Spatial–Temporal Evolution and Correlation Analysis of Human Activity Intensity and Resource Carrying Capacity in the Region around Poyang Lake, China, from 2010 to 2020

Abstract

To explore the influence of human activities on natural resources and the correlation between these two characteristics, taking the region around Poyang Lake in China as the research area. Utilizing data from 2010, 2013, 2015, 2017, and 2020, including remote sensing, gross domestic product (GDP), and population density data, the human activity intensity (HAI) and resource carrying capacity (RCC) were calculated using relevant models. Analyzing the dynamic change characteristics, the temporal and spatial correlations, and the interaction between the two aspects, bivariate spatial autocorrelation and other methods were used. The results show that: (1) from 2010 to 2020, there was a significant increasing trend in the HAI, while the RCC remained at a generally stable level. (2) In both the temporal and spatial dimensions, there was a certain positive correlation between the HAI and RCC. Their interaction, however, is primarily based on the intrinsic resources and human activity factors within each region, with limited observable factor flow between regions. (3) Regions with the medium level of HAI have greater development potential. The results are expected to enrich the research on human activities and natural resources in the region around Poyang Lake in China and provide a reference for relevant planning.

1. Introduction

With the growth of the population and the rapid development of the social economy, the conflict between human beings and the natural environment is intensifying day by day. How to make resources and the environment meet the growing material needs of people and the requirements for the sustainable development of the regional economy, while ensuring the stability of their own productivity is the focus of current sustainable development research [1,2,3]. As the nexus connecting the social, environmental, and economic systems, the concept of resource and environmental carrying capacity is crucial for assessing the interplay between human economic and social activities and the natural environment. The concept of carrying capacity is practiced in the study of population problems [4]. Later, on this basis, several concepts and corresponding studies, such as land resource carrying capacity, environmental carrying capacity, and resource environmental carrying capacity, were carried out [5]. Currently, numerous scholars have extensively researched and explored the concept of resource and environmental carrying capacity. In terms of research focus, this encompasses evaluating carrying capacity [6], establishing index systems [7], analyzing influencing factors [8], and assessing spatiotemporal dynamics [9]. Regarding the scope of study, it primarily spans across various levels, such as countries [10], provinces [11], cities [12], and plains [13]. As for research methodologies, prevalent approaches include ecological footprint analysis [14], the entropy weight TOPSIS model [15], system dynamics modeling [16], and the planetary boundary framework [17], among others.

Resources, the environment, and human activities constitute the triad of elements defining the carrying capacity of the resource–environment system [18], which are independent and interrelated open systems. Many studies have analyzed the interaction between human activities and the natural environment. For example, Pi studied the impact of human activities on local vegetation change in Guizhou Province [19]. Liu assessed human activities in China’s important ecological protection areas from the perspective of land-use change [20]. Huang analyzed the coupling change characteristics of human activity and landscape patterns in the Three Gorges Reservoir area and discussed the evolution of regional human–land relationships [21]. The scarcity of resources determines that there is an upper limit on resource consumption, and reasonable human activities can make full use of limited resources and further expand the carrying capacity of resources. For instance, in formulating the evaluation criteria for the water resources carrying capacity in the Yangtze River Economic Belt, Liu employed escalating demographic expansion and the pursuit of an elevated quality of life as indicators, reflecting the localized pressures on water resources [22]. Similarly, Xie took into account the impacts of inter-basin water transfers and water diversion projects, forecasting the prospective water resources carrying capacity in the Huang-Huai-Hai River Basin [23]. Xu evaluated the suitability of land resources and human activities on the Qinghai-Tibet Plateau based on the status quo on land use (construction land and cultivated land) [24]. Current research on the relationship between human activities and resource carrying capacity (RCC) primarily focuses on the singular impact of individual human activities on resources, such as land-use changes [25], artificially designated nature reserves [26], and population density [27], etc. Alternatively, when conducting assessments of RCC, researchers may consider incorporating indicators related to human activities. However, human activities themselves constitute a complex, multi-indicator, integrated process. Therefore, it is essential to comprehensively assess the overall impact of human activities through multi-indicator calculations, treating them as a holistic entity. This approach is crucial for a thorough analysis of the influence of human activities on RCC and the interactive relationships between the two, contributing to the rational and effective development and utilization of resources for sustainable development.

Building upon this foundation, taking the Poyang Lake region as the research scope, the human activity intensity (HAI) [28,29] and RCC in the study area, during 2010–2020, were calculated by using remote sensing images, gross domestic product (GDP), population density, and other data, and the spatial–temporal evolution characteristics of the two aspects were analyzed. Using bivariate spatial autocorrelation and other methods, the correlation between the HAI and the RCC and the interaction characteristics of each unit were analyzed. The study aims to contribute to the existing body of research on human activities, natural resources, and ecological environments in the Poyang Lake region of China. Simultaneously, it seeks to provide insights for similar regions, such as nature reserves and areas abundant in water resources, to assist in the formulation of relevant planning and policies.

2. Study Area and Data Source

2.1. Study Area

The region around Poyang Lake is located in the northeast of Jiangxi Province, China located at 113°57′–118°29′ E, 26°29′–30°06′ N, with the largest freshwater lake (Poyang Lake) in China as the core, and consists of six prefecture-level cities, including Nanchang city, Jiujiang city, Shangrao city, Fuzhou city, Yingtan city, and Jingdezhen city (Figure 1). The study area is vast, with a total area of approximately 70,687 square kilometers, including plains, hills, and mountains. The study area exhibits diverse terrain, characterized by higher elevations in the south and lower elevations in the north, with a central position occupied by the Poyang Lake Plain, a significant plain within the Yangtze River Basin. The region is flanked by mountains on three sides, adding to its distinctive geographical features. It has abundant land and water resources, superior agricultural production conditions, and abundant cultivated land, mineral resources, and natural and cultural tourism resources.

The region around Poyang Lake is an important economic development zone in Jiangxi Province, and it is also a demonstration area for ecological civilization construction, with remarkable advantages in terms of natural resources and the environment. Simultaneously, being a vital element in the development of the central region of China, its progress holds significant strategic importance. However, in recent years, environmental problems have begun to emerge in this region, such as the decline in wetland resources [30], the intensifying contradiction between the supply and demand for land resources [31], and the obvious decline in forest resources [32].

2.2. Data Source

Considering the availability of data, a relatively representative year is selected during the study period, so the data years involved are 2010, 2013, 2015, 2017, and 2020. Land-use data were derived from the Geospatial Data Cloud (http://www.gscloud.cn/, accessed on 1 December 2023), through the interpretation of the Landsat images selected for the study area. After preprocessing the data, a supervised classification was conducted to extract information on the land-use categories, including built-up land, cultivated land, aquaculture land, and forest land. The accuracy of the classification results was validated. Transportation data, such as data on the roads and railways for each year, were obtained from the CSDN website (https://www.csdn.net/, accessed on 1 December 2023). Population density spatial distribution data and GDP were acquired from the Chinese Academy of Sciences, Resource and Environmental Science Data Center (http://www.resdc.cn/, accessed on 1 December 2023). Water resource utilization intensity data were compiled from the Statistical Yearbook of Jiangxi Province and the Water Resources Bulletin of Jiangxi Province. The Digital Elevation Model (DEM) was obtained from the Geospatial Data Cloud (http://www.gscloud.cn/, accessed on 1 December 2023). Spatial distribution data on the soil texture, average annual precipitation, and average annual temperature for the years 2010, 2013, and 2015, originated from the Chinese Academy of Sciences, Resource and Environmental Science Data Center (http://www.resdc.cn/, accessed on 1 December 2023). Average precipitation and average temperature data for the years 2017 and 2020 were processed using monthly data from the National Earth System Science Data Center (http://www.geodata.cn/, accessed on 1 December 2023).

The spatial resolution of the involved data is consistently less than 1 km. Following the relevant literature, the study area was subdivided into unit grids of one square kilometer. Based on this grid, the data within the study area were resampled, ensuring uniform spatial resolution [33].

3. Research Methodology

3.1. HAI Model

Many factors contribute to measuring the impact of human activities. Drawing from the relevant literature on HAI, this study, based on the specific conditions of the Poyang Lake region, employs an existing evaluation index system for HAI and corresponding intensity values under each evaluation indicator [34], as outlined in

Table 1. Evaluation model of HAI.

Objective	Evaluation Indicators	Weight	Specific Indicators	Indicator Description
HAI	Land-use types (HAI1)	0.25	Built-up land	The value of HAI1 in built-up land area is 0.4.
			Arable land	The value of HAI1 in cultivated land area is 0.4.
			Aquaculture land	The value of HAI1 in aquaculture land area is 0.4.
			Other land-use types	The value of HAI1 in the area related to other land-use types is 0.
	Traffic data (HAI2)	0.25	Roadway	Within 500 m on both sides of a highway (including provincial roads and expressways), the value of HAI2 is 0.8.
			Railway	Within 500 m on both sides of the railway, the value of HAI2 is 0.8.
	Economic data (HAI3)	0.25	GDP	The value of HAI3 is determined through continuous assignment within the interval of 0.1–1, using the natural breakpoint grading method.
	Population data (HAI4)	0.25	Population density	When the population density is greater than 1000 km2/people, the value of HAI4 is 1. When the population density is less than 1000 km2/people, the value of HAI4 is calculated as 0.3333 × log (population density + 1).

. Additionally, to ensure model adaptability and independence, equal weights were assigned to four evaluation indicators, namely land-use types, GDP, population density, and road data. The comprehensive HAI index for the study area was then computed through a weighted overlay. The numerical results obtained through the model are used to quantify the intensity of the human activities within the region.

3.2. RCC Mode

The design of the evaluation index system for RCC should be informed by the specific conditions and goals of the study area. In the case of the Poyang Lake region, as a significant economic zone in Jiangxi Province, it is dedicated to establishing an ecological civilization construction demonstration area. When selecting evaluation indicators, careful consideration should be given to the distinctive development characteristics and resource advantages inherent to the region. The evaluation of RCC requires a comprehensive evaluation of the land, water, forest, climate, and other related resources in the study area. Based on the evaluation model of RCC from prior research [34], the analytic hierarchy process is utilized to ascertain the weights for each index. Subsequently, a weighted summation calculation is executed to derive the RCC value for the study area, as presented in

Table 2. Evaluation model for RCC.

Objective	Evaluation Indicators	Weight	Calculation Method	Indicator Description
RCC	Land-use intensity (RCC1)	0.448	Available-land area/total-land area	Land meeting the criteria of a slope less than 15 degrees, elevation below 1000 m, and soil texture classified as clayey loam is defined as usable land.
	Water-use intensity (RCC2)	0.218	The total scale of domestic and industrial water use/total scale of water resources	The actual water resource utilization is equal to the sum of residential water use, urban public water use, industrial water use, environmental water use, and agricultural water use. The total water resources equal the sum of surface water resources and groundwater resources minus redundancy.
	Forest coverage rate (RCC3)	0.117	Forest area/total-land area	The forest coverage rate is a crucial indicator reflecting the resources of an ecosystem in a given area. It enhances ecosystem functions, such as water conservation, regulation of runoff, soil retention, and water quality protection. It serves as a significant indicator for assessing RCC.
	Climatic potential productivity (RCC4)	0.217	Thornthwaite memorial model [35]	The calculation of this indicator involves annual average evapotranspiration, mean evaporation, annual precipitation, and annual average temperature.

.

3.3. Bivariate Spatial Autocorrelation Analysis

Spatial autocorrelation is a significant indicator that reflects the degree of correlation between a specific geographical phenomenon or its attribute value within a particular region and the same phenomenon or attribute value in adjacent regional units [36]. To delve deeper into the spatial correlation characteristics among diverse variables, Anselin introduced a bivariate spatial autocorrelation analysis method [37]. In contrast to previous emphasis on individual variables, bivariate spatial autocorrelation allows for the characterization of spatial relationships among various geographical elements and the assessment of correlation levels between a positional variable and others [38]. In this study, the bivariate spatial autocorrelation tool in the GeoDa1.18.0 is utilized to analyze the spatial correlation between HAI and RCC.

Bivariate spatial autocorrelation includes global spatial autocorrelation and local spatial autocorrelation. Global spatial autocorrelation uses Moran’s I index to reflect the overall degree of spatial correlation among elements. When I is greater than zero, it indicates a positive spatial correlation between the elements, with a larger I value indicating a more pronounced spatial correlation. When I is less than zero, it indicates a negative spatial correlation, with smaller I values suggesting greater spatial difference. When I equals zero, the spatial correlation of the elements presents randomness. The calculation formula for Moran’s I [33] is as follows:

$${{\displaystyle I}}_{\mathrm{hr}}=\frac{n{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{\mathrm{j}=1}^n{W}_{ij}\left(\frac{y_{i,h}-\overline{y_h}}{{\sigma }_h}\right)\left(\frac{y_{i,c}-\overline{y_c}}{{\sigma }_c}\right)}}}{\left(n-1\right){\displaystyle \sum _{i=1}^n{\displaystyle \sum _{\mathrm{j}=1}^n{W}_{ij}}}}$$

(1)

where I_hr denotes the bivariate global autocorrelation coefficient of HAI and RCC, y_i,h denotes the value of HAI for the i-th assessment grid, y_i,c denotes the value of RCC for the i-th assessment grid, σ_h denotes the variance of HAI for all the evaluation grids, and σ_c denotes the variance of HAI for all the evaluation grids.

After analyzing the overall correlation degree among the elements, local spatial autocorrelation is utilized to further investigate the clustering level of similar attributes between a specific spatial unit and its surrounding area [39]. The local indicators of spatial association (LISA) is a tool for analyzing hotspots and spatial clustering or dispersion patterns in geographical data, providing a spatial statistical analysis tool to reveal regional correlations. It is commonly applied to study the distribution characteristics of economic, social, or environmental indicators in geographical space. The spatial agglomeration can be visualized using a LISA map to identify the specific locations of spatial agglomeration [40]. This helps analyze the local feature differences in the spatial distribution of the two elements.

4. Results and Analysis

4.1. Analysis of Spatiotemporal Variations in HAI

The study area was divided into evaluation units with an area of one square kilometer. The values on the HAI in the study area were resampled using the evaluation units, and the resulting data were then used to analyze the spatial and temporal characteristics of the HAI. Considering the specific conditions of the study area and referring to relevant existing research [36], the values on the HAI were divided into five levels: low level (HAI ≤ 0.2), relatively low level (0.2 < HAI ≤ 0.4), medium level (0.4 < HAI ≤ 0.6), relatively high level (0.6 < HAI ≤ 0.8), and high level (HAI > 0.8). Figure 2 illustrates the spatial distribution of the HAI for each period from 2010 to 2020.

From Figure 2, the overall distribution pattern of the HAI in the study area shows that the HAI levels are higher in the central region and lower in the surrounding areas. The areas with a high or relatively high level of HAI are mainly distributed near the urban area of Nanchang, relying on major transportation arteries to spread to the surrounding areas. The areas with a low and relatively low level of HAI are mainly located in the surrounding regions and the Poyang Lake Nature Reserve. The surrounding areas of the study region are mostly hilly and mountainous, which inherently have poorer resource conditions and lag behind in regard to economic development compared to the central plain areas. The nature reserves established in the Poyang Lake Basin are an integral part of China’s nature reserve system, with a strong focus on environmental protection and effective management. Hence, the HAI in these two types of areas is comparatively low. The spatial distribution of the HAI aligns with the natural and economic geographical spatial differentiation features of the Poyang Lake region.

The variation in the HAI during the period from 2010 to 2020 is significant. (1) All levels of the HAI show a tendency of fragmentation distribution. From 2010 to 2015, the spatial distribution of the HAI is relatively cohesive and continuous; however, it begins to break and exhibit a discontinuous distribution in 2017. In 2020, areas of all levels exhibit small and fragmented distributions, especially the medium level areas which are uniformly distributed within the study area. The fragmented distribution of the HAI levels indicates that with the passage of time, human activities are no longer concentrated in a single center, but are gradually expanding their influence across various areas. Consequently, there are no distinct boundaries between the different intensity levels. (2) In 2010, the HAI is mainly dominated by low and relatively low levels, accounting for over 90% of the total study area. As time progressed, the proportion of these two levels gradually decreased. During the study period, the proportion of areas with a medium, relatively high, and high level of HAI show an increasing trend. Among them, the proportion of medium level areas shows the largest increase, rising from 3.58% to 31.46% over the course of 10 years. The intensity of human activity generally exhibits an increasing trend year by year, but it does not entail an absolute transition from low HAI levels to high levels. During the study period, the area of each HAI level shows unsteady increases and decreases and undergoes transformations among them.

4.2. Analysis of Spatiotemporal Variations in RCC

The evaluation unit was also used for resampling the RCC in the study area. Based on the actual conditions of the study area and the data characteristics for each period, the RCC was categorized into five levels: low level (RCC ≤ 0.3), relatively low level (0.3 < RCC ≤ 0.4), medium level (0.4 < RCC ≤ 0.5), relatively high level (0.5 < RCC ≤ 0.7), and high level (RCC > 0.7). Figure 3 shows the spatial distribution of the RCC for each period from 2010 to 2020.

The spatial distribution of the RCC in the study area exhibits a pattern of a “higher level in the central region, lower level in the surrounding areas”. The central region, relying on the Poyang Lake Plain, boasts abundant water resources, fertile soil, and flat terrain, making it suitable for agricultural and economic development. Given its favorable intrinsic conditions, the region exhibits a relatively high RCC, as evident in areas such as Donghu District and Nanchang County. The western, eastern, and southern regions are relatively difficult to develop because of their rugged terrain. In addition, most of these areas are mountains and hills, resulting in limited availability of land and water resources. Therefore, the RCC in this part of the region is generally low, such as Wuyuan County, Wuning County, Zixi County, and other regions.

The RCC shows no significant change from 2010 to 2020 in terms of time. Figure 3 illustrates a stable spatial distribution of the RCC across different levels in the region, showing no distinct migrations or changes. Of note, the area of the RCC at the medium level demonstrates a steady annual increase, gradually expanding from the central plain area towards the surrounding regions. The RCC has remained stable over the observed time period, indicating a favorable condition in the study area, without signs of environmental degradation due to excessive resource consumption.

4.3. The Correlation between HAI and RCC

4.3.1. Spatial Correlation

Utilizing bivariate spatial autocorrelation, the spatial correlation between the HAI and the RCC during the period from 2010 to 2020 was calculated. The results (Figure 4) show that the value of Moran’s I for both variables during the period from 2010 to 2020 are 0.631, 0.741, 0.576, 0.335, and 0.485, respectively (p < 0.05). This indicates a certain positive spatial correlation between the HAI and the RCC, suggesting a clustering tendency in space. The majority of data points in Figure 4 are concentrated in the first quadrant, indicating a notable clustering of higher levels of HAI with higher levels of RCC.

The spatial aggregation areas of the HAI and the RCC were further explored using the bivariate local spatial autocorrelation method based on z-tests (p = 0.05). Figure 5 represents the local indicators of spatial association (LISA) clustering results. From the LISA clustering results, it can be observed that areas with significant local associations between the HAI and the RCC are mainly located in Nanchang city and the surrounding areas. The LISA clustering patterns for the two variables can be classified into five types: high–high (HH), low–low (LL), low–high (LH), high–low (HL), and non-significant (NS).

The HH clustering refers to areas where the central region has high levels of HAI, and the surrounding areas also have high levels of RCC. These areas are mainly distributed in the Poyang Lake Basin, such as the urban areas of Nanchang and Jiujiang. Such regions possess favorable resource conditions to support social development, attracting a significant population concentration, resulting in higher levels of HAI. Human activities have a positive impact on resource consumption to a certain extent, as they can enhance resource utilization efficiency. The dynamic interplay between the two establishes a stable state termed “human activity-resource carrying,” wherein regions with high HAI also demonstrate elevated RCC.

The LL clustering refers to areas where the central region has low levels of HAI, and the surrounding areas also have low levels of RCC. The LL clustering is mainly distributed in the peripheral areas of the study area, such as the southwest of Jiujiang city, Jingdezhen city, the eastern part of Shangrao city, and the southern part of Fuzhou city. These regions, despite having abundant forest resources, are not suitable due to unfavorable terrain, soil texture, and lower climate productivity potential, resulting in low HAI.

The LH clustering refers to areas where the central region has low levels of HAI, but the surrounding areas also have high levels of RCC. The LH clustering is mainly situated in proximity to the HH clustering. These regions possess favorable resource conditions, but exhibit a marginally lower level of economic development compared to the HH clustering, consequently resulting in a slightly reduced HAI. These areas demonstrate a lesser degree of resource utilization, implying substantial untapped development potential.

The HL clustering refers to areas where the central region has high levels of HAI, but the surrounding areas also have low levels of RCC. The HL clustering is primarily distributed in the eastern and northwestern parts of the study area. These regions possess relatively poor inherent resource conditions. However, efficient and intensive resource utilization enables them to accommodate higher levels of HAI. The proportionate area of the HL clustering demonstrates a growing trend, implying an enhancement in resource utilization efficiency within the study area.

The NS clustering refers to areas where there is no significant clustering relationship between the HAI and the RCC within the evaluation grid units. These areas account for approximately 50% of the total area and are primarily distributed in the central part of the Poyang Lake Basin. Considering the earlier analyses, the HAI in the NS clustering areas is generally at a moderate to low level, and the RCC is at a medium level. The values and spatial distribution of both indicators are relatively even in these areas, without distinct high-value or low-value centers.

4.3.2. Temporal Correlation

In view of the practical scenario, the influence of human activities on resource consumption shows a discernible lag effect. To delve deeper, it is essential to statistically analyze the lagged correlation between the HAI and the RCC. Therefore, an analysis is conducted to measure the influence of the HAI at the beginning of each period on the RCC during that specific period. The results are presented in

Table 3. The impact of different levels of HAI on RCC from 2010 to 2020.

HAI Level at the Beginning of the Period	The Number of Assessment Units Where RCC Levels Have Changed.
	2010–2013			2013–2015			2015–2017			2017–2020			Total
	Increase	No change	Decrease	Increase	No change	Decrease	Increase	No change	Decrease	Increase	No change	Decrease	Increase	No change	Decrease
high level	2	29	0	0	6	9	78	86	10	0	43	0	80	164	19
relatively high level	60	244	23	22	334	180	308	707	61	30	490	6	420	1775	270
medium level	357	1950	110	682	4693	1759	552	3660	236	441	5699	117	2032	16,002	2222
relatively low level	3168	29,712	1463	6120	40,641	7895	4774	20,899	1859	2832	42,345	766	16,894	133,597	11,983
low level	808	38,046	1988	1069	11,611	2939	8336	34,759	1635	1699	23,032	460	11,912	107,448	7022

.

According to Table 3, it can be observed that the RCC generally remains stable or increases across different levels of HAI. Under the influence of the HAI, the number of evaluation units where the level of RCC remains unchanged is over 62%, and the number of units where it increases is generally greater than the number of units where it decreases. This indicates that resource consumption in the study area is not exceeding the capacity, demonstrating the ability to accommodate an increase in the HAI. The impact on the RCC varies with different levels of HAI. In total, under the influence of the medium level of HAI, the level of RCC shows that the number of evaluation units that increased is less than the number that decreased. This indicates that a medium level of HAI leads to higher resource consumption, coupled with relatively low resource conversion efficiency, resulting in a reduction in RCC. Furthermore, the impact of human activities on RCC demonstrates discernible stages and temporal fluctuations. The RCC shows an increasing trend from 2010 to 2013. However, from 2013 to 2015, there is an abnormal decline in the RCC, where the number of levels decreasing exceeded the number of levels increasing. Starting in 2015, the number of units with an increase in the RCC begins to surpass the number of units with a decrease, indicating a restoration of the increasing trend. The occurrence of this situation may be related to the implementation of the Ecological City Cluster Planning of Poyang Lake (2015-2030) in Jiangxi Province. This plan clearly emphasizes achieving overall green transformation and uses the sustained improvement of the ecological environment as a criterion for regional development. It aims to establish a livable urban spatial layout characterized by “ecological pattern assurance, pleasant living environment, accessible low-carbon transportation, and integrated ecological facilities” [41]. With the effective implementation of this plan, the allocation of resources in the research area has been optimized, alleviating the impact of human activities on resource consumption, and promoting the coordinated development of resources, the environment, and socioeconomics.

4.3.3. Mutual Influence

From the analysis above, it can be observed that there is a mutual influence between the resource carrying capacity and human activities within the study area. This aligns with the evaluation foundation of carrying capacity from the resource intensification perspective: at a macro scale, the region’s inherent resource conditions determine its carrying capacity; at a micro scale, the movement of resources and elements caused by human activities can influence the carrying capacity [18]. To further investigate the interaction between the HAI and the RCC, this study takes the LISA clustering results from the HAI and the RCC in 2020 as an example (Figure 6) and tabulates the number of units for various combination patterns (

Table 4. Spatial autocorrelation patterns of LISA clustering for HAI and RCC.

Categories of LISA Clustering for HAI.	Categories of LISA Clustering for RCC.
	2010				2013				2015				2017				2020
	HH	LL	LH	HL	HH	LL	LH	HL	HH	LL	LH	HL	HH	LL	LH	HL	HH	LL	LH	HL
HH	9328	993	6	-	9392	225	2	-	6756	231	-	-	4430	3551	2	-	7797	3828	6	-
LL	12	16,875	-	2	286	20,786	-	1	-	20,263	-	-	1377	13,172	-	3	649	10,490	-	4
LH	8	5	-	-	6	2	-	-	15	61	-	-	5	15	-	-	1	3	-	-
HL	1	100	-	-	-	16	-	-	-	298	-	-	1	32	-	-	1	4	-	-

).

Table 4 and Figure 6 show the 10 combination patterns found in the study area, with the highest number of units observed in patterns such as “[LL]-[LL]” and “[HH]-[HH]”. The surrounding areas of the study region mostly exhibit the “[LL]-[LL]” combination pattern. This pattern signifies that the HAI and RCC are low in the central area, and both are also low in the surrounding regions. The regions characterized by the “[LL]-[LL]” combination pattern struggle to utilize the resources from the surrounding areas. However, although this type of combination has the highest proportion, its quantity has decreased over time, indicating an overall improvement in the development of the study area. In the central region of the study area, such as the urban area of Nanchang, Nanchang County, and Jinxian County, the predominant combination pattern is “[HH]-[HH]”. This pattern signifies a state of “balanced supply”, where the region’s internal resource provision and the resource consumption caused by human activity have reached an equilibrium. The areas characterized by the “[HH]-[HH]” combination pattern represent well-developed and healthy regions. The “[HH]-[LL]” combination pattern is primarily found in Jiujiang city and Shangrao city. Despite the relatively poor RCC in both the central and peripheral areas, the flow of resources and elements brought about by human activities allows these regions to accommodate higher intensities of human activity. In contrast, the “[LL]-[HH]” combination pattern is mainly found in areas near the Poyang Lake Basin, such as Yujiang County and the northern part of Yugan County. Despite the superior internal resource conditions in the central area and its surroundings, the intensity of human activity remains low due to the constraints of the Poyang Lake Basin Nature Reserve.

There are relatively few combination patterns in the study area involving the “outflow” and “inflow” of regional elements. For example, the “[HH]-[LH]” combination pattern, where the regions in this pattern utilize the “inflow” of elements from the surrounding areas to accommodate higher levels of HAI in the central area, shows a low prevalence over the 10-year period. The highest proportion of this combination pattern is merely 0.03% of the effective combination quantity for that year. Furthermore, combination models exhibiting an “outflow” trend, such as “[HH]-[HL]” and “[HL]-[HL],” have a count of zero in the study area, indicating a minimal “outflow” of elements. Therefore, overall, the “outflow” and “inflow” of elements between various regions in the study area do not manifest prominently. The region primarily relies on the efficient utilization of its own resources.

5. Conclusions and Discussion

5.1. Conclusions

By exploring the spatiotemporal characteristics of the HAI and the RCC in the Poyang Lake area and their correlation, the following conclusions are drawn:

(1)

From 2010 to 2020, there was a significant increase in the HAI in the Poyang Lake area, showing a trend of continuous growth year by year.

(2)

During the period of 2010 to 2020, there was no significant change observed in the RCC in the Poyang Lake area, and there was no indications of environmental degradation due to excessive resource consumption.

(3)

There is a certain positive correlation between the HAI and the RCC for both spatial and temporal dimensions. Spatially, there was a positive spatial correlation between the two, with variations in their correlation. Temporally, under different levels of HAI, the RCC primarily showed stability and growth. The influence of the HAI on the RCC varied across different levels and time periods.

(4)

There is an interactive relationship between the HAI and the RCC. The inherent resource conditions of a region determine its carrying capacity limit. However, factors such as element flow and resource intensification resulting from human activities positively influence the carrying capacity of the resource and environmental units to some extent. In this study area, the primary aspect is the efficient utilization of resources by each assessment unit, and inter-regional element flow is not pronounced.

5.2. Discussion

The assessment of RCC is of great scientific value and practical significance for addressing global resource and environmental challenges and advancing the construction of ecological civilization. It stands as a crucial foundation for the scientific formulation of national land spatial planning. However, evaluating RCC is a complex system fraught with numerous uncertainties, impacting the practical application of findings due to low repeatability. This study, which examines human activities and resource consumption in China’s Poyang Lake region from 2010 to 2020, depicts the present state of resource consumption and utilization influenced by human activities. The chosen study area, marked by diverse topography and abundant resources, including large lakes and numerous nature reserves, allows the applied evaluation model to be highly applicable to most other regions. It is worth noting that the intensive use of resources resulting from human activities and the flow of resource elements between regions are of significant importance. Taking the moderate-intensity regions in regard to HAI as an example, during the period from 2010 to 2020, this type of area experienced the largest increase in area, severe spatial fragmentation, yet a decreasing trend in RCC. This suggests that regions with moderate intensity in regard to HAI during this period did not exhibit high resource utilization, indicating significant development potential for the future. Attention should be focused on such areas.

However, it is important to note that this study has certain limitations. The analysis of the correlation between HAI and RCC is based on the interaction between the results of these two evaluation factors, which may not comprehensively capture the intrinsic connection between HAI and RCC. Future research could delve more deeply into understanding the extent of the impact of human activities on RCC. Additionally, the chosen study duration in this paper spans only ten years. To enhance the scientific rigor and comprehensiveness of the study findings, subsequent research efforts could consider expanding the temporal scope to thoroughly investigate the long-term variations in HAI and RCC.