Zoning optimization of ecosystem service value and landscape ecological risk in typical karst areas in southwest China

Abstract

Scientifically evaluating the current ecological environment situation in karst areas and constructing a reasonable ecological zoning system are of great significance for promoting the rational allocation of regional land resources, maintaining ecological security, and driving sustainable development. Based on land use data from 1973 to 2020, the temporal and spatial dynamic evolution of the ecosystem service value (ESV) and landscape ecological risk (LER) were quantitatively analyzed using an ecosystem service model and ecological risk model, and four types of ecological zones were constructed using the Z-score standardization method. The findings indicated that (1) The land use types in Puding County were mainly shrub land (330.61 km²) and dry land (277.34 km²). During the research period, the area of construction land increased most significantly due to the conversion from dry land and paddy fields, with an increased area of 96.55 km². The overall landscape fragmentation and heterogeneity increased, and the landscape diversity remained at a relatively high level. (2) The ecosystem services in Puding County were predominantly regulating services, and the ESV showed a fluctuating trend of decreasing first and then increasing, with a small overall change of 15.11%. The value supplied by shrub land was the highest, accounting for 24.85% of the total value of the study area. The overall ESV showed a spatial allocation pattern of “high in the northeast and low in the southwest”. (3) The LER in the study area is mostly low, showing a pattern of being “high in the middle and low around”. Accounting for approximately 64% of the total area, the lower and moderate ecological risk zones were the most prevalent. (4) Based on the ESVs and LERs, the research area has been divided into four ecological zones: ecological risk improvement zone, ecological comprehensive restoration zone, ecological function enhancement zone, and ecological conservation and maintenance zone. The different ecological zones had a weak spatial distribution continuity, and the low ESV - low LER zone (III) was the most common. The research results of this study aim to provide an important reference basis for the optimization and scientific management of the ecological environment in karst areas.

Introduction

Ecosystems are continuously exposed to a variety of risks when delivering services due to rapid urbanization and changes in land use and cover, among other factors¹. As an important manifestation of regional sustainable development, the ecosystem service value (ESV) refers to various benefits that humans directly or indirectly obtain from ecosystems^2,3. Against the backdrop of the continuous stress exerted on ecosystem functions by the rapid urbanization process and intense human activities, there exists a spatial coupling relationship between ESV and the regional land use pattern⁴. The methods for evaluating ESV mainly include the functional value method and the equivalent factor method based on land use. Among them, the equivalent factor method can achieve a comprehensive quantitative evaluation of multiple ecosystem service functions by constructing a standardized system of ecological value coefficients, and it has strong operability. Therefore, it has been widely applied in the research field of ecosystem services. In current ESV research, on the one hand, the focus is on using various models and methods to accurately quantify the dynamic changes of ESV in different regions within a specific time period, and to analyze the evolution characteristics of ESV from the spatiotemporal dimension⁵. On the other hand, it delves into multiple factors influencing the changes in ESV, covering natural factors such as climate and terrain, as well as human factors such as land use changes, the intensity of human activities, and the level of economic development. By means of geographical detectors, ecological contribution models and other tools, it reveals the action mechanisms of these factors^6,7.

Landscape Ecological Risk (LER), as an important branch of ecological risk assessment, refers to the possibility that the structure and function of the regional ecosystem are damaged and negative ecological effects are generated under the interference of natural processes or human activities^8,9,10. In the evaluation method system, the LER evaluation method based on land use change can intuitively present the health status of the regional ecosystem and its spatial heterogeneity driven by both natural and human activities¹¹. Compared with the “source-sink” ecological risk assessment framework oriented by ecological processes, the landscape index method focuses more on analyzing ecological risk from the perspective of landscape pattern, and systematically takes into account the superposition effect and cumulative impact of multiple risk sources^9,12. This method breaks through the limitations of single-element analysis and realizes a multi-dimensional and comprehensive assessment of ecological risk. In recent years, the research field of LER has shown a development trend of multi-scale and multi-method integration, with studies mostly focusing on issues such as the construction of LER quantitative evaluation models, scale effects, and driving mechanisms^8,13,14.

So far, most scholars have carried out extensive research on ESV and LER respectively, and achieved fruitful research results. However, due to the internal connection and coupling mechanism between ESV and LER, independent research on the two can no longer fully reveal the internal change characteristics of the ecosystem environment¹⁰. Although some scholars have conducted integrated research on ecosystem services closely related to human well-being and LER, most of these studies focus on analyzing the mutual relationship between the two and theoretical discussions^15,16. There is a lack of research integrating the two to form an effective analytical framework that can directly serve the formulation of management strategies and zoning decisions, resulting in research outcomes that are difficult to meet the practical needs of ecological service improvement and ecological risk management. In addition, as an important means to promote regional sustainable development and ecological security construction, ecological zoning can not only accurately identify regional protection targets and ecological issues, but also effectively achieve differentiated ecological restoration and refined management and control in different regions¹⁷. Therefore, scientific ecological zoning and differentiated management strategies are crucial for optimizing the pattern of territorial space development and protection, as well as improving the efficiency of regional ecological environment protection¹⁸. With the continuous innovation of remote sensing and GIS technologies, ecological zoning research has shown significant evolutionary characteristics. Early studies mostly focused on single-dimensional analyses of single natural elements such as vegetation types and topographical features^19,20, while current research has shifted toward constructing comprehensive evaluation systems with multi-dimensional indicator coupling, emphasizing the inclusion of interrelated elements of ecosystems²¹. The ecological zoning method based on ESV and LER, by integrating a two-dimensional evaluation framework of human well-being (positive dimension) and potential risks (negative dimension), systematically reveals the dynamic characteristics of regional ecosystem security. It not only emphasizes the protection of ecosystem integrity and the identification of uniqueness but also provides important methodological support for the spatial governance of complex ecosystems.

However, current research has limitations in theoretical construction and spatial scale. On the one hand, empirical studies based on the “value-risk” framework are still in the exploratory stage, and a standardized and universal research paradigm has not yet been formed. There is a lack of systematic theoretical support in key aspects such as indicator selection, weight assignment, and zoning threshold determination. On the other hand, spatial research scales exhibit obvious regional agglomeration characteristics, mainly focusing on economically developed provincial and municipal areas²², urban agglomerations²³, and nature reserves²⁴, among others. It is worth noting that as a fragile ecosystem with special hydrogeological structures and ecological processes, karst areas have relatively scarce research on the coupling of ESV and LER. Few studies have comprehensively incorporated ecosystem services and ecological risks into the same research framework. Carrying out ecological zoning research with a “value-risk” two-dimensional coupling in this special geographical unit may serve as a beneficial supplement to the existing research system and provide a new research perspective for this ecosystem with significant regional heterogeneity.

Guizhou Province is one of the most typical regions in China’s karst regions where karst landforms are widely distributed, with complete morphological types and obvious regional differentiation characteristics²⁵. It is characterized by fragile ecosystems, small environmental carrying capacity and low quality of land resources²⁶. Puding County, a typical karst county in Guizhou Province, serves as both a significant ecological function region in the higher sections of the Pearl River and Yangtze and the hub of economic development in Guizhou’s central region²⁷. Puding County has experienced remarkable increases in the social economy and urbanization processes during the last few decades. The landscape has drastically changed due to unreasonable human activities and unfavorable natural factors. This has led to frequent problems such as land degradation, soil erosion, and a decline in the county’s ecosystem service functions, which has resulted in certain ecological risks, which has threatened the sustainable development of land use. Quantifying the evolutionary trajectories and characteristics of ESV and LER over long time series, breaking through the limitations of single elements in traditional ecological zoning, constructing a theoretical framework and methodological system for ecological zoning based on ESV and LER, and further establishing a differentiated ecological management and control zoning scheme that takes into account both ecological protection and social and economic development are of great significance for the maintenance and service improvement of ecosystem functions. Thus, this study uses land use data to quantitatively analyze the spatiotemporal evolution characteristics of ESV and LER in Puding County from 1973 to 2020, and based on this, carries out the construction of ecological zoning and proposes a differentiated management plan for ecological zoning in Puding County. The scientific problems addressed and research objectives of this study are as follows: First, from the integrated perspective of ESV and LER, this study aims to investigate the spatiotemporal variation characteristics of ESV and LER in Puding County and carry out ecological zoning, so as to deepen the understanding of the interaction between ecosystem services and ecological risks in karst regions. This is expected to provide strong support for helping decision-makers formulate ecological management measures, enhance the level of ecosystem services, and promote regional ecological balance and sustainable development. The second objective is to construct an ESV-LER zoning method for Puding County at the county scale based on long-term time-series data, breaking through the traditional ecological zoning models that rely on administrative boundaries or focus on single elements. By emphasizing the continuity and integrity of ecosystems, this approach not only expands the research boundaries of ecological management and spatial optimization in karst areas but also holds significant importance for effectively revealing the correlation patterns and spatial laws of territorial ecological security, thus enabling comprehensive ecological security construction in the region.

Materials and methods

Overview of the study area

Puding County is located in the western and central regions of Guizhou Province, southwest China (Fig. 1), under the administrative jurisdiction of Anshun City, Guizhou Province. It is situated in the watershed area between the Yangtze River system and the Pearl River system, with a total area of 1091 km² at elevations of 1047–1802 m. The county has a tropical monsoon humid climate, with an average annual temperature of 15.1 °C and rainfall of 1378.2 mm. The region is characterized by a highly developed karst landform, with high terrain in the north and south and flat terrain in the center. The county’s exposed carbonate rock area accounts for 79.2% of the total land area. The most common soil types include lime, yellow-brown, yellow, purple, meadow, and paddy soils and the predominant vegetation is subtropical species. As a typical representative of the ecologically vulnerable karst areas in Southwest China, Puding County faces multiple ecological challenges. Long-term karst erosion and irrational human activities have led to prominent problems of rocky desertification and soil erosion, posing a serious threat to agricultural production and ecological security. Vegetation degradation has resulted in a significant decline in biodiversity, and the ecosystem service functions have been continuously deteriorating. These ecological dilemmas not only restrict regional sustainable development but also serve as a microcosm of ecological governance in the karst regions of Southwest China, making Puding County an important sample area for the study of rocky desertification control and ecological restoration.

Fig. 1

Geographic location and altitude of the study area. Geographic location of Guizhou in China (a); location of the study area in Guizhou (b); DEM of the study area (c).

Data sources and processing

The remote sensing images required for this study were obtained from the University of Maryland and the Geospatial Data Cloud (http://www.gscloud.cn/). The images include Landsat 1–3 MSS imagery from 1973 with a spatial resolution of 60 m, Landsat 4–5 MSS imagery from 1989, and Landsat 7 ETM + imagery from 2003, 2010, and 2020, all with a spatial resolution of 30 m. To improve the accuracy and timeliness of interpretation, this study selected remote sensing images with consistent seasons and cloud cover below 5% as the basic data. First, in the Erdas 9.2 software, spectral enhancement and radiometric enhancement were performed on the five periods of imagery, respectively, followed by standard false-color composite synthesis, geometric precise correction, and atmospheric correction (Supplementary Table S1). Second, supervised classification and human-computer interactive interpretation methods were used for visual interpretation to extract nine land cover types, including forest land (closure degree > 30%), shrub land, other forest land, grass land, paddy fields, dry land, water bodies, construction land, and unutilized land, with field verification and map unit correction. Finally, the five periods of land use data were uniformly converted to the WGS_1989_UTM_Zone_48 coordinate system, and all raster data were uniformly adjusted to a resolution of 30 m. Furthermore, Puding County’s grain output and grain price for 2020 were acquired via review and compilation of data from the National Agricultural Product Cost-Benefit Data Compilation (https://www.yearbookchina.com/) and the Anshun Statistical Yearbook (http://www.anshun.gov.cn/) (Fig. 2).

Fig. 2

Framework of the study.

Research methods

Evaluation unit division and Spatial visualization

In order to calculate the ESV and LER of Puding County and achieve their spatial visualization, this study takes into account the topographical characteristics and area size of the study area, and conducts equidistant sampling based on landscape ecology theory²⁸. Based on previous studies²⁹, this research applied ArcGIS 10.8 software to divide the land use layer of the study area into 1 km × 1 km grids, which served as the assessment units. The spatial analysis tools in ArcGIS 10.8, such as the sampling and clipping tools, were used to resample and calculate the land use data so as to obtain the land use data based on grid units.

The natural break method is a classification method that achieves the optimal grouping of data based on the statistical distribution rules of the data and the idea of clustering. This method can ensure the homogeneity of the data within the groups, enhancing the scientificity and rationality of the grouping results. Its results are intuitive and easy to interpret, which has led to its extensive application in the research of relevant fields of ESV and LER^30,31. In this study, using the ArcGIS 10.8 software platform and based on the 2020 dataset, the natural break method was applied to classify the ESV of the study area into five zones: low ESV zone (ESV < 0.021 × 10⁸), relatively low ESV zone (0.021 × 10⁸ ≤ ESV < 0.028 × 10⁸), moderate ESV zone (0.028 × 10⁸ ≤ ESV < 0.035 × 10⁸), relatively high ESV zone (0.035 × 10⁸ ≤ ESV < 0.042 × 10⁸), and high ESV zone (ESV ≥ 0.042 × 10⁸). Additionally, the LER was divided into five levels: low ecological risk (LER < 0.0170), relatively low ecological risk (0.0170 ≤ LER < 0.0210), moderate ecological risk (0.0210 ≤ LER < 0.0250), relatively high ecological risk (0.0250 ≤ LER < 0.0300), and high ecological risk (LER > 0.0300). Finally, Ordinary Kriging interpolation was used to visualize and quantify the spatial information of ESV and LER.

Landscape pattern index

The landscape pattern indices can quantitatively reveal the dynamic changes in landscape patterns³². Referring to previous research results³³, this study selected a total of 11 landscape indices at the type level and landscape level. The type level indices included the number of patches (NP), patch density (PD), largest patch index (LPI), perimeter-area fractal dimension (PAFRAC), interspersion and juxtaposition index (IJI), and cohesion index (COHESION); The landscape level indices includes NP, landscape shape index (LSI), contagion (CONTAG), splitting index (SPLIT), Shannon diversity index (SHDI), and Shannon evenness index (SHEI). The calculation is based on the land use data of Puding County from 1973 to 2020, and is completed using Fragstats4.2 software.

Measurement of the ESV

Costanza et al. first proposed a methodology for assessing the ESV³⁴, the Value Equivalent Method (VEM), but this method has not been able to match well with the actual status of ecosystem services in China, and there are problems such as severe underestimation of the ESV of arable land and failure to include multiple ecosystem services in the assessment system. Therefore, Chinese scholars Xie et al. drew on this methodology and developed an ESV per unit area equivalent system that fits China’s actual situation through a questionnaire survey of numerous professionals with ecological backgrounds³⁵. Together with the real conditions of the study area and pertinent study findings^24,36, the ESV equivalent table per unit area for terrestrial Chinese terrain, proposed by Xie et al., was consulted to appropriately revise the the primary and secondary classifications and their ESVs in the value comparison Table³⁵. The value equivalent of the forest land and other forest land was based on broad-leaved mixed forest with conifers and coniferous forest as a reference. According to Xie et al. the average grain production’s market value per unit, area in the study area for a given year is equal to one-seventh of the monetary worth of an equivalent factor of the ESV³⁷. In our analysis, we used the average grain production in Puding County in 2020 (6228.57 kg/hm²) and the average grain purchase price (2.93 yuan/kg) without considering factors that affect price fluctuations. Based on the above data, the price of an individual ESV equivalent factor in Puding County was determined to be 2607.10 yuan/hm²(Eq. (1)). The unit area ecosystem service value of Puding County from 1973 to 2020 is shown in Table 1. The ESV of every land use type in the grid unit was determined using the grid land use data(Eq. (2)), and the entire ESV in Puding County was determined using Eq. (3). The following are the precise calculation formulas:

$$\:VC{_{i}} = 1/7 \times Q \times V{_{i}}$$

(1)

$$\:ES{V_j} = \sum {{A_{ij}} \times V{C_i}}$$

(2)

$$\:ESV = \sum\nolimits_{j = 1}^n {ES{V_j}}$$

(3)

where $\:ESV{_{j}}$is the ESV of the jth land use type in the unit grid (yuan); Q is the market value of grain yield per unit area; V_i is the equivalent value of i types of land use; A_ij is the distribution area of the jth grid and ith land use type (hm²); VC_i is the ith land use type’s ESV factor (yuan/hm²); and n is the entire amount of categories for land use.

Table 1 Ecosystem service value equivalent per unit area for puding County (yuan/hm²).

Construction of LER

The landscape pattern indices capture and condense the information of regional landscape pattern, and are an important indicator to reveal the spatial structure configuration characteristics of landscape components. This study is based on the proportion of various land use types, starting from the perspective of landscape pattern, combined with the specific conditions of the study area and existing research results³⁸, using vulnerability index, landscape disturbance index, landscape separation index, and the area ratio of landscape components, to construct the LER of the study area (Supplementary Table S2). The following are the precise calculation formulas:

$$\:LER = \sum\nolimits_{i = 1}^n {\frac{{{A_{ki}}}}{{{A_k}}}} R$$

(4)

$$\:{R_i} = {V_i} \times {D_i}$$

(5)

$$\:D = a{F_i} + b{S_i} + c{T_i}$$

(6)

Where n is the total quantity of the various scenery types; A_ki is the area of the ith landscape type in risk community k; A_k is the entirety of risk community k; R_i is the loss index; V_i is the vulnerability index; D_i is the interference index; F_i is the fragmentation index; S_i is the separation index; and T_i is the dominance index.

The weights a, b, and c represent the respective landscape indices, satisfying a + b + c = 1, and are assigned values of 0.5, 0.3, and 0.2 based on relevant research³⁹. By referring to relevant studies and combining with the actual situation of the study area, the vulnerability grading of land use types in the study area was carried out, and normalization processing was performed to obtain the vulnerability indices of different land use types^24,40,41.

Ecological service value-landscape risk ecological zoning quadrant division

Z-score standardization is a method of transforming data into a standard normal distribution, which can effectively eliminate the dimensional differences between indicators, thus achieving comparability among the indicators⁴². Z-score standardization not only enhances the comparability of data, but also has the characteristics of improving the stability and interpretability of the data, and thus has been gradually applied to the construction of ecological zoning⁴³. Four quadrants were delineated based on the Z-score standardization of 1186 grid units in the study area. This was conducted in compliance with the computation of the ESV and LER. Then plot the standardized ESV and LER values on a two-dimensional coordinate system. X and Y represent the standardized ESV and LER, respectively. Quadrants I, II, III, and IV represent the high ESV-high LER, low ESV-high LER, low ESV-low LER, and high ESV-low LER quadrants, respectively. The exact computation formulas are as follows:

$$\: x = \frac{{{x_i} - \overline x }}{s}$$

(7)

$$\: \overline x = \frac{1}{n}\sum\nolimits_{i = 1}^n {{x_i}}$$

(8)

$$\:S = \sqrt {\frac{1}{n}\sum\nolimits_{i = 1}^n {{{\left( {x - \overline x } \right)}^2}}}$$

(9)

where x is the ESV and LER after grid unit standardization; x_iis the ESV and LER of the ith grid unit; $\:\overline x$ is the research area’s average value; S is the standard deviation of the study area; and n is the total number of square units.

Results

Spatiotemporal characteristics of land use change

As can be seen from Fig. 3, the predominant land use type in Puding County, forest land, accounted for 563.63 km² by 2020, or around 52.00% of the study area’s total area (shrub land > forest land > other forest land). Dry land, making up about 33.96% of the entire study area, was the next largest land use type. With a decline of 56.72% from 1973 to 1989, the change in the forest land area was the most noticeable. Since Puding County adopted the returning farmland to forest policy, the amount of forest land continued to increase from 1989 to 2020, and the growth of shrub land increased by 5.13 times during the same period. The amount of cultivated land and grass land, significantly decreased during the study period, decreasing by 24.44% and 56.85% respectively. Puding County has been affected by rapid urbanization and population expansion since 1973, and the construction land has increased year after year, and the area has expanded by 96.55 km², an increase of 72.01 times. The area of water bodies increased by 3.46 times during the study period.

As illustrated in Fig. 4, from 1973 to 1989, the transfer of land use was mainly manifested as the transfer of forest land to dry land and paddy fields, with a transfer area of as much as 203.03km². This is chiefly owing to the surge in population during that span of time, which led to a large amount of forested land being reclaimed as arable land to alleviate the demand for food, the occurrence of a large amount of deforestation, reclamation of steep slopes and other acts of destruction of forests and meadows, and the aggravation of the regional problems of rocky desertification and water and soil erosion. A significant amount of dry land, covering an area of 109.43 km², was converted to forest land between 1989 and 2003 due to the policy of restoring farmland to forest. Due to Puding County’s urbanization and reservoir building during this time, there was a surge in the area of construction land and water bodies. From 2003 to 2010, the area of diverse land use types converted to other types increased significantly, and the land use shift accelerated, which was manifested as mutual conversion between dry land and paddy fields, with conversion areas of 51.12 km² and 63.98 km², respectively. The largest transfer from 2010 to 2020 involved the conversion of cultivated land and shrub land, with 92.35 km² of dry land being converted to shrub land. The area of the other forest land decreased by 61.09 km² in the same period in which a significant portion of other forest land was turned into forest land. With the ongoing development of regional urbanization, a great deal of land of various types has been successively transferred to construction land.

Fig. 3

Area of land use types in Puding County from 1973 to 2020.

Fig. 4

Land use transfer matrix for Puding County from 1973 to 2020. (a)Represents the land use transfer from 1973 to 1989; (b)Represents the land use transfer from 1989 to 2003; (c) Represents the land use transfer from 2003 to 2010; (d) Represents the land use transfer from 2010 to 2020.

Characteristics of landscape pattern change

Tables 2 and 3 show that the NPs and LSI in Puding County exhibited increasing trends from 1973 to 2020, while the CONTAG and the SPLIT decreased and the SHDI and SHEI remained relatively stable. Overall, the study area’s landscape arrangement tended to be fragmented and simplified, with weakening trends of connectedness and aggregation between patches. The NP exhibited a “V-shaped” increasing trend, with an increase as high as 27.77%. The LSI showed a “W-shaped” fluctuation during the study period, with an overall increase of 4.20. The CONTAG generally decreased by only 0.87%, indicating that the connectivity among landscapes remained in a relatively stable development state during the study period. From 1973 to 1989, due to the continuous expansion of dry lands and paddy fields as dominant landscapes, the SPLIT sharply decreased, with a decrease of 81.58%, and the landscape tended to become concentrated. However, after 1989, the patches of dry land and paddy fields tended to be discrete. From 1973 to 2020, the SHDI and SHEI exhibited similar change trends, and both reached their peaks in 2010.

From the perspective of the landscape type level, a substantial increasing trend occurred across all measures for the construction land, indicating that the degree of human interference gradually increased during the study period and the landscape tended to become more complex and fragmented. Notably, in 2010, the PD and NP of the arable land reached 6.05 patches/km² and 6563. This may have been an outcome of the conversion of a significant quantity of cultivated land to forest and construction land, which greatly increased the fragmentation of the cultivated land. As the advantageous landscapes in Puding County, the forest land and arable land were distributed in a local area. Over the course of the study, the forest land’s LPI value declined at first and subsequently increased, resulting in a cumulative rise of 7.47. Its impact on the landscape as a whole grew and expanded. In contrast, the LPI values of dry land and paddy fields showed opposite changing trends. There was first an increase and then a decrease in the number and density of unused land patches, indicating that the land was being used efficiently. The PAFRAC values of all of the landscape types in Puding County were stable at around 1.5, and the mosaic structure of the patch morphology tended toward a random state similar to Brownian motion. The NP and PD of the forest land were the largest (the proportion of shrub land was the highest), and the PAFRAC of the water area was the largest (close to 1.5), indicating that its distribution was complex and unstable. The aggregation degree of the cultivated land was the greatest, the connectivity between the patches of cultivated land was the best, and the transport of material and energy was relatively smooth.

Table 2 Changes in the landscape level index in puding County from 1973 to 2020.

Table 3 Changes in the landscape type level index in puding County from 1973 to 2020.

Spatial and Temporal characteristics of the ESV

ESV Estimation

As shown in Table 4, Puding County’s ESVs were 3408.58 million yuan, 2417.06 million yuan, 3243.29 million yuan, 3593.32 million yuan, and 3923.56 million yuan in 1973, 1989, 2003, 2010, and 2020. The ESV of the study area increased by 514.98 million yuan during the 50-year study period, exhibiting a V-shaped trend, i.e., initially decreasing and then gradually increasing. With a contribution that represented almost 60.08% of the overall ESV, the forest ESV had the greatest rate of contribution and shrub land > forest land > other forest land. The cultivated land was the next largest, accounting for 16.57% of the ESV in the study area. The ESVs of the water bodies and forest land typically exhibited fluctuating increasing trends. The ESV of the water bodies experienced the greatest variation within the study period, with an increase of up to 499.28 million yuan. As an outcome of the substantial increase in the forest area since 1989, the ESV of the forest land continuously increased, reaching 2.54 times that in 1989 by 2020. This indicates that forest ecosystems are of irreplaceable importance in enhancing ESV in Puding County, and they play a pivotal role in maintaining ecological balance, promoting biodiversity and supplying natural resources. The ESVs of the cultivated land and grassland exhibited fluctuating decreasing trends. The ESV of the cultivated land reached a peak of 784.27 million yuan in 1989, and then, it exhibited a steadily decreasing trend.

From the perspective of the single ESV (Table 5), it can be summarized as three cascades. Climate regulation and hydrological regulation comprised the first cascade of the single ESV, accounting for over 50.00% of the ESV. From 1973 to 2020, they initially decreased and then increased, with increases of 45.74 million yuan and 380.00 million yuan, respectively. The second cascade included the following: soil conservation > gas regulation > biodiversity > environmental purification. The ESVs of these four categories were 377.78 million yuan, 340.16 million yuan, 281.13 million yuan, and 235.09 million yuan. Food production, raw material production, water resource supply, nutrient cycling maintenance, and aesthetic landscape comprised the third cascade of the individual ESV, accounting for less than 12.00% of the ESV. Influenced by the thin soil layer and easy nutrient loss in the karst region, the study area is weak in maintaining the nutrient cycling function, providing a value of less than 50 million yuan. In addition, the value of water supply services was persistently negative during the study period, which was mostly ascribed to the heavy consumption of water resources by paddy fields and the serious leakage of groundwater resources due to the unique geological structure of the karst region, which in turn led to the comparative scarcity of surface water resources.

Table 4 Changes in ecosystem service values of different land use types from 1973 to 2020.

Table 5 Value and proportion of the individual ecosystem services from 1973 to 2020.

Visualization of the ESV space

Figure 5 illustrates the great spatial variability of the ESV in Puding County, which was mostly represented by the rough spatial distribution of high in the northeast and low in the southwest. Specifically, the northeast region, which was primarily composed of forest and water bodies, contained the majority of the highest-value areas and higher-value areas. This is primarily due to the favorable hydrothermal conditions in this part of the country, with better vegetation growth and cover to the extent that the ESV contribution is higher. The highest-value locations were dispersed in a dense, continuous pattern, whereas the higher-value areas were dispersed haphazardly around the highest-value areas. The main land use types in the moderate-value areas and lower-value areas were dry land and paddy fields, which were relatively dispersed and distributed in strips around the periphery of the high value areas, with a certain degree of spatial transition. The lowest-value areas were principally located in the northwestern part of the study area, which primarily consisted of unused land and construction land, as well as some dry land and paddy fields. The population density in this region is relatively high, and it has been severely damaged by human interference, with most of the ecological land having been encroached upon. According to time series analysis, from 1973 to 1989, a significant portion of the ecological land was converted into arable land, which resulted in the widespread conversion of high value areas into low value areas. However, after 1989, the change trend was the opposite. In addition, the spatial distribution of the ESV in Puding County significantly changed over time, and the difference between the high and low values was at least 13 million yuan after 2003.

Fig. 5

Spatial distribution of ecosystem service value from 1973 to 2020. (a)~(e) represent the spatial distribution of ecosystem service values in Puding County from 1973 to 2020.

Spatial and Temporal characteristics of the LER

Quantitative characteristics of LER

The average LERs in Puding County for the years 1973, 1989, 2003, 2010, and 2020 were 0.0216, 0.0215, 0.0205, 0.0203, and 0.0224, respectively. The aggregate LER exhibited a decreasing-increasing trend. According to our findings (Fig. 6), there were basically three degrees of LER for the various land use types over the 50-year study period: the vegetation coverage area of the unutilized land was small, the ecological sensitivity was high, and the average LER was 0.1566, comprising the first level of the LER. It is noteworthy that the high degree of separation of the unutilized land in 2020 made its LER as high as 0.5140, demonstrating that the unutilized land, although occupying a small area, is still a potentially big threat threat to the ecological security construction in the study area. The other forest land, water bodies, and grassland comprised the second level of LER. Due to elements such as topography, terrain, and human involvement, their fragmentation was high and their ecological stability was low. The forest land, shrub land, paddy fields, dry land, and construction land comprised the third level of the LER, mainly due to these types’ concentrated distribution and stable landscape pattern.

From the perspective of the ecological risk contribution weight, the dry land had the highest ecological risk contribution weight, reaching 28.00%. It initially increased and then decreased throughout the study period, peaking at 49.66% in 2010. The next category was the other forest, paddy fields, and shrub land, all of which had a high potential for ecological harm. The ecological risk weights of the grassland, forest land, water bodies, and unused land were relatively small, all of which did not exceed 10%. Despite having relatively low contribution weights, these types were crucial to the development of ecological security in Puding County. The contribution weight of the LER of the construction land was the smallest, and it did not exceed 3.00%. During this stage, the contribution weight of construction land showed a continuous increasing trend. This also somewhat confirms that the security of the ecological pattern may have been in danger from the increase in the construction land area.

Fig. 6

Landscape ecological risk and its contribution weight from 1973 to 2020. The bar chart represents the landscape ecological risk of different land use types from 1973 to 2020; the pie charts (a)~(e) represent the contribution weights of landscape ecological risk of different land use types in Puding County from 1973 to 2020 in turn.

Change in LER structure

The classes of LER zones in the research area were determined via statistical analysis (Table 6). We discovered an increasing prevalence of a serrated pattern in the lowest ecological risk regions. In 2010, the percentage of the lowest ecological risk areas was the greatest (32.65%). In 1973, 1989, 2003, 2010, and 2020, the areas with lower ecological risk covered 433.55 km², 358.09 km², 509.79 km², 446.42 km², and 322.91 km², respectively, exhibiting a fluctuating decreasing trend. The evolution trajectory of the moderate ecological risk zone regularly mirrored that of the lower ecological risk zone. The area of the higher ecological risk region and highest ecological risk region exhibited a range of increases, and the highest ecological risk zone in 2020 increased to 4.50 times that of 1973. High-intensity human activity in Puding County during the 50-year study period significantly altered the regional landscape pattern. The policy of returning farmland to forest and grassland, ecological restoration initiatives, and reservoir construction increased the area of water and forest, and the quality of the environment gradually improved. However, the acceleration of urbanization increased landscape fragmentation, weakened anti-interference ability, enhanced vulnerability to external influences, and increased recovery difficulty, which greatly contributing to the further expansion of the high ecological risk area.

Considering the structure of the LER levels, the lower and moderate ecological risk levels accounted for 63.43% of the study area, dominating the LER of Puding County during the five periods. This demonstrates that the stability of the landscape pattern of Puding County has always been maintained at a relatively high level, and that the policy of forestation and soil and water conservation measures implemented in recent years have favorable ecological benefits for Puding County. The areas with the lowest ecological risk levels and higher ecological risk levels were both relatively modest. Throughout the research period, their respective proportions ranged from 10.11 to 32.65% and 8.23 to 18.50%. The highest ecological risk category had the smallest area, with an area ratio of no more than 8.15%. It should be noted that despite the relatively small size of the highest ecological risk areas, it exhibited a significant trend of continuous growth over the past 20 years. This may have contributed to the increase in the LER in the study area.

Table 6 Area and proportion of landscape ecological risk levels from 1989 to 2020.

Spatial distribution characteristics of LER

As can be seen from Fig. 7, the spatial distribution of the LER of Puding County changed dramatically during the 50-year study period, with distinct spatial specificity and temporal instability. The study area’s central and western regions contained the vast majority of the highest ecological risk areas, and these areas were aggregated. The main landscape types were unutilized land and other forest land types. The moderate ecological risk and higher ecological risk regions were spread out and randomly distributed around the high ecological risk areas. The main landscape types were water bodies, grasslands, and other landscape types that were vulnerable to human interference, had low stability, and were relatively scattered. The lower and lowest ecological risk areas mainly exhibited an irregular distribution pattern with a patchy distribution pattern and were concentrated in the northern and southwestern parts of the study area. While the primary landscape types in the lowest ecological risk areas were shrub land and forest landscapes with favorable ecological surroundings, the predominant landscapes in the lower ecological risk areas were mostly concentrated and stable landscape types such as dry land and paddy fields.

Fig. 7

Spatial distribution of landscape ecological risk zones from 1973 to 2020. (a)~(e) represent the spatial distribution of landscape ecological risks in Puding County from 1973 to 2020.

Ecological space zoning construction

Ecological Spatial zoning pattern

Through Z-score standardization processing, the ecological zoning quadrant map, geographical distribution (Figs. 8 and 9), and ecological zoning statistics of the study area were derived (Table 7; Fig. 10). The results indicate that the distribution of the four types of ecological zones in the study area was relatively uniform, and based on the overall distribution trend, all ecological zones underwent varying degrees of change.

The low ESV-low LER area (III), which accounted for the largest proportion and had a distribution area of over 30%, gradually shrank after reaching its peak (437.21 km²) in 1989. The area decreased by 27.32 km² overall throughout the study period, mostly in the northwestern and southern parts of the study area, which primarily consisted of dry land, paddy fields, and construction land. The variation range of the ESV was 609.29–788.01 million yuan, the variation range of the LER was 0.0130–0.0171.

The high ESV-high LER area (I) was primarily distributed in the marginal areas in the west, north, and northwest, and its distribution was highly coincident with that of the water bodies, grasslands, and other forest lands. The change in its area exhibited a V-shaped fluctuating decreasing trend, reaching the lowest value of 186.44 km² in 2010. The decrease from 1973 to 2010 was 41.91%. The ESV in this area exhibited a saw-toothed growth pattern with a range of 936.30–1434.60 million yuan. The LER also exhibited an erratic increasing trend, with an average of 0.029.

The distribution area of the low ESV-high LER area (II) was the smallest, except for a considerable distribution in the southeast region in 1989, other research periods showed a highly dispersed distribution state. Unused land, some scattered paddy fields, and dry land accounted for the majority of the land use types in this region. During the research period, this region was in a state of continuous growth, which, to a certain extent, increased the potential hidden dangers to the safety of the regional ecological functions. From 1973 to 2020, the ESV in this region continued to increase from 267.70 million to 466.00 million, and the LER was high (> 0.0262), with a rise of 16.60% over the study period.

The high ESV-low LER area (IV) underwent significant alteration, exhibiting an increasing zigzag trend. From 1973 to 1989, the area shrank to its minimum value of 187.64 km², but from 1989 to 2010, it expanded by 173.26 km².This zone exhibited a somewhat dense distribution in the northern region of the study area and a general blocky distribution pattern dispersed over the entire study area. The primary land types in this zone included forest land, shrub land, and a minor amount of grassland. The ESV and LER remained within the ranges of 6.05 to 15.72 million yuan and 0.0152 to 0.0180, respectively, exhibiting relatively severe changes in the ESV overall, but there was no significant change in the LER.

The results of the study revealed that the distribution of the four types of eco-zones in the study region was relatively even, and in terms of the overall distribution trend, all the eco-zones changed to various degrees. At the same time, the phenomenon of synchronized increases in ESV and LER in some ecological subregions reflects that, although the implementation of ecological protection policies has significantly improved the quality of the regional ecological environment, these policies have, to a certain extent, interfered with the intrinsic stability of the ecosystem. Such interventions may lead to dissimilation of ecosystem structure and function, resulting in fragmentation and fragility of regional landscapes, which in turn increases the ecological risk in some areas. Therefore, when formulating and implementing ecological conservation strategies, their underlying negative externalities must be comprehensively taken into account. Comprehensive or targeted approaches should be adopted to balance the relationship between ecological conservation and ecosystem stability, ensuring that ecological conservation measures promote biodiversity and ecological services while maintaining the long-term stability and resilience of ecosystems.

Fig. 8

Quadrant distribution of ecological zones in Puding County from 1973 to 2020. (a)~(e) represent the distribution of ecological zoning quadrants in Puding County from 1973 to 2020.

Fig. 9

Spatial distribution of ecological zones from 1973 to 2020. (a)~(e) represent the spatial distribution of ecological zones in Puding County from 1973 to 2020.

Table 7 Changes in ecological zoning in puding County from 1973 to 2020.

Fig. 10

Quadrant distribution of ecosystem service value and landscape ecological risk in ecological zones. I, II, III, IV represent high ESV-high LER area, low ESV-high LER area, low ESV-low LER area, and high ESV-low LER area, respectively.

Ecological zoning control scheme

According to the ESV and LER characteristics of the grid units, we divided Puding County into ecological risk improvement zone, ecological comprehensive restoration zone, ecological function enhancement zone, and ecological conservation and maintenance zone (Fig. 11). The specific ecological zoning control plan is described below.

• (1) Ecological risk improvement zone (I): This zone is common in high ecological environment quality areas that are vulnerable to disturbance from human activities. The principal categories of land use include grassland, other forest land, and water bodies. From the perspective of the protection of ecosystem service functions, the core objective of ecological restoration is to maintain the integrity of the ecosystem, enhance its ability to resist interference, and effectively reduce the negative impacts of human activities on the ecosystem service functions. Based on the ecological characteristics of the region, it is necessary to strictly control the intensity of human production and living activities. Measures such as demarcating the ecological protection red line and establishing an ecological compensation mechanism should be taken to prevent soil erosion and vegetation degradation. In view of the problems such as high ecological risks and poor landscape connectivity in this region, it is recommended to adopt a composite restoration model combining natural restoration with artificial intervention. On the one hand, biological engineering measures such as afforestation and forest tending can increase the coverage area of green spaces, the degree of habitat fragmentation is reduced, and the construction of ecological corridors is promoted. On the other hand, in combination with the development characteristics of the karst landform in Puding County, and aiming at the problem of weak surface water storage capacity, small water conservancy facilities like ponds and water cellars are built according to local conditions at the downstream of slope catchment areas to optimize the regulation system of surface runoff. In addition, the water area protection system needs to be strictly enforced, an ecological governance plan with water storage and soil conservation as the core should be formulated, and engineering, biological and agricultural technical measures should be comprehensively applied to construct a multi-level soil and water conservation system.

• (2) Ecological comprehensive restoration zone (II): This zone is mainly composed of unused land and highly fragmented dry land and paddy fields. It covers a small area, has a scattered distribution, and contains areas with high landscape fragility. From the perspective of county-level spatial governance, it is necessary to systematically carry out land use planning and comprehensive renovation work. For dry lands and paddy fields with severe rocky desertification and slopes exceeding 25 degrees, the policy of returning farmland to forest should be implemented through artificial afforestation projects to promote their transformation into forest ecosystems. A sound ecological compensation mechanism should be established, and economic incentive measures should be adopted to coordinate the relationship between ecological protection and land use, optimize the regional ecological landscape structure, and achieve the dual goals of improving the ecological environment quality and preventing and controlling ecological risks. In terms of land development control, it is necessary to strictly limit the development intensity of unused land. By scientifically demarcating ecological protection boundaries and optimizing the land use layout, the landscape fragmentation can be reduced. In view of the “mosaic” spatial distribution characteristics of this area, it is recommended to clarify the main responsibilities of ecological restoration at the township level in accordance with the principle of proximity, and establish a grid management mechanism as a supporting system. It is recommended to combine 3 S technology to achieve refined supervision, so as to improve the management efficiency of land resources and promote regional ecological restoration and sustainable development.

• (3) Ecological function enhancement zone (III): This type of area is the most widely distributed in the study area, mainly consisting of dry land, paddy fields, and construction land, which are used for human production and living. It is a key area for Puding County to improve vegetation cover and optimize the quality of the ecological environment. The main development focus is to enhance the ESV of the region. Based on the regional ecological foundation and the characteristics of social and economic development, it is necessary to establish a collaborative governance system for ecological protection and spatial development. Firstly, strengthen the control of ecological space, strictly enforce the cultivated land protection system, and strictly prohibit the transfer of agricultural land for non-agricultural purposes and for uses other than grain production, so as to curb the trend of ecological function degradation. Secondly, guided by the enhancement of the value of ecosystem services, systematically promote the construction of urban ecological corridor networks to enhance the connectivity and stability of the ecosystem. In addition, implement the project to improve the quality of urban green spaces, scientifically plan diversified ecological spaces such as urban parks and rooftop greening. While improving the human settlement environment, it can effectively increase regional vegetation coverage and achieve the synergistic improvement of ecological and social benefits.

• (4) Ecological conservation and maintenance zone (IV): In Puding County, this zone is a vital ecological conservation area with a healthy ecological setting. This area is mainly composed of forest ecosystems, including shrub land and forest land, which play a decisive supporting role in the regulation services of Puding County. For this region, human development and construction activities should be strictly restricted, and acts of destruction such as illegal logging and forest destruction for reclamation should be severely cracked down on in accordance with the law. In terms of ecological restoration and functional enhancement, in light of the hydrological characteristics of karst regions, small-scale rainwater collection systems should be constructed in ecological conservation and water retention areas according to local conditions. Through facilities such as water collection ponds and infiltration ditches, effective interception and infiltration of rainwater can be achieved, reducing the scouring intensity of surface runoff and improving the soil water content and fertilizer retention capacity. Meanwhile, measures such as replanting and supplementary afforestation, as well as closing mountains for afforestation, should be taken to optimize the forest community structure, further enhancing the service functions of the ecosystem, such as carbon sequestration, oxygen release, and water conservation.

Fig. 11

Optimization strategy for ecological zoning in Puding County. H-H, L-H, L-L, H-L represent ecological security type regulation zone, ecological improvement type regulation zone, ecologically improved conservation zone, and ecological protection type conservation zone, respectively.

Discussion

Spatiotemporal evolution characteristics of land use and ESV

The study’s findings indicate that throughout the last 50 years, Puding County’s land usage has undergone significant change. There have been frequent transitions between cultivated land, forest land, and grassland areas, and the areas of construction land and water bodies have grown as well. This result is consistent with the findings of Qin et al. regarding the land use change in Puding County³³. A substantial portion of the forests and grassland were converted to dry land and paddy fields between 1973 and 1989 because of the problems of food security brought about by the population boom. This phenomenon is consistent with the trend of land use change caused by population pressure in many regions, further confirming the negative impact of population growth on the ecological environment⁴⁴. During the research period, the area of construction land surged. Constrained by the natural conditions such as the significant topographic undulations and uneven development of karst landforms in the study area, the development of urbanization often has difficulty expanding into areas with complex terrains. Instead, it relies on occupying dry land and paddy fields with relatively gentle terrains. From 1973 to 2020, the ESV of Puding County initially decreased and then increase, increasing by an aggregate of 514.98 million yuan. This was mostly attributable to the implementation of the Grain for Green Project and the reservoir construction in Puding County, both of which considerably encouraged the growth of shrub land, forest land, and water bodies, increasing the local ESV. This conclusion is essentially compatible with previous research findings regarding the ESV in karst areas^45,46. It fully demonstrates the crucial role of policy intervention in ecological restoration. However, compared with the research results of Zhang et al. and Shang et al. on the ESV in non-karst areas, the restoration process of the ESV in Puding County is relatively slow^47,48. This difference may be due to the constraints of the special geological conditions in the karst area, such as the thin soil layer and the increased risk of rocky desertification. These factors jointly limit the self-restoration ability of the ecosystem and the rapid improvement of its service functions. From 1973 to 1989, the forest area in Puding County decreased by 286.73 km², corresponding to a decrease of 1269.99 million yuan in terms of the ESV. Moreover, from 1989 to 2020, the area of forest land increased, and the value of ecosystem services also gradually increased. At the same time, it shows that as an important support for the value of the ecosystem services in Puding County, the forest ecosystem should be given more attention in terms of protection in the future. Notably, between 1973 and 2010, the water resource supply service in the single ESV was always negative. The widespread distribution of karst landforms in Puding County results in high water seepage and a poor water storage capacity, may have been the cause of this phenomenon. In addition, the dry land and paddy fields in Puding County covered a large area, and a significant amount of water resources were needed to support agricultural production. This conclusion is remarkably close to the research results of He and Yao regarding water resource supply services in karst areas⁴⁹.

Change characteristics of the landscape pattern and LER

As a core indicator for evaluating the regional ecological security pattern, the dynamic change of LER profoundly reflects the interactive influence between human activities and the natural environment. During the period from 1973 to 2020, the LER in Puding County showed obvious characteristics of spatio-temporal differentiation, and on the whole, it experienced an evolutionary process of “first decreasing and then increasing”, which highly coincides with the research conclusion of scholars such as Xia et al. regarding karst areas⁵⁰. This indicates that during the stage of policy-driven ecological restoration, measures such as closing mountains for afforestation and returning farmland to forest and grassland have effectively alleviated the ecological pressure. However, due to the increase in regional landscape fragmentation, the LER has risen again. The intensification of local ecological vulnerability still highlights the high sensitivity of the ecosystem in karst areas to human activities and the limitations of restoration, suggesting that the ecological security pattern in karst areas still faces challenges. Puding County’s LER increased significantly between 2010 and 2020. This increase may have been due to the rapid urbanization of the study area and the dispersed distribution of the construction land during this period, which increased number and density of patches, the fragmentation and vulnerability of the landscape, and the LER. This is basically consistent with the research results of Wang et al. and Xu et al.^51,52. Based on the existing research achievements, the expansion of construction land has a significant two-way effect and spatial heterogeneity on LER. Firstly, the orderly and centralized expansion of construction land can effectively reduce landscape fragmentation, enhance landscape connectivity and integrity, and thus inhibit the rise of the LER level. Secondly, the decentralized and disorderly expansion of construction land and the unreasonable land use structure are likely to lead to problems such as low land use efficiency and the squeezing of ecological space, exacerbate the discretization of landscape patches and functional imbalance, and thus significantly increase the regional LER level. Therefore, in future urban development, the disorderly expansion of construction land should be strictly curbed. Urban development boundary lines should be established, and the layout of construction land should be optimized to avoid exacerbating ecological risks due to patch fragmentation. At the same time, the protection of forest land and water area land should be strengthened to maintain the regional ESV at a high level, giving full play to their key role in regulating LER and ESV.

Construction of ecological zoning based on ESV and LER

The karst region, with its distinctive geological structure and hydrological characteristics, is considered to be one of the most vulnerable regions in the world in terms of natural ecosystems. In recent years, driven by both anthropogenic interference and the natural environment, karst regions have faced many challenges, such as serious soil erosion, intensifying rocky desertification and poverty, which have put forward new requirements for regional ecological management. Ecological zoning is an important instrument for realizing the integrity and uniqueness of regional ecosystem management, which not only contributes to a comprehensive understanding of the actual state of regional ecosystems, but also effectively mitigates the negative impacts of human activities on the sustainable development of ecosystems⁵³. This study combines the unique ecosystem characteristics of the karst region and integrates two ecological indicators, ESV and LER, to construct ecological zoning. Studies have shown that although the value of the ecosystem services in Puding County has shown a steady upward trend as a whole since 1989, the landscape fragmentation in the region has continued to intensify, and the LER value has increased simultaneously, further reflecting the urgency of constructing ecological zoning by combining the two dimensions of the ecosystem and human well-being. This contradictory phenomenon of the improvement of the value of ecosystem services coexisting with the exacerbation of LERs reveals the limitations of evaluating the value of single ecosystem services, and further highlights the necessity and urgency of integrating the dual perspectives of the structural functions of the ecosystem and the needs of human well-being to construct a multi-dimensional ecological zoning system. The study identified that the area proportion and spatial distribution pattern of different ecological zones in Puding County continuously evolved from 1973 to 2020, revealing the obvious instability of the ecosystem structure in the region. Based on ESV and LER, ecological zones are constructed, and differentiated ecosystem management measures and future development plans are proposed for different ecological zones in Puding County. It aims to realize the optimal management and sustainable development of the ecosystem, and also provides an essential reference and scientific basis for the enhancement of ecosystem service functions and the coordinated development of ecology-economy-society in the karst region, which is of great theoretical and practical value.

Uncertainty analysis and prospects

In terms of the research data, the establishment of an ecological zoning pattern in Puding County based on the ESV and LER can better link human well-being with ecological environment changes, which is more conducive to the construction of targeted ecological management and control programs. However, in this study, the ESV and LER were calculated from land use data. Compared with other regions, karst areas have weak water and soil conservation capacities and large terrain fluctuations, which have significant spillover effects on the ecological environment. Therefore, there are still some differences between the real condition and the analysed ESV and LER. According to the research methods, because it is difficult to clearly define the advantages and disadvantages of construction land on ecological environments, the equivalent value of the ESV per unit area was set to 0 based on the research findings of scholars such as Xie et al., which may have had a slight influence on the precision of the ESV assessment³⁵. In addition, in this study, we did not examine the key variables influencing the creation of the ecological zones, and the connection between the ESV and LER still requires further study.

Conclusions and prospects

This study takes a county in the karst area as the spatial unit. From the perspective of integrating ESV and LER, it constructs ecological zoning based on identifying the spatio-temporal changes of these two factors in the karst region. The main conclusions are as follows: (1) Shrub land and dry land are the main types of land use in Puding County, accounting for more than 56% of the total area of the study area; over the 50 years, the transferred area of land use in Puding County reached 764.69 km², with the most dramatic transfers occurring between dry land and shrub land and into construction land. (2) In 1973, 1989, 2003, 2010, and 2020, Puding County’s ESV was 3408.58 million yuan, 2417.06 million yuan, 3243.29 million yuan, 3593.32 million yuan, and 3923.56 million yuan, respectively, with shrub land contributing the most to the county’s ESV. The ESV was high in the northeast and low in the southwest in terms of location. The LER in Puding County is relatively low, with dry land contributing the largest weight to LER, reaching 28.00%. Spatially, the LER exhibits a distribution pattern of “high in the central part and low in the surrounding areas”. (3) The low ESV-low LER area (III) accounted for the greatest proportion of the study area (> 30%), and it was primarily located in the northwestern and southern parts of the study area. The low ESV-high LER area (II) had the smallest distribution area and exhibited a widely dispersed distribution.

Compared with traditional zoning systems that focus on single natural elements or unidirectional ecological functions, this study comprehensively considers ecosystems and human well-being. By deconstructing the multidimensional interactive relationship between ESV and LER, this study constructed a more comprehensive and integrated ecological zoning framework in karst geomorphic units, providing a new research path for ecological spatial governance in complex geographic units. This is not only a new attempt to integrate ESV and LER analysis in karst areas but also provides universal theoretical support and scientific basis for the scientific and optimized allocation of land resources in karst areas. It has exploratory academic value in improving the ecological governance technology system and promoting the coordinated development of human-land systems. In addition, based on the ecosystem characteristics and spatial differentiation patterns of the study area, this research divided the study area into four different ecological management zones. Differentiated control strategies were proposed according to the characteristics of each zone, which not only responds to the special hydrogeological conditions and ecological process requirements of karst areas but also provides scientific references for the precise formulation of regional land use planning and ecological protection policies. This is of great practical significance for promoting the transformation of ecological security construction from an extensive management model to a refined governance model.

Although this study has made certain marginal contributions to the accumulation of knowledge in related fields, future research should consider strengthening in-depth exploration in the following two aspects: On the one hand, given that the dynamic evolution of ESV and LER is comprehensively driven by multi-dimensional factors such as natural backgrounds, human activity intensity, and policy systems, future research should focus on identifying and analyzing the key factors driving changes in ecological zoning. Furthermore, more precise ecological zoning strategies should be formulated based on the action mechanisms and effects of different influencing factors. On the other hand, because of the complex interplay between biological processes and landscape patterns, the outcomes of the LER assessment are inconsistent. Therefore, it is necessary to improve the identification of high-risk uncertain areas and to simulate the various risk preferences of decision-makers to lessen the influence of emotional cognition on the assessment outcomes. Moreover, we should further explore the multi-scale effects and nonlinear characteristics of the relationship between the different types of landscape components and the configuration characteristics and LER to better grasp the response of LER to landscape patterns, reduce the uncertainty of the risk assessment results.

Data Availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.