A method for studying the influence of landscape pattern and climate change on future water quality in a watershed

Abstract

This invention discloses a research method for landscape patterns and the impact of climate change on future water quality in watersheds, belonging to the field of climate change's impact on the ecological environment. It addresses the uncertainty in selecting the optimal research scale and relevant indicators when analyzing the impact of climate and landscape pattern changes on future watershed water quality. The invention utilizes DEM data and land use data to delineate the study area boundaries and research scale. It extracts surface landscape type information using unsupervised classification, visual interpretation, and field verification methods, constructs a predictive model for future watershed water quality, and obtains the characteristics of future watershed water quality under climate and landscape pattern changes. This invention is suitable for watershed studies at various scales, selecting the optimal research scale; it overcomes high redundancy among indicators, helps optimize the "source and sink" landscape pattern for reducing non-point source pollution within the watershed, and is of great significance for ecological assessment and non-point source pollution control within watersheds under climate change conditions; it has wide applications.

Description

Technical Field

[0001] This invention belongs to the field of the impact of climate change on the ecological environment, and specifically relates to a research method for landscape patterns and the impact of climate change on the future water quality of watersheds. Background Technology

[0002] Climate change is a globally recognized fact, impacting various spheres of the Earth system, including rocks, atmosphere, water, and biota. A watershed is a geographical unit connecting upstream, midstream, and downstream areas via water. The water cycle is heavily constrained by the watershed's landscape pattern and inevitably affected by local climate change, leading to alterations in both the quality and quantity of water resources. For example, considering the ecological processes of non-point source pollution, pollutants accumulate in water bodies through different source and sink landscape types. The source and sink composition and spatial structure of the landscape pattern inevitably influence the transport of non-point source pollutants, thus affecting water quality. Furthermore, the uneven spatial and temporal distribution of future climate factors such as increased temperatures and extreme precipitation will exacerbate eutrophication and other water problems. Therefore, when analyzing the spatiotemporal evolution of water quality based on the watershed's landscape pattern, it is crucial to pay close attention to the impact of non-stationary factors such as climate change. This is of significant practical importance for understanding the watershed's water environment characteristics, non-point source pollution processes, and comprehensive control and management. The method of dividing the watershed into minimum hydrological response units based on digital elevation models (DEMs) and land use data can determine the spatial scale of the study. The spatial heterogeneity of the watershed determines the complexity of landscape components and landscape structure. Unsupervised classification, visual interpretation, and field verification methods can maximize the accuracy of extracting key elements such as watershed landscape types and patterns. The prediction of future watershed landscape patterns and the statistical downscaling of future climate elements can provide high-precision underlying surface information and climate information to accurately grasp the characteristics of future water quality within the watershed. Under the socio-economic development conditions where agricultural pollutant emissions are under pressure from intensive agricultural development and it is difficult to achieve reduction, the watershed needs quantitative scientific theories and methods to support the impact of non-point source pollution landscape composition types, landscape spatial structure, and their configuration on water quality. However, in existing landscape pattern studies on water quality, the relationship between landscape pattern indices and ecological processes is not very rigorous, so the calculated results of landscape pattern indices are not ecologically meaningful. Fundamentally, the calculation results of traditional landscape pattern indices do not reflect spatial variability. Furthermore, existing studies often overlook the potential impacts of climate change on water quality, as well as the scale effects in the relationship between landscape patterns and water quality indicators. The influence of climate conditions and landscape patterns on certain water quality indicators can vary significantly across different research scales. Therefore, research findings on the impact of climate conditions and landscape patterns on water quality inevitably face a series of problems. Summary of the Invention

[0003] The purpose of this invention is to propose a research method for the impact of landscape patterns and climate change on the future water quality of a watershed, characterized by the following steps:

[0004] Step 1: The spatial research scale of the watershed where the study area is located is divided, including the watershed boundary, the upper, middle and lower reaches of the watershed, and several typical small watersheds;

[0005] Step 2: Unsupervised classification, visual interpretation, and field verification methods are used to extract watershed landscape type and landscape pattern information;

[0006] Step 3: Use the stepwise clustering downscaling (SCD) method to obtain high-precision climate indicators for future temperature, precipitation, and evapotranspiration.

[0007] Step 4: Use the FLUS model to obtain information on the future landscape types, landscape components, and landscape patterns within the watershed;

[0008] Step 5: Use the watershed future water quality prediction model (SCA) to obtain water quality index information for the future watershed;

[0009] Step 6: Study water quality characteristics under the conditions of climate and landscape pattern changes, and study the impact of climate and landscape pattern changes on future water quality from multiple temporal and spatial perspectives.

[0010] Step 1 includes:

[0011] (1) Use the Fill function in ArcSWAT 10.3 to fill depressions in the DEM data;

[0012] (2) Calculate the direction of water flow and the amount of water accumulation to determine the outlet point of the small watershed. The scope of the small watershed is all grid cells upstream of the watershed outlet.

[0013] (3) By trying different runoff accumulation rates for river network extraction, the smaller the threshold area, the denser the river network, the larger the number of small watersheds, and the more spatial variations in watershed topography are considered when extracting small watersheds. By comparing with the river system map, the threshold is finally determined to ensure that the river network is as similar as possible to the actual situation;

[0014] (4) Use the Watershed tool to extract the scope of the study watershed, and combine Google satellite maps and the topographic features of the watershed where the research is located to determine the boundaries of the study area and the boundaries of each sub-watershed;

[0015] (5) Several typical small watersheds are finally generated and named in order from upstream to downstream;

[0016] (6) Based on the typical small watershed division, the watershed is divided into three parts: upper, middle and lower reaches, according to the topography, climate and hydrology, vegetation and soil, elevation characteristics and the altitude of the watershed outlet in the study area.

[0017] Step 2 includes:

[0018] (1) Import the relevant high-resolution remote sensing images into ArcGIS, add and calibrate the geospatial reference coordinate system, and perform image stitching to obtain the initial multispectral image data;

[0019] (2) The mosaic image was cropped using the boundary vector data of the study area to obtain the original remote sensing image of the study area. The image was then interpreted using a combination of unsupervised classification and visual interpretation, as well as field verification.

[0020] (3) Based on the research objectives, field survey data of the research watershed, digital elevation model (DEM) data and the "Classification of Current Land Use", a preliminary landscape classification system for the research watershed is constructed. Then, based on the landscape type characteristics of the research area and the needs of the research data, the data is merged through cluster analysis, and finally the logically ordered classification standard for the landscape type of the research watershed is determined.

[0021] (4) In order to improve the interpretation accuracy and classification accuracy, verification points were selected for each type of landscape on the image, and the GPS fixed point information collected in the field was used for investigation, verification and correction to obtain the overall interpretation accuracy. The Kappa coefficient was verified to make it meet the requirements of the research.

[0022] (5) In ArcGIS 10.3, process and analyze vector data, extract different landscape information according to the research content, and output the landscape type map of the study area by adding legends, north arrows, important cities and latitude and longitude grids and defining scale.

[0023] Step 3 includes:

[0024] (1) Select large-scale forecast factors, download and screen climate change variables from GCM (Global Climate Model) and reanalysis datasets including GFDL-CM2.0, Had CM3 and NCARCCSM3, and perform default value and dimensionality tests on the input and output datasets; construct the matrix of input and output variables;

[0025] (2) Based on the classification principles and merging criteria, all samples are classified into the corresponding categories;

[0026] (3) Establishment of a stepwise clustering tree, that is, to establish the statistical relationship between large-scale climate forecasting factors and regional climate forecast quantities;

[0027] (4) The effectiveness of the stepwise clustering downscaling model (SCD) was calibrated and validated;

[0028] (5) Based on the stepwise clustering tree, high-precision output variable values ​​are generated according to the input variables, including the simulation prediction results of temperature, precipitation and evapotranspiration;

[0029] Step 4 includes:

[0030] (1) Based on the actual situation of the watershed and the availability of data, select the driving factors of landscape pattern change, including topographic factors of elevation and slope, traffic accessibility factors and limiting transformation factors.

[0031] (2) Standardize the driving factors of land use change and calculate the suitability probability map of landscape component types in each pixel;

[0032] (3) By using the Markov prediction model, the target for the number of changes in each landscape component type is preset, and the ease or difficulty of conversion between different landscape types and the restricted areas for landscape type conversion are determined.

[0033] (4) Set model parameters, including the number of simulation iterations and the size of the domain, to ultimately simulate the changes in the watershed landscape pattern;

[0034] Step 5 includes:

[0035] (1) Based on historical climate data and landscape pattern data of temperature, precipitation and evapotranspiration, a comprehensive and low-redundancy forecast index system closely related to water quality indicators was selected.

[0036] (2) Establish the statistical relationship between the forecast indicator system and water quality indicators;

[0037] (3) Use historical observation data to calibrate and validate the future water quality prediction model (SCA);

[0038] (4) Based on the water quality prediction model, generate future water quality index data, including the simulation prediction results of ammonia nitrogen, total phosphorus and chemical oxygen demand.

[0039] Step 6 includes:

[0040] (1) Research on future climate change in the basin, including the characteristics of future climate change in the basin, including temperature, precipitation and evapotranspiration, as well as the characteristics of extreme climate events such as heat waves and rainstorms.

[0041] (2) Selection of multi-temporal and spatial research scales: Based on the landscape type, landscape pattern and hydrological period characteristics of the study area, representative time scales such as dry season, normal water period and wet season and spatial scales including typical small watersheds, upper, middle and lower reaches of the watershed and the whole watershed are selected as the research scales for landscape patterns.

[0042] (3) Analysis of the relationship between the forecast index system of climate and landscape pattern and water quality index, which is used to study the impact of climate and landscape pattern changes on future water quality at multiple time and space scales.

[0043] The beneficial effects of this invention are as follows: This invention considers the scale effect of ecological processes, studying the impact of climate change and landscape patterns on water quality at different spatiotemporal scales. Addressing the significant spatiotemporal variability of non-point source pollution, it improves the method for determining the research scale, making it suitable for watershed studies at various scales. The construction of climate indicators, landscape pattern indices, and water quality indicators resolves the issue of high redundancy among indicators. This invention helps optimize the "source-sink" landscape patterns for reducing non-point source pollution within watersheds and facilitates the selection of the optimal research scale. It also has significant implications for ecological assessment and non-point source pollution control within watersheds under climate change conditions. It can be widely applied to geographic information systems and remote sensing technology in the study of ecological processes related to non-point source pollution in watersheds. Attached Figure Description

[0044] Figure 1 Extracting the boundary of a watershed and dividing it into sub-watersheds; where (a) is an elevation map; (b) is the watershed river network; (c) is the watershed boundary; and (d) is a typical sub-watershed.

[0045] Figure 2 This is a map showing the distribution of water resources in the upper, middle, and lower reaches of a certain river basin.

[0046] Figure 3 A summary of images from the preprocessing stage of remote sensing data for a watershed, including (a) the cropped watershed image; (b) the interpreted watershed land use and cover map; and (c) the edited watershed landscape type map. Detailed Implementation

[0047] This invention proposes a research method for the impact of landscape patterns and climate change on the future water quality of a watershed, comprising the following steps:

[0048] Step 1, spatial research scale division of the watershed where the study area is located, including (a) elevation map, (b) watershed river network, (c) watershed boundary, and (d) typical small watershed;

[0049] Step 2: Unsupervised classification, visual interpretation, and field verification methods are used to extract watershed landscape type and landscape pattern information;

[0050] Step 3: Use the stepwise clustering downscaling (SCD) method to obtain high-precision climate indicators for future temperature, precipitation, and evapotranspiration.

[0051] Step 4: Use the FLUS (Future Land Use Simulation) model to simulate and obtain information on the future landscape types, landscape components, and landscape patterns within the watershed;

[0052] Step 5: Use the watershed future water quality prediction model (SCA) to obtain water quality index information for the future watershed;

[0053] Step 6: Study water quality characteristics under the conditions of climate and landscape pattern changes, and study the impact of climate and landscape pattern changes on future water quality from multiple temporal and spatial perspectives.

[0054] This study addresses the uncertainty surrounding the optimal research scale and relevant indicator selection when analyzing the impacts of climate and landscape pattern changes on future watershed water quality. It utilizes DEM and land use data to delineate the study area boundaries and research scale. Unsupervised classification, visual interpretation, and field verification methods are employed to extract surface landscape type information. Based on historical climate and landscape pattern data, high-precision future climate data and landscape pattern information for the watershed are obtained. By calculating the correlation coefficients between historical landscape pattern indices, climate indicators, and water quality indicators, a predictive model for future water quality in the watershed is constructed. Future water quality data is then obtained to study the future water quality characteristics of the watershed under conditions of climate and landscape pattern changes.

[0055] Example 1

[0056] A research method for studying the impacts of landscape patterns and climate change on the future water quality of a watershed includes the following steps:

[0057] Step 1, as follows Figure 1 The example described uses a watershed containing a secondary tributary. The spatial research scale of this watershed is divided, including the watershed boundaries (e.g.,...). Figure 1 As shown in c), several typical small watersheds (such as...) Figure 1 (as shown in d) and the upper, middle and lower reaches of the basin (such as Figure 2 The division shown in the figure;

[0058] Specifically, using the watershed spatial research scale division method described in this invention may include the following steps:

[0059] (1) Use the Fill function in ArcSWAT 10.3 to perform operations such as... Figure 1 Fill the depressions in the DEM data shown in a;

[0060] (2) Calculate the direction of water flow and the amount of water accumulation to determine the outlet point of the small watershed. The scope of the small watershed is all grid cells upstream of the watershed outlet.

[0061] (3) By experimenting with different runoff accumulation rates for river network extraction, a smaller threshold area results in a denser river network and a larger number of small watersheds. This requires greater consideration of spatial variations in watershed topography when extracting small watersheds. The threshold is ultimately determined by comparing with a river system map to ensure the river network most closely approximates the actual situation. Figure 1 As shown in b;

[0062] (4) Using the Watershed tool, the extent of the study watershed was extracted. Combined with Google satellite maps and the topographic features of the watershed, the boundaries of the study area and the boundaries of each sub-watershed were determined, such as... Figure 1 cd shown;

[0063] (5) Several typical small watersheds are finally generated, such as Figure 1 As shown in d, they are named in order from upstream to downstream;

[0064] (6) Based on the typical small watershed division, and according to the topography, climate, hydrology, vegetation, soil, elevation characteristics, and the altitude of the watershed outlet in the study area, the watershed is divided into three parts: upper, middle, and lower reaches. Figure 2 As shown;

[0065] Step 2, the unsupervised classification, visual interpretation and field fixed-point verification method is used to extract watershed landscape type and landscape pattern information;

[0066] Specifically, the method for extracting watershed landscape type and landscape pattern information according to the present invention may include the following steps:

[0067] (1) Import the relevant high-resolution remote sensing images into ArcGIS, add and calibrate the geospatial reference coordinate system, and perform image stitching to obtain the initial multispectral image data;

[0068] (2) Use as follows Figure 1 The boundary vector data of the study area described in c is used to crop the stitched image to obtain, as shown in the figure. Figure 3 The original remote sensing image of the study watershed shown in image a was interpreted using a combination of unsupervised classification and visual interpretation, along with indoor reading and field verification. The interpreted remote sensing image is shown below. Figure 3 As shown in b;

[0069] (3) Based on the research objectives, field survey data of the research watershed, DEM data and the Classification of Current Land Use (GB / T21010-2018), a preliminary landscape classification system for the research watershed is constructed. Then, according to the landscape type characteristics of the research area and the needs of the research data, the classification is merged through cluster analysis, and finally the logically ordered classification standard for landscape types of the research watershed is determined, including: cultivated land, forest land, grassland, river and canal, reservoir and pond, construction land and unused land.

[0070] (4) In order to improve the interpretation accuracy and classification accuracy, verification points were selected for each type of landscape on the image, and the GPS fixed point information collected in the field was used for investigation, verification and correction to obtain the overall interpretation accuracy. The Kappa coefficient was verified to make it meet the requirements of the research.

[0071] (5) Process and analyze vector data in ArcGIS 10.3, extract different landscape information according to the research content, and output the data by adding legends, north arrows, important cities, latitude and longitude grids, and defining scales, etc. Figure 3 The landscape type map of the study area shown in c includes the following main landscape types: cultivated land, forest land, grassland, rivers and canals, reservoirs and ponds, construction land, and unused land.

[0072] Step 3, the stepwise clustering downscaling method (SCD) is used to obtain high-precision climate indicators such as future temperature, precipitation and evapotranspiration;

[0073] Specifically, using the stepwise clustering downscaling method described in this invention may include the following steps:

[0074] (1) Select large-scale forecast factors, download and screen climate change variables from GCM (Global Climate Model) and reanalysis datasets including GFDL-CM2.0, Had CM3 and NCARCCSM3, and perform default value and dimensionality tests on the input and output datasets; construct the matrix of input and output variables;

[0075] (2) Based on the classification principles and merging criteria, all samples are classified into the corresponding categories;

[0076] (3) Establishment of a stepwise clustering tree, that is, to establish the statistical relationship between large-scale climate forecasting factors and regional climate forecast quantities;

[0077] (4) The effectiveness of the stepwise clustering downscaling model (SCD) was calibrated and validated;

[0078] (5) Based on the stepwise clustering tree, high-precision output variable values ​​are generated according to the input variables, including the simulation prediction results of temperature, precipitation and evapotranspiration; Step 4, the FLUS model is used to obtain the future landscape pattern information of the watershed, including the landscape type, landscape components and landscape pattern information in the watershed.

[0079] Specifically, the method for obtaining future landscape pattern information of a watershed using the present invention may include the following steps:

[0080] (1) Based on the actual situation of the watershed and the availability of data, select the driving factors of landscape pattern change, including topographic factors of elevation and slope, traffic accessibility factors and limiting transformation factors.

[0081] (2) Standardize the driving factors of land use change and calculate the suitability probability map of landscape component types in each pixel;

[0082] (3) By using the Markov prediction model, the target for the number of changes in each landscape component type is preset, and the ease or difficulty of conversion between different landscape types and the restricted areas for landscape type conversion are determined.

[0083] (4) Set model parameters, including the number of simulation iterations and the size of the domain, to ultimately simulate the changes in the watershed landscape pattern;

[0084] Step 5: The predictive model (SCA) for future water quality in the basin is used to obtain information on future water quality indicators in the basin.

[0085] Specifically, using the watershed future water quality prediction model described in this invention may include the following steps:

[0086] (1) Based on historical climate data and landscape pattern data of temperature, precipitation and evapotranspiration, a comprehensive and low-redundancy forecast index system closely related to water quality indicators was selected.

[0087] (2) Establish the statistical relationship between the forecast indicator system and water quality indicators;

[0088] (3) Use historical observation data to calibrate and validate the prediction model for future water quality;

[0089] (4) Based on the water quality prediction model, generate future water quality index data, including simulation prediction results of ammonia nitrogen, total phosphorus and chemical oxygen demand;

[0090] Step 6, the study of water quality characteristics under the conditions of climate and landscape pattern change, is used to study the impact of climate and landscape pattern change on future water quality from multiple temporal and spatial perspectives;

[0091] Specifically, the study of water quality characteristics under climate and landscape pattern change conditions described in this invention may include the following steps:

[0092] (1) Research on future climate change in the basin, including the characteristics of future climate change in the basin, including temperature, precipitation and evapotranspiration, as well as the characteristics of extreme climate events such as heat waves and rainstorms.

[0093] (2) Selection of multi-temporal and spatial research scales: Based on the landscape type, landscape pattern and hydrological period characteristics of the study area, representative time scales such as dry season, normal water period and wet season and spatial scales including typical small watersheds, upper, middle and lower reaches of the watershed and the whole watershed are selected as the research scales for landscape patterns.

[0094] (3) Analysis of the relationship between the forecast index system of climate and landscape pattern and water quality index, which is used to study the impact of climate and landscape pattern changes on future water quality at multiple time and space scales.

Claims

1. A research method for the impact of landscape patterns and climate change on the future water quality of a watershed, characterized in that, Includes the following steps: Step 1: Spatial scale division of the watershed where the study area is located, including the watershed boundary, the upper, middle and lower reaches of the watershed, and several typical small watersheds; Step 2: Unsupervised classification, visual interpretation, and field verification methods are used to extract watershed landscape type and landscape pattern information; Step 3: Use a stepwise clustering downscaling method to obtain high-precision climate indicators for future temperature, precipitation, and evapotranspiration. Step 3 includes: (1) Select large-scale forecast factors, download and screen climate change variables from GCM and reanalysis datasets including GFDL-CM2.0, Had CM3 and NCARCCSM3, and perform default value and dimensionality tests on the input and output datasets; construct the matrix of input and output variables; (2) Based on the classification principles and merging criteria, all samples are assigned to the corresponding categories; (3) Establishment of a stepwise clustering tree, that is, to establish the statistical relationship between large-scale climate forecasting factors and regional climate forecast quantities; (4) The effectiveness of the stepwise clustering downscaling model was calibrated and verified; (5) Based on the stepwise clustering tree, high-precision output variable values ​​are generated according to the input variables, including the simulation prediction results of temperature, precipitation and evapotranspiration; Step 4: Use the FLUS model to obtain information on the future landscape types, landscape components, and landscape patterns within the watershed; Step 4 includes: (1) Based on the actual situation of the watershed and the availability of data, select the driving factors of landscape pattern change, including topographic factors of elevation and slope, traffic accessibility factors and limiting transformation factors; (2) Standardize the driving factors of land use change and calculate the suitability probability map of landscape component types in each pixel; (3) By using the Markov prediction model to pre-set the target for the number of changes in each landscape component type, the ease of conversion between different landscape types and the restricted areas for landscape type conversion are determined; (4) Set model parameters, including the number of simulation iterations and the size of the domain, to ultimately simulate the changes in the watershed landscape pattern; Step 5: Use the watershed future water quality prediction model to obtain water quality index information for the future watershed. Step 5 includes: (1) Based on historical climate data and landscape pattern data of temperature, precipitation and evapotranspiration, a comprehensive and low-redundancy forecast index system closely related to water quality indicators was selected. (2) Establish the statistical relationship between the forecast indicator system and water quality indicators; (3) Use historical observation data to calibrate and validate the prediction model for future water quality; (4) Based on the water quality prediction model, generate future water quality index data, including the simulation prediction results of ammonia nitrogen, total phosphorus and chemical oxygen demand; Step 6: Study water quality characteristics under the conditions of climate and landscape pattern changes, and study the impact of climate and landscape pattern changes on future water quality from multiple temporal and spatial perspectives.

2. The research method for the impact of landscape pattern and climate change on the future water quality of the watershed according to claim 1, characterized in that, Step 1 includes: (1) Use the Fill function in ArcSWAT 10.3 to fill depressions in the DEM data; (2) Calculate the direction of water flow and the amount of water accumulation to determine the outlet point of the small watershed. The scope of the small watershed is all grid cells upstream of the watershed outlet. (3) By trying different flow accumulation amounts to extract the river network, the smaller the threshold area, the denser the river network, the larger the number of small watersheds, and the more spatial changes in the topography of the watershed are considered when extracting small watersheds; by comparing with the river system map, the threshold is finally determined to ensure that the river network is as similar as possible to the actual situation. (4) Use the Watershed tool to extract the scope of the study watershed, and combine Google satellite maps and the topographic features of the watershed where the research is located to determine the boundaries of the study area and the boundaries of each sub-watershed; (5) Several typical small watersheds are finally generated and named in order from upstream to downstream; (6) Based on the typical small watershed division, the watershed is divided into three parts: upper, middle and lower reaches, according to the topography, climate and hydrology, vegetation and soil, elevation characteristics and the altitude of the watershed outlet in the study area.

3. The research method for the impact of landscape pattern and climate change on the future water quality of the watershed according to claim 1, characterized in that, Step 2 includes: (1) Import the relevant high-resolution remote sensing images into ArcGIS, add and calibrate the geospatial reference coordinate system, and perform image stitching to obtain the initial multispectral image data; (2) The mosaic image was cropped using the boundary vector data of the study area to obtain the original remote sensing image of the study area. The image was then interpreted using a combination of unsupervised classification and visual interpretation, as well as field verification. (3) Based on the research objectives, field survey data of the research watershed, digital elevation model data and land use status classification methods, a preliminary landscape classification system for the research watershed is constructed. Then, based on the landscape type characteristics of the research area and the needs of the research data, cluster analysis is used to merge the data and finally determine the logically ordered landscape type classification standard for the research watershed. (4) In order to improve the interpretation accuracy and classification accuracy, verification points were selected for each type of landscape on the image, and the GPS fixed point information collected in the field was used for investigation, verification and correction to obtain the overall interpretation accuracy. The Kappa coefficient was verified to make it meet the requirements of the research. (5) Process and analyze vector data in ArcGIS 10.3, extract different landscape information according to the research content, and output the landscape type map of the study area by adding legends, north arrows, important cities and latitude and longitude grids and defining scale.

4. The research method for the impact of landscape pattern and climate change on the future water quality of the watershed according to claim 1, characterized in that, Step 6 includes: (1) Research on future climate change in the basin, including the characteristics of future climate change in temperature, precipitation and evapotranspiration, as well as the characteristics of extreme climate events such as heat waves and rainstorms. (2) Selection of multi-temporal and spatial research scales: Based on the landscape type, landscape pattern and hydrological period characteristics of the study area, representative time scales of dry season, normal water season and wet season and spatial scales including typical small watersheds, upper, middle and lower reaches of the watershed and the whole watershed are selected as the research scales of landscape pattern. (3) Analysis of the relationship between the forecast index system of climate and landscape pattern and water quality index, which is used to study the impact of climate and landscape pattern changes on future water quality at multiple time and space scales.

Images