtained. The land use data is from the Global Land Use / Land Cover (LULC) dataset, and the elevation data (DEM) is from the Data Center of Resources and Environmental Sciences, Chinese Academy of Sciences.

[0035] In addition to land use and elevation data, other environmental variables included 19 bioclimatic variables, the Normalized Difference Vegetation Index (NDVI), slope, aspect, and distance to the nearest water source. The bioclimatic variables were sourced from WorldClim, the NDVI from the Data Center of Resources and Environmental Sciences, Chinese Academy of Sciences, slope and aspect were extracted from elevation data in ArcGIS 10.8, and the distance to the nearest water source was calculated using the Euclidean distance tool in ArcGIS 10.8.

[0036] After collecting habitat range (AOH) data, 25 initial environmental variables were selected using the variance inflation factor (VIF) test and Pearson correlation analysis. The vifstep function from the usdm R package was used to remove variables with VIF values ​​exceeding 10. Pearson correlation analysis was then performed on the remaining variables, retaining those with correlation coefficients below 0.75. This process was repeated 10 times, ultimately selecting 10 variables with no strong correlation as environmental factors for the species distribution model. These variables included: BIO3 (isotherm), BIO8 (average temperature of the wettest quarter), BIO9 (average temperature of the driest quarter), BIO14 (precipitation in the driest month), BIO15 (precipitation seasonality), BIO18 (precipitation in the warmest quarter), land use, slope, aspect, and distance to the nearest water source.

[0037] Species occurrence records were collected from BirdLife, the Global Biodiversity Information Platform (GBIF), and the Integrated Digital Biocollection (iDigBio) website, and merged with AOH-converted point data as species distribution point information.

[0038] Species distribution point data and environmental variables are input into megaSDM for species distribution modeling.

[0039] (Reference for Variance Inflation Factor (VIF) test and Pearson correlation analysis: Naimi, B. (2015). USDM: Uncertainty analysis for species distribution models.)

[0040] Step S2: Biovelocity Calculation. Based on the species binary habitat suitability maps generated from the species distribution model for the current and future periods, the forward and backward biovelocities of the species are calculated.

[0041] Forward biovelocity is defined as the rate at which a species migrates from its current location to the nearest suitable future habitat, reflecting the speed at which a species needs to move to track its climatic niche.

[0042] Backward biovelocity is defined as the rate required to move from a future habitat to the nearest current distribution area, reflecting how easy or difficult it is for an existing species to migrate to a future habitat.

[0043] In this embodiment, the target species include mammals and birds. In the species binary habitat suitability map, 1 represents suitable habitat and 0 represents unsuitable habitat. R v.4.3.1 is used to calculate the required forward biovelocities and backward biovelocities for each species, and the biovelocities of all species belonging to the same taxa are averaged to obtain the average forward and backward biovelocities for both species groups.

[0044] Step S3: Species richness calculation. Based on the species distribution model, current and future binary habitat suitability maps are generated, and weights are assigned to each species according to its endangerment level: Vulnerable (Vu) = 2, Endangered (EN) = 4, Critically Endangered (CR) = 8.

[0045] The weighted sum of habitat suitability for all species of the same type within each 1 km² grid cell is calculated using the following formula:

[0046]

[0047] Where Si represents whether species i has a suitable (1) or unsuitable (0) habitat in this grid cell, and Wi represents the threat level of each species.

[0048] After completing the weighted calculation of habitat suitability for the two types of species in the region, a weighted species richness map was drawn based on the obtained data.

[0049] Step S4: Habitat quality calculation. Using the habitat quality module in the InVEST model, land use data, biodiversity threat factors, and sensitivity factors are integrated to generate a habitat quality map of the target area.

[0050] Current land use and land use change data are derived from statistics compiled by the Data Center for Resources and Environmental Sciences, Chinese Academy of Sciences. Biodiversity threats include paddy fields, dry land, urban land, rural settlements, other construction land, and unused land.

[0051] The threat and sensitivity parameters in the model were determined based on data from previous studies in temperate and subtropical regions.

[0052] Step S5: Climate Stability Calculation. Principal Component Analysis (PCA) is performed on the current and future climate datasets for dimensionality reduction. First, 19 bioclimatic variables are statistically obtained and converted into independent principal component (PC) axes. The first four principal components are selected based on their cumulative explanation of more than 90% of the total variability of the climate variables.

[0053] The difference between current and future climate conditions is quantified using standardized Euclidean distance, calculated as follows:

[0054]

[0055] Where xi and yi represent the current and future climate values ​​of the i-th principal component axis, respectively.

[0056] A higher CCI value indicates greater climate change intensity and greater climate instability.

[0057] Finally, the calculated CCI values ​​are normalized to 0-1 and inverted to obtain the climate stability map.

[0058] Step S6: Calculate the connectivity index and reachability index.

[0059] Connectivity index CI calculation:

[0060] Forward biovelocity reflects the rate of migration required by a species to track its climatic niche; therefore, a lower forward biovelocity indicates greater connectivity between current distribution and future habitat.

[0061] The connectivity index CI is calculated by combining the forward biovelocity and average dispersal rate of a species, using the following formula:

[0062] Where V for V represents the forward velocity of a living organism. mean This represents the average dispersal rate for each species. For mammals and birds, a threshold of 1.5 km / year is used. The parameter 0.5 indicates that when the forward biovelocity is the same as the average dispersal rate, the probability of connectivity between the current distribution and future habitat is 0.5.

[0063] Accessibility Index AI Calculation:

[0064] Backward biovelocity reflects how easy it is for existing species to migrate to future habitats; therefore, the lower the backward biovelocity, the higher the accessibility of future habitats.

[0065] First, regions where new species have appeared are extracted based on a species binary habitat suitability map. Then, the accessibility index AI is calculated by combining the species' backward biovelocity and average dispersal rate, using the following formula:

[0066] Where V back V represents the backward velocity of a living organism. mean This represents the average dispersal rate of each species.

[0067] Ultimately, the connectivity index CI and accessibility index AI for the two species were obtained.

[0068] Step S7: Multiply the weighted species richness, habitat quality and climate stability grids normalized to 0-1 to obtain the habitat sustainability index (HPI).

[0069] Define two types of climate shelters:

[0070] 1. Stepping stone climate refuge. This refers to transitional habitats that provide temporary suitable habitats during the migration process of species adapting to climate change.

[0071] 2. New habitat-type climate refuges. These are suitable habitats newly formed due to climate change that can provide long-term survival space for species.

[0072] The potential index for stepping stone type climate refuges is CRPiss = HPIj × CIj.

[0073] Where HPIj represents the habitat sustainability index of species j, and CIj represents the connectivity index CI of species j.

[0074] The potential index for new habitat-type climate refuges is CRPInh = HPIj × AIj.

[0075] HPIj represents the habitat sustainability index of species j, and AIj represents the accessibility index AI of species j.

[0076] The CRPI data was calculated, and the top 25 percentile regions in the Climate Shelter Potential Index map were extracted as the distribution areas of potential climate shelter hotspots for the two types of shelters.

[0077] Example 2:

[0078] Application of a regional climate refuge identification method based on climate change and species distribution models. This method was implemented in the middle and lower reaches of the Yangtze River in China, including parts of Hubei, Hunan, and Jiangxi provinces, as well as Jiangsu, Zhejiang, Shanghai, Anhui, Shaanxi, and Henan provinces (municipalities), covering a total area of ​​approximately 780,800 square kilometers.

[0079] The average and standard deviation of the biovelocities of two species in the middle and lower reaches of the Yangtze River in China are shown in Table 1.

[0080] Climate change scenarios: SSP2-4.5 and SSP5-8.5.

[0081] Table 1

[0082]

[0083] For details on the forward and backward biovelocities of two types of species in the middle and lower reaches of the Yangtze River in China, please see [link to relevant documentation]. Figure 2 and Figure 3 .

[0084] Under SSP2-4.5 and SSP5-8.5 climate scenarios, the distribution of the Climate Refuge Potential Index (CRPI) for two species in the middle and lower reaches of the Yangtze River was simulated up to the 2090s (2081-2100). Details are as follows: Figure 4 and Figure 5 As shown in the figure, the higher the CRPI value of a region, the greater its potential as a climate refuge. Potential climate refuges are mainly distributed in the western and southeastern parts of the middle and lower reaches of the Yangtze River. Specifically, under the SSP2-4.5 scenario, the distribution patterns of stepping stone climate refuges for mammals and birds are highly similar, with hotspots mainly concentrated in three areas: western and northwestern Hunan Province, the southern mountainous area bordering Hunan and Jiangxi Provinces, and southeastern Anhui Province; while birds also show high potential in areas near water sources. New habitat refuge hotspots show significant heterogeneity between the two species: mammal hotspots are mainly concentrated in the border area between Hubei and Hunan Provinces, but are almost non-existent in the southeastern part of the middle and lower reaches of the Yangtze River; bird hotspots are mainly distributed in southeastern Jiangxi Province and western Hubei Province.

[0085] The spatial pattern of refuge hotspots under the SSP5-8.5 climate scenario is similar to that under the SSP2-4.5 scenario, but the distribution range of mammalian stepping stone refuge hotspots is significantly reduced in southern and southeastern Jiangxi Province.

[0086] The top 25 percentile regions were extracted from the climate refuge potential index map of the middle and lower reaches of the Yangtze River as refuge hotspots. The distribution areas of the two types of potential climate refuge hotspots were then overlaid with existing protected areas to obtain the spatial distribution of protected and unprotected climate refuge hotspots.

[0087] Under the SSP2-4.5 scenario, 32.3% of protected areas can serve as stepping stone climate refuges, 25.6% can serve as new habitat-type climate refuges, and 21.6% of protected areas possess both of these refuge functions.

[0088] Under the SSP5-8.5 scenario, 35.5% of protected areas can serve as stepping stone climate refuges, 27.5% can serve as new habitat-type climate refuges, and 22.7% of protected areas possess both of these refuge functions.

[0089] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for identifying regional climate refuges based on a climate change and species distribution model, characterized in that, Includes the following steps: Step S1: Species distribution modeling. First, based on the species distribution information, land use data and elevation data of the target area, the habitat range (AOH) of the species is selected, and species distribution modeling is carried out in combination with environmental variables. Step S2: Biovelocity Calculation. Based on the species binary habitat suitability map generated by the species distribution model for the current and future periods, the forward biovelocity and backward biovelocity of the species are calculated. Step S3: Species richness calculation. Based on the species distribution model, current and future binary habitat suitability maps are generated, and weights are assigned to each species according to its endangerment level. Then, for each 1km² area... 2 In the grid cells, the habitat suitability weighted sum of all species of the same group is calculated to obtain a weighted species richness map; Step S4: Habitat quality calculation, integrating land use data, biodiversity threat factors and sensitivity factors to generate a habitat quality map of the target area; Step S5: Climate stability calculation. Principal component analysis (PCA) is performed on the current and future climate datasets to reduce dimensionality. The climate change intensity value (CCI) is calculated using Euclidean distance. A climate stability map is obtained based on the CCI. Step S6: Calculation of connectivity index and accessibility index. The connectivity index CI is obtained based on the forward biovelocity and average dispersal rate of the species; the accessibility index AI is obtained based on the backward biovelocity and average dispersal rate of the species. Step S7: Identification of potential climate refuges. The weighted species richness, habitat quality, and climate stability data are normalized and multiplied to obtain the Habitat Sustainability Index (HPI). Based on the HPI and the Connectivity Index (CI), the Stepping Stone Climate Refuge Potential Index (CRPI) is calculated. ss Based on the Habitat Sustainability Index (HPI) and Accessibility Index (AI), the New Habitat Type Climate Refuge Potential Index (CRPI) was obtained. nh Based on the derived CRPI ss and CRPI nh To identify the distribution areas of climate refuge hotspots.

2. The method for identifying regional climate refuges based on climate change and species distribution models according to claim 1, characterized in that: In step S1, after collecting habitat range AOH and converting it into point data, 10 variables with no strong correlation were selected from 25 initial environmental variables through variance inflation factor (VIF) test and Pearson correlation analysis as environmental factors for species distribution model. Speciation records were added as a supplement to the above species distribution points. Both were input into megaSDM for species distribution modeling.

3. The method for identifying regional climate refuges based on climate change and species distribution models according to claim 1, characterized in that: In step S2, the forward and backward biovelocities required for each species are first calculated, and then the rate results of all species belonging to the same group are averaged to obtain the average forward and backward biovelocities for each species.

4. The method for identifying regional climate refuges based on climate change and species distribution models according to claim 1, characterized in that: In step S3, the endangered status weights are assigned as follows: Vulnerable (Vu) = 2, Endangered (EN) = 4, Critically Endangered (CR) = 8; the species habitat suitability weighted sum is calculated using the following formula: Where S i W represents whether species i has a suitable (1) or unsuitable (0) habitat in this grid cell. i This indicates the threat level for each species.

5. The method for identifying regional climate refuges based on climate change and species distribution models according to claim 1, characterized in that: In step S4, biodiversity threat factors include paddy fields, dry land, urban land, rural settlements, other construction land, and unused land.

6. The method for identifying regional climate refuges based on climate change and species distribution models according to claim 1, characterized in that: In step S5, several current and future bioclimatic variables are transformed into independent principal component (PC) axes. The CCI is calculated by selecting the first four principal components, using the following formula: Where x i and y i These represent the current and future climate values ​​of the i-th principal component axis, respectively.

7. The method for identifying regional climate refuges based on climate change and species distribution models according to claim 1, characterized in that: In step S6, Where V for V represents the forward velocity of a living organism. mean This represents the average dispersal rate of each species.

8. The method for identifying regional climate refuges based on climate change and species distribution models according to claim 1, characterized in that: In step S6, Where V back V represents the backward velocity of a living organism. mean This represents the average dispersal rate of each species.

9. The method for identifying regional climate refuges based on climate change and species distribution models according to claim 1, characterized in that: In step S7, the stepping stone climate refuge potential index CRPiss = HPI j ×CI j HPI j CI represents the habitat sustainability index for species class j. j The connectivity index CI represents species class j; the new habitat-type climate refuge potential index CRPI represents the potential of new habitat-type climate refuges. nh =HPI j ×AI j HPI j AI represents the habitat sustainability index for species class j. j AI represents the accessibility index of species class j.

10. The method for identifying regional climate refuges based on climate change and species distribution models according to claim 1, characterized in that: In step S7, the top 25 percentile regions in the two types of climate refuge potential index maps are extracted as climate refuge hotspots and compared with protected areas to reveal protection gaps in the existing protected area network.

Images