Spatial Differentiation and Driving Mechanisms of Ecosystem Service Value Change in Rural Land Consolidation: Evidence from Hubei, China

Abstract

Rural land consolidation projects (RLCPs) have become one of the largest organized human activities to change land use patterns and impact terrestrial ecosystems, and it may also be an important precondition to improving ecosystem service value (ESV). Evaluating the change in ecosystem service value (ESV) is an important basis for measuring the effectiveness of RLCPs. Therefore, this paper, taking RLCPs implemented at County Level in Hubei Province, China, as an example, uses the improved ESV evaluation model to analyze the spatial differentiation of ESV change in RLCPs and then adopts geographic detectors and a geographically weighted regression model to identify the dominant factors affecting the ESV change in RLCPs. The results showed that (1) although RLCPs make the unevenness of land use obvious, they reduce the complexity of land use evidently and improve the dominance of land use significantly; (2) The ESV of RLCPs in 71 counties of Hubei Province increased, with an average increase of USD 2.37 × 10⁷ a⁻¹. The ESV increase is large in central Hubei, while small in eastern and western Hubei. However, the increase rate of ESV is high in eastern and central-north Hubei, while low in western and central-south Hubei. This indicates that RLCPs can effectively promote ESV, but there are significant regional differences, and (3) the ESV increase is positively correlated with GDP and construction scale, but negatively linked with investment and per capita income of rural residents. The ESV increase rate is negatively associated with cultivated land proportion and land use diversification index change, but it is positively related to the change in the land use evenness index. However, their driving effects have significant spatial heterogeneity.

1. Introduction

Ecosystem services refer to natural environmental conditions and their utility that are formed and maintained by ecosystems and their processes [1], including support services, regulation services, supply services, and cultural services [2]. With the rapid increase in population and development of the economy, ecosystem services are constantly being utilized by human beings. The Millennium Ecosystem Assessment points out that more than 60% of the global ecosystem services are declining, and the trend is expected to continue indefinitely [1,2]. As an important basis for natural asset accounting and ecological environment protection, the quantitative assessment of ecosystem service value (ESV) can be more persuasive in terms of the ecological benefits obtained by adhering to high-quality economic development and enhancing public awareness of environmental protection [3,4]. At present, the ESV evaluation is mainly based on three aspects. The first aspect is evaluating the comprehensive value of ecosystem services at different spatial scales, i.e., global, regional, and local [5,6,7,8,9]. The second aspect is assessing the comprehensive ESV of different ecosystem types, i.e., cultivated land, forest land, grassland, wetland, and unused land [10,11,12,13,14]. The third is evaluating the value of individual ecosystem services, such as primary product production, gas regulation, soil and water conservation, and biodiversity maintenance [14,15,16,17,18,19]. However, it should be emphasized that the purpose of ESV assessment is not to put an accurate and absolute price tag on ecosystem services, but to allow marginal changes in ESV to be reflected in the decision-making process on ecosystem conservation and land management [20,21,22]. The evaluation accuracy of ESV has not been effectively solved in previous studies, which is why future research should focus on analyzing the marginal change in ESV brought about by public policies or project implementation in order to weaken the error of absolute value to a certain extent, and provide policy implications for continued and effective implementation of public policies or projects [11,12].

Rural land consolidation projects (RLCPs), as sustainability-oriented projects, have been implemented worldwide to promote agricultural production and achieve rural development [23,24]. With the strategic adjustment of land use policy in China, RLCPs have acquired richer connotations and functions, which enabled comprehensive RLCPs. The RLCPs required by the Ministry of Natural Resources of the People’s Republic of China refer to a governance activity that combines the comprehensive application of relevant policies and the adoption of advanced engineering technologies to adjust the land use structure and optimize the spatial layout of the land to ensure sustainable land use, ultimately meeting people’s production, life, and ecological needs, including comprehensive effects such as food security, modern agriculture, precision poverty alleviation, and ecological restoration. Combining the above definition with the actual situation of RLCPs in China, In terms of RLCP objects, besides increasing the area and improving the quality of cultivated land, increasing the income of farmers, accelerating the internal adjustment of agriculture, promoting the migration of labor force, and improving the living environment in rural areas are also involved [25,26,27]. In terms of the contents of RLCPs, they are projects that mainly involve the implementation of agricultural land consolidation, construction land consolidation, area environmental consolidation, ecological restoration, and historical and cultural protection [28,29,30]. Among them, agricultural land consolidation refers to supplementing the quantity of cultivated land, improving the quality of cultivated land, and improving land utilization efficiency through land consolidation, reclamation, and development. Construction land consolidation refers to idle and inefficient construction land consolidation to meet the dual goals of regional development land demand and cultivated land protection. Ecological protection and restoration refer to a comprehensive improvement in the rural environment to improve rural ecological functions and maintain biodiversity, improving the ability to prevent natural disasters, and maintaining the rural natural landscape. Historical and cultural protection in RLCPs aim to enable rural areas to achieve industrial development led by distinctive rural culture, which requires the protection and utilization of rural natural and cultural resources, and the preservation of rural unique local culture.

RLCPs have become one of the largest organized human activities to change land use patterns and impact terrestrial ecosystems [31,32,33,34]. RLCPs will inevitably have direct or indirect positive or negative impacts on hydrology, soil, vegetation, atmosphere, organisms, and other environmental conditions in the project and surrounding areas [33,34,35,36]. The study of the ecological impact of RLCPs has become an important concern in the evaluation of the benefits of land consolidation projects. Especially in China, quantitatively evaluating the economic effects of RLCPs by analyzing ESV changes has become a major concern for many scholars to provide a reference for policymakers in conducting post-RLCP benefit evaluations. For example, Lu et al. used an improved equivalent factor method to reveal the ESV profit and loss patterns of land consolidation projects with different properties and landforms, and effectively quantitatively evaluated the ecological environment changes caused by the implementation of land consolidation projects in villages in plains, hills, and mountainous areas of Hubei Province, China [19]. Liu et al. used GIS and landscape indicators to evaluate the ecological effects of RLCPs in typical RLCP locations [37]. Zhou et al. evaluated the ecological benefits of an RLCP at a township level [38]. However, the existing research mainly focuses on small areas such as villages and towns [19,37,38] and pays less attention to the county level. Specifically, these studies ignored the characteristics of China’s RLCPs in terms of regions, types, goals, and modes, leading to an inaccurate assessment of the ecological effects of RLCPs. On the other hand, the driving mechanisms in ESV change in RLCPs at the county level remain unexplored. Taking Hubei Province, one of the pioneering zones for RLCP implementation in China, as a case study, we use the modified ESV evaluation model to analyze the spatial differentiation of ESV change in RLCPs. Then, we adopt a geographic detector and a geographically weighted regression model to identify the dominant factors affecting the ESV change in RLCPs, which is suitable for promoting the ecological transformation of RLCPs. Additionally, it also provides a scientific basis and support for strengthening the construction of ecological civilization and protecting the ecological environment.

2. Materials and Methods

2.1. The Effects of RLCPs on ESV

Based on the theory of ecosystem services, RLCPs will promote changes in ESV by changing land use patterns and causing ecosystem changes that affect support services, regulation services, supply services, and cultural services of the ecosystem.

Implementing RLCPs can have an impact on ecosystem support services. RLCPs supplement the quantity and improve the quality of cultivated land by implementing land consolidation, land reclamation, and land development. This has increased food production capacity. The result of improving food production capacity means that RLCPs can improve ecosystem support services.

The implementation of RLCPs has an impact on both ecosystem regulation and supply services. RLCPs affect soil physicochemical properties and hydrological processes such as surface runoff, soil infiltration, and deep infiltration by changing land use/cover types. These changes in soil habitats have an impact on the above ecosystem services in the area where RLCPs are implemented, specifically manifested as direct impacts on crop production, climate challenges, soil and water conservation, and other aspects.

The implementation of RLCPs will also have an impact on ecosystem cultural services. In practice, RLCPs focus on the improvement and protection of rural landscapes, manifested as promoting the development of industries such as ecotourism while emphasizing the protection of existing rural landscapes and culture. From this, the role of RLCPs in improving ecosystem cultural services is becoming increasingly evident.

Above all, this paper proposes that while RLCPs have an effect on the ecosystem service, there is an ESV change in RLCPs.

2.2. Study Area

Hubei Province, located in central China and the middle reaches of the Yangtze River, is one of the major grain-producing areas in China. It extends over an area of 18.59 × 10⁴ km², equivalent to 1.94% of the national territory, of which mountains account for 55.5%, hills and hillocks comprise 24.5%, and plains and lakes encompass 20%. Hubei is one of the provinces with the earliest start, numerous projects, and remarkable effects in RLCPs in China, and it is also one of the pioneers and demonstration provinces for RLCP implementation mode innovation. The RLCPs in Hubei Province is divided into 3 first-level types: low mountain and hilly consolidation type areas in eastern Hubei (I), plain consolidation type areas in central Hubei (II), and mountain plateau consolidation type area in western Hubei (III); it can be further subdivided into 7 secondary modes: low mountain consolidation mode areas (I-1), hilly consolidation mode areas (I-2), hillock consolidation mode areas (II-1), plain consolidation mode areas (II-2), water network polder consolidation mode areas (II-3), river (ditch) valley basin consolidation mode areas (III-1), and karst dam consolidation mode areas (III-2). According to the Department of the Natural Resources of Hubei Province, Hubei carried out 406 key RLCPs in 2013, with a cumulative construction scale of 461,190 hectares, a total investment of USD 2041.16 million, and 11,069 hectares of newly cultivated land, which were completed by the end of 2016. For the convenience of research, we have merged the municipal districts and the counties that have not implemented land consolidation projects, and we obtained a total of 71 county-level units. The serial numbers of each research unit and its consolidation mode are shown in Figure 1. Considering that the construction period of RLCPs takes half a year to three years, we chose 2012 and 2017 to represent the year before and after RLCPs, respectively.

2.3. Research Methods

2.3.1. Indices of Land Use Structure

A change in land use structure will change ecosystem function and structure, leading to a change in ESV. The land use diversification index (LDI), land use diversity index (LVI), land use dominance index (LAI), and land use evenness index (LEI) are widely used to characterize land use structure from a macro perspective [39,40]. Both LDI and LVI indicate the complexity and richness of land use types, and the higher the value, the more complex the land use types. LAI refers to the difference between the maximum and actual LVI values. LEI is defined as the uneven distribution degree of various land use type areas. Their formulas are as follows.

$$LDI=1-\sum S_i^2/\sum S_i{}^2$$

(1)

$$LVI=-\sum P_{i}ln{P}_i$$

(2)

$$LAI=\mathit{ln}m+\sum P_{i}ln{P}_i$$

(3)

$$LEI=-\sum P_{i}ln{P}_i/\mathit{ln}m$$

(4)

where S_i and P_i represent the area and proportion of the i-th land use type, respectively; m represents the maximum area of each type of land use.

2.3.2. ESV Calculation

The equivalent factor method, derived from the unit value-based approach, has been widely applied for the calculation of the comprehensive ESV. Xie et al. developed a method based on a survey of 700 ecological experts to estimate the ESV in China [41,42,43]. This method is convenient and cost-effective for performing a comprehensive assessment of ESV. However, a growing body of research reveals that ecosystem service functions are regulated by different ecological mechanisms and processes, which are closely related to local ecological conditions [41]. Thus, assessment based on regional uniform equivalent factors cannot reflect regional differences in ecosystem services functions, thus limiting the practical application of ESV assessment in environmental management. To address these limitations, Xie et al. [41,42,43] updated equivalent coefficients of ecosystem services in China based on a combination of methods. This study obtained the table of equivalent ecosystem services coefficients in Hubei Province [41,42,43]. The specific steps are as follows.

First, we selected the indicators of average precipitation, food production, and social development stage to adjust the equivalent coefficients of ecosystem services [19,43,44], by which the adjusted equivalent coefficients can reflect the regional variations of ecological conditions in Hubei Province. The formulas are as follows:

$${\lambda }_t=G_t/G_{0t}=W_t/W_{0t}$$

(5)

where λ_t is the regional correction coefficient in year t; G_t and G_0t are the average grain yield (kg/hm²) of Hubei Province and China in year t, respectively; and W_t and W_0t are the average annual precipitation (mm) of Hubei Province and China in year t, respectively.

$$l_t=H_{t}L/\left(1+e^{-\left(1/E_{nt}-3\right)}\right)+h_{t}L/\left(1+e^{-\left(1/E_{0nt}-3\right)}\right)$$

(6)

$$T_t=l_{1t}/l_{0t}$$

(7)

$$P_t=GD{P}_{1t}/GD{P}_{0t}$$

(8)

where l_t is the coefficient of the social development stage in year t; T_t is the coefficient of regional willingness to pay for ecological environmental protection in year t; P_t is the coefficient of regional payment capacity in year t; L is the value of the social development coefficient in the ideal stage; H_t and h_t are the proportion of the urban and rural population in the total population in year t, respectively; E_nt and E_0nt are Engel’s coefficients of urban and rural areas in year t, respectively; l_1t and l₀ are the coefficients of the social development stage in Hubei Province and China in year t, respectively; and GDP_1t and GDP_0t are the GDP per capita of Hubei Province and China in year t, respectively.

Second, Xie et al. defined that the ESV of the standard equivalent factor is equal to 1/7 of the food production value that farmland can provide [19,41,42,43]. We select the sowing area and output value of rice, wheat, and corn in Hubei Province to adjust the economic value of one equivalent factor. The formulas are as follows:

$$V_t=\frac{1}{7}{{\displaystyle \sum }}_{m=0}^n\left(\frac{A_{mt}{P}_{mt}{Q}_{mt}}{M_t}\right)$$

(9)

$$V{C}_{ijt}={\lambda }_t\times T_t\times P_t\times V_t\times E_{0ijt}$$

(10)

where V_t is the economic value of one equivalent factor in year t (USD/hm² a⁻¹); VC_ijt is the economic value of one equivalent factor of the i-th ecosystem type (or land use type) and the j-th service type in year t; A_mt, P_mt, and Q_mt are the sown area, selling price, and total output of the m-th crops in year t, respectively; and M_t is the total sown area of three crops in year t.

Third, Xie et al. [43] divided the ecosystem types or land use types into 6 categories in the equivalent table of terrestrial ecosystem services in China, which differs from the land use type in the project area, so it is necessary to readjust them. Specifically, ESV of cultivated land is the average value of water fields and dry land; ESV of garden land is weighted with 80% of the corresponding value of forest land and 20% of the corresponding value of grassland; ESV of forest land is the average value of broad-leaved forests and shrub forests; ESV of grass land is the average value of grasslands, shrubs, and meadows; ESV of traffic land is replaced by the corresponding value of bare land; other land types include field ridge, sandy land, and bare land, and their ESV is replaced by the corresponding value of bare land; and the ESV of construction land is zero.

Forth, based on the modified equivalent coefficients of ecosystem services and areas of different land use types, the ESV formulas are as follows.

$$ES{V}_0={{\displaystyle \sum }}_i{{\displaystyle \sum }}_j{A}_{0i}\times V{C}_{0ij}$$

(11)

$$ES{V}_1={{\displaystyle \sum }}_i{{\displaystyle \sum }}_j{A}_{1i}\times V{C}_{1ij}$$

(12)

$$ES{V}_{gl}=ES{V}_1-ES{V}_0$$

(13)

$$ES{V}_{gl1}=ES{V}_1-ES{V}_0/ES{V}_0\times 100$$

(14)

where ESV₀ and ESV₁ are the total ESV before and after RLCP; A_0i and A_1i are the areas of the i-th ecosystem type before and after RLCP; VC_0ij and VC_1ij refer to the value coefficient of the i-th ecosystem type and the j-th service type before and after RLCP; and ESV_gl and ESV_gl1 represent the amount and rate of change in ESV, respectively.

2.3.3. Spatial Autocorrelation

Spatial autocorrelation is commonly used to calculate the correlation degree of spatially dependent or heterogeneous data and interpret their spatial mechanism [45]. It can be divided into global spatial autocorrelation and local spatial autocorrelation [45]. The former can be used to investigate the overall spatial correlation and differences in ESV change, as expressed by Moran’s I statistic [46]. The latter can be used to further explain the spatially non-stationary and heterogeneous characteristics of ESV change, which can be described by local Moran’s I [46]. The formulas are as follows:

$$I=n\frac{{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{j=1}^n{W}_{ij}(x_i-\overline{x})\left(x_j-\overline{x}\right)}{{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{j=1}^n{W}_{ij}{(x_i-\overline{x})}^2}$$

(15)

$$I_i=\frac{n\left(x_i-\overline{x}\right){{\displaystyle \sum }}_{j=1}^n(x_j-\overline{x})}{{{\displaystyle \sum }}_{j=1}^n{(x_j-\overline{x})}^2}$$

(16)

where I and I_i are Moran’s I and local Moran’s I indices, respectively; n is the number of research units; x_i and x_j are the attribute values of spatial units I and j, respectively; x is the average value of x_i; and W_ij is the spatial weight matrix.

2.3.4. Geographic Detector and Geographically Weighted Regression Model

The geographic detector is widely used to detect spatial heterogeneity and discover its driving factors, and it is based on the assumption that if the independent variable has a significant effect on the dependent variable, the spatial distribution of the independent and dependent variables should be similar or even highly consistent [47,48]. Therefore, we use the geographic detector to determine the dominant factors affecting the ESV change in RLCPs. The expression is as follows:

$$q=1-\frac{1}{N{\sigma }^2}{{\displaystyle \sum }}_{h=1}^L{N}_h{\sigma }_H^2$$

(17)

where q is the explanatory power of explanatory variables for ESV change or change rate, whose values range from 0 to 1; h = 1, ..., L is the stratification of variable Y or factor X; N_h and N are the numbers of cells in stratum h and full area, respectively; and ${\sigma }_h^2$ and ${\sigma }^2$ are the variances of Y values in stratum h and full area, respectively.

The geographic detector is a method of global spatial analysis, in which the regression parameters are the same in different geographic locations. However, the regression parameters differ in different geographical locations [49]. If we would only adopt the geographic detector, the estimated parameters would be the average values of the regression parameters in the whole study area, which cannot reflect the real spatial characteristics of the regression parameters of the driving factors that affect ESV change in RLCPs. To solve this problem, Fortheringham et al. presented a geographically weighted regression (GWR) model, which is an extension of the ordinary linear square model and incorporates the geographic location of the sample data into the regression parameters [50,51]. Therefore, we use the GWR model to analyze the heterogeneity of the influence of the driving factors that affect ESV change. The formula is as follows:

$$Y_i={\beta }_0\left(u_i,v_i\right)+{{\displaystyle \sum }}_{i=1}^{k=1}{\beta }_k(u_i,v_i)X_{ik}+{\epsilon }_i$$

(18)

where Y_i is the explained variable, i.e., the ESV change amount or rate of the i-th county; X_ik is the explanatory variable, i.e., the k-th explanatory variable of the i-th county; (u_i, v_i) is the geographic location of the i-th county; β₀(u_i, v_i) and β_k(u_i, v_i) are the constant and the k-th regression parameter of the i-th county; and ε_i is random error item.

2.4. Data Source

The data used in this paper include RLCP data and socioeconomic data. The RLCP data were obtained from the Department of Natural Resources of Hubei Province. The socioeconomic data were derived from China Statistical Yearbook, China Water Conservancy Statistical Yearbook, Hubei Statistical Yearbook, and National Agricultural Products Cost–benefit Data Compilation.

3. Results

3.1. Characteristics of Land Use Change

Table 1. The area of different land use types before and after land consolidation.

	Cultivated Land	Garden Land	Forest Land	Water	Ditch	Grassland	Rural Roads	Other Lands	Construction Land
Before consolidation/hm2	366,944	14,949	2624	28,966	14,957	4709	10,273	16,894	871
After consolidation/hm2	374,371	14,786	1587	27,471	14,933	1719	11,431	13,968	659
Area change/hm2	7427	−163	−1037	−1495	−24	−2990	1158	−2925	−211
Range change/%	2.02	−1.09	−39.52	−5.16	−0.16	−63.50	11.27	−17.32	−24.26

shows the areas of different land use types before and after RLCPs. After RLCPs, the area of cultivated land and rural roads increased, while the area of other lands, barren grassland, water, forest, construction land, garden, and ditch decreased. However, the increase rate of rural roads is significantly higher than that of cultivated land, which indicates that the main purpose of RLCPs in Hubei Province is no longer to increase the area of cultivated land and promote agricultural production, but to pay more attention to the construction and improvement in rural infrastructure [28,39,40].

Figure 2 shows the changes in land use structure indices before and after RLCPs. After RLCPs, the LDI, LVI, and LEI indices of each county show a decreasing trend, while the LAI index shows an increasing trend. RLCPs can promote the transformation of land use types, such as barren grassland, water, and forest converted into cultivated land, which increases the proportion of cultivated land. Although RLCPs can reduce the complexity of land use and significantly improve the dominance of land use, they can also lead to uneven land use.

3.2. Spatial Differentiation of ESV Change

3.2.1. Spatial Distribution Characteristics of ESV Change

Figure 3 shows the spatial distribution of the amount and rate of ESV increase. Grading similar amounts and rates of ESV increase to the same level is convenient for comparison. Therefore, using the Jenks natural breaks classification method [52,53], the amounts and rates of ESV increase were divided into four levels: low, medium-low, medium-high, and high. As can be seen in Figure 3a, the amount of ESV increase is characterized as high in the middle, while low in the east and west. Specifically, the ESV amount in the whole province, the low mountain and hilly areas in eastern Hubei, the plain areas in central Hubei, and the mountainous plateau area in western Hubei increased by an average of USD 3.54 × 10⁶ a⁻¹, USD 2.84 × 10⁶ a⁻¹, USD 5.28 × 10⁶ a⁻¹, and USD 1.52 × 10⁶ a⁻¹, respectively. The high-level zones are only found in Xiantao City, the medium-high-level zones are mainly located in the small hillock areas, the medium-low zones are mainly distributed in the plain and water network polder areas, and the low-level zones are mainly distributed in eastern and western Hubei. As shown in Figure 3b, the increasing rate of ESV is high in eastern and central-northern Hubei, while low in western and central-southern Hubei. Specifically, the entire province, the low mountainous and hilly areas in eastern Hubei, the plain areas in central Hubei, and the mountain plateau area in western Hubei are characterized by an average increase rate of 10.49%, 12.46%, 10.63%, and 8.76%, respectively. Additionally, the low and medium-low level zones are mainly concentrated in western Hubei, while the zones with medium-high and high levels are mainly distributed in central and eastern Hubei.

3.2.2. Spatial Agglomeration Characteristics of ESV Change

Figure 4 shows the Lisa cluster map of the amount and rate of increase in ESV. Moran’s I value of the amount of ESV increase was 0.1041, passing the significance test at 0.01 levels. Moran’s I value of the rate of ESV increase was 0.3396, passing the significance test at 0.001 levels. These results show that there is an obvious spatial autocorrelation between the amount of ESV increase amount and the rate of ESV increase. In terms of the amount of ESV increase, the high-high and low-high areas are relatively small and mainly distributed in central Hubei, while the low-low areas are relatively large and mainly concentrated in northeast and southwest Hubei. This indicates that RLCPs in Hubei Province were widely distributed with an obvious diffusion effect. In terms of the rate of ESV increase, the high-high areas are mainly located in Wuhan, Ezhou, and Huanggang, and these areas are the hot spots of high-standard farmland construction in Hubei Province. The low-low areas are mainly located in Shiyan. The low-high and high-low areas are mainly concentrated in southern Hubei. These results suggest that with the economic development, each county increased its RLCPs investment, and regional differences gradually decreased.

3.3. Driving Mechanism of ESV Change

3.3.1. Driving Factors Selection

In addition to RLCPs, natural, economic, social, policy, and other factors will influence the change in ESV. Firstly, natural factors, especially cultivated land factors, are the sufficient conditions for RLCPs. The main purpose of RLCPs is to improve the quality of cultivated land and increase the area of cultivated land. Secondly, RLCPs belong to the engineering measures of capital investment, and good economic conditions are a necessary condition for its implementation. Thirdly, RLCPs are an important tool for promoting rural revitalization and achieving regional sustainable development. Therefore, social factors such as population and farmers’ income are also important factors influencing the layout of RLCPs. Fourthly, policy factors, especially investment scale and construction scale, not only indicate the frequency of RLCPs, but also represent the difficulty of RLCPs. Finally, an apparent linear relationship is noticed between indicators of land use change and ESV change. Based on the above analysis, 15 factors were selected as the driving factors of ESV change (

Table 2. Driving factors of ESV change in RLCPs.

Driving Factors		Factors Definition/Unit	Code
Natural factors	Cultivated land area	Cultivated land area (hm2)	X1
	Cultivated land proportion	Cultivated land area/Total land area (%)	X2
	Per capita cultivated land area	Cultivated land area/Total population (hm2·person−1)	X3
Economic factors	GDP	GDP (USD 100 million)	X4
	Per capita GDP	Total GDP/Total population (USD·person−1)	X5
	Average investment in fixed assets	Social fixed assets investment/Total land area (104 USD·km−2)	X6
Social factors	Per capita income of rural residents	Per capita net income of rural residents (USD)	X7
	Population density	Total population/Total land area (person·km−2)	X8
	Urbanization rate	Non-agricultural population/Total population (%)	X9
Policy factors	Investment scale	Total investment scale of RLCPs (104 USD)	X10
	Construction scale	Total construction scale of RLCPs (hm2)	X11
Land use change factors	LDI variation	LDI before RLCP-LDI after RLCP	X12
	LAI variation	LAI before RLCP-LAI after RLCP	X13
	LEI variation	LEI before RLCP-LEI after RLCP	X14

).

3.3.2. Geographical Detector Results

The geographical detector is used to identify the driving factors of the increase in ESV amount and rate, and the results are shown in

Table 3. Estimated results of geographical detector.

Increase Amount	X1	X2	X3	X4	X5	X6	X7	X8	X9	X10	X11	X12	X13	X14
q value	0.11	0.14	0.01	0.16	0.11	0.09	0.17	0.10	0.06	0.27	0.57	0.03	0.03	0.03
sig.	0.14	0.05	0.86	0.04	0.12	0.22	0.01	0.16	0.30	0.00	0.00	0.60	0.63	0.53
Increase Rate	X1	X2	X3	X4	X5	X6	X7	X8	X9	X10	X11	X12	X13	X14
q value	0.01	0.24	0.12	0.10	0.05	0.09	0.12	0.08	0.07	0.02	0.07	0.13	0.24	0.28
sig.	0.99	0.03	0.25	0.83	0.83	0.99	0.23	0.74	0.38	0.81	0.89	0.13	0.01	0.00

. In terms of the amount of ESV increase, X4, X7, X10, and X11 pass the significance test at 0.05 level, and the order of q values from large to small is X11, X10, X7, and X4. This indicates that the amount of ESV increase is significantly affected by policy, social, and economic factors. In terms of the rate of ESV increase, only X2, X13, and X14 passed the significance test at 0.05 level, and the order of q values from large to small is X14, X13, and X2. This shows that the rate of ESV increase is significantly affected by land use change factors and natural factors.

3.3.3. GWR Results

Spatial Differentiation of the Dominant Factor Affecting the Amount of ESV Increase

Figure 5 describes the spatial differentiation of the dominant influencing the amount of ESV increase. It can be seen in Figure 5a that there is a positive correlation between GDP and the amount of ESV increase, indicating that the higher the level of economic development, the higher the total amount of ESV increased by RLCPs. Specifically, RLCPs belong to the capital investment engineering measures, and the higher the regional GDP, the greater the scale and amount of investment in RLCPs, and the greater the ecological effect of RLCPs. From the spatial perspective, the impact intensity of GDP on the amount of ESV increase shows a decreasing trend from southwest to northeast, which is opposite to the GDP of each unit. This is because the lower the economic level, the stronger the motivation for promoting the economy by RLCPs. The per capita income of rural residents is negatively correlated with the amount of ESV increase (Figure 5b). The impact intensity of per capita income of rural residents on ESV increases from west to east, which is consistent with the income level of rural residents in each unit. It may be that the higher the farmers’ income, the higher their awareness of RLCPs. There is a negative correlation between the investment scale and the amount of ESV increase (Figure 5c), which may be that the higher the investment, the higher the cost of RLCPs, and then the number of projects or the construction scale is reduced. The effect of investment scale on the amount of ESV increase shows a decreasing trend from southeast to northwest, which is in direct proportion to the investment scale of RLCPs. At the same time, the positive impact of construction scale on the amount of ESV increase gradually increased from west to east (Figure 5d). The reason is that RLCPs mainly affect ESV by changing the land use structure, so its construction scale has a substantial impact on the amount of ESV increase [28,34,36].

Spatial Differentiation of the Leading Factor Affecting the Rate of ESV Increase

Figure 6 describes the spatial differentiation of the dominant factors influencing the rate of ESV increase. In plains and hilly areas, the cultivated land proportion is negatively correlated with the rate of ESV increase (Figure 6a). The impact intensity of cultivated land proportion on the rate of ESV increase gradually weakens around the Jianghan Plain. This is mainly due to the high proportion of cultivated land and low proportion of ecological land in these areas. The RLCPs of each unit are mainly based on the internal potential, and the rate of ESV increase is low. There is a positive correlation between the cultivated land proportion and the rate of ESV increase in mountainous areas. The more the terrain fluctuates, the lower the rate of land use. Each unit increases the cultivated land proportion through RLCPs, which will improve ESV. The relationship between LDI variation and the rate of ESV increase is similar to the cultivated land proportion. Except for some units in Enshi Prefecture and Shiyan City, the LEI variation is positively correlated with the rate of ESV increase. A common general rule is that the more the LEI index decreases, the lower the rate of ESV increase. The influence intensity is the strongest in the water network polder areas, and gradually weakens in a semicircular shape from south to north and from east to west, which is consistent with the distribution of cultivated land resources and the land use intensity in each unit.

4. Discussion

4.1. Strengthening Research on ESV Change under the Background of RLCP

We have made corresponding modifications to the ESV evaluation method, which is realized by modifying the average precipitation, food production, socioeconomic development stage, as well as land use type, etc. The improved evaluation method is more consistent with the actual situation of Hubei Province in terms of specific application scales. The empirical results based on the improved value equivalent method show that after RLCPs, the ESV of the RLCP areas in 71 counties of Hubei Province increased. This result indicated that RLCPs can effectively improve the ecological environmental quality.

Currently, China is vigorously promoting the implementation of RLCPs, which is characterized by rich content, diverse models, diverse objectives, and comprehensive means. It has played an important role in optimizing spatial land layout, ensuring food security, and improving rural production and the living environment and ecosystem. At present, China is in a critical period of economic transformation. While ensuring sustained and stable economic growth, it is also implementing relevant policy measures such as RLCPs to reduce the damage caused by economic development to resources and the environment, ensure food demand, narrow development gaps, and improve the quality of economic development. Studying the ESV variation patterns and driving mechanisms in China’s RLCPs provides a reference for other developing countries to achieve these goals in their development process.

However, this study has several research limitations. First, due to the limitation of data acquisition, the first year after the RLCPs were put into use was selected to represent the year after consolidation, which results in a smaller change in the amount and rate of ESV. After RLCPs, the supply, regulatory, and cultural services are in dynamic change. The changes in supply and partial cultural services can be observed in a short time, but the changes in regulatory and partial cultural services need to be reflected over a long period of time [19,25,26,38]. Therefore, the ESV change in RLCPs requires long-term fixed-point observation and analysis. Second, the change in the relationship between ecosystem services under the background of RLCPs was not quantitatively analyzed. The relationship between supply, regulatory, and cultural services is contradictory and antagonistic; that is, these three categories of ecosystem services do not increase or decrease simultaneously, with some services increasing, some services decreasing, and some services remaining unchanged [6,12,18]. In practice, the impact of RLCPs on ecosystem supply and regulatory services is reflected in ensuring the food production and quality of the region. Therefore, the development of the region relies on agricultural planting. The impact of RLCPs on ecosystem cultural services is reflected in strengthening the natural resources and historical and cultural characteristics of the region through relevant measures, vigorously developing the ecological tourism industry, etc. Its characteristic is that it will, to some extent, abandon the grain cultivation it originally relied on for rural development. Ensuring food security and development are equally important. Hence, the trade-offs, dependencies, or synergy relationships between ecosystem services affected by RLCPs should be deeply explored to provide support and guidance for the spatial layout of RLCPs and the optimal management of ecosystem services. Similarly, in the selection of content and policy formulation for the implementation of future RLCPs in China, appropriate content needs to be selected through the balance of these contents to ensure the positive benefits brought by the implementation of RLCPs.

4.2. Implementing Differentiated RLCPs Policy

Continue to implement the high-standard basic farmland consolidation projects in central Hubei: Central Hubei includes the Jianghan Plain and the middle and lower reaches of the Hanjiang River Plain, with flat terrain, abundant cultivated land resources, and a high level of economic development. The investment scale and the construction scale of RLCPs far exceed those in eastern or western Hubei. The amount of ESV increase in central Hubei is much larger than that in eastern or western Hubei, but the rate of ESV increase is relatively low. Additionally, the proportion of cultivated land and the land use diversity index change has a significant impact on ESV change in Central Hubei. Therefore, we should continue with the implementation of basic farmland consolidation projects, but we should also strengthen the protection of land use diversity in the RLCP process and try to retain water areas and other land types that have a greater impact on the ecosystem in order to achieve good ecological effects.

Expanding the implementation of low hilly consolidation projects in eastern Hubei: The terrain of eastern Hubei is mainly hilly, with a small number of plains and mountains. Although the investment scale and the construction scale of RLCPs are far smaller than those of central or western Hubei, the newly increased rate of cultivated land is much higher than that of central or western Hubei, and the rate of ESV increase is also higher than that of central and western Hubei. Therefore, we should expand the implementation of consolidation projects in the area of low hills in eastern Hubei in order to ensure the dynamic balance of cultivated land in Hubei Province and achieve the dual effects of production and ecology.

Strengthen the investment in ecological conservation projects in western Hubei: Western Hubei includes Qinba Mountain and Wuling Mountain, and is characterized by poor natural conditions, weak ecological foundation, and low per capita income. Meanwhile, both the investment and construction scale of RLCPs are relatively small. The amount and rate of ESV increase in western Hubei are low; GDP and construction scale are positively correlated with the amount of ESV increase, while the investment scale and per capita income of rural residents are negatively correlated with the amount of ESV increase. Therefore, western Hubei can more actively implement small-scale RLCPs. What should be emphasized the most is the protection of natural forests and water conservation, while increasing the area and improving the quality of cultivated land.

5. Conclusions

RLCPs have become one of the largest organized human activities aiming to change land use patterns and impact terrestrial ecosystems. This study used Hubei Province as a case study to explore the spatial differentiation and driving mechanisms of ESV change in RLCPs. We use the improved ESV evaluation model to analyze the spatial differentiation of the amount and rate of ESV change due to RLCPs, and then adopt the geographic detectors and the geographically weighted regression model to identify the dominant factors affecting the amount and rate of ESV change due to RLCPs. The main results and conclusions are as follows:

(1)

Although RLCPs make the unevenness of land use obvious, they evidently reduce the complexity of land use and significantly improve the dominance of land use. After RLCPs, the land use diversification index, land use diversity index, and land use dominance index of each county decreased, while the land use dominance index increased. This also indicates that RLCPs can improve ecosystem complexity and stability, and ultimately improve ESV;

(2)

The ESV of the RLCPs areas in 71 counties of Hubei Province increased, with an average increase of USD 2.37 × 107 a⁻¹. The amount of ESV increase is large in central Hubei, while small in eastern and western Hubei. The rate of ESV increase is high in eastern and central-northern Hubei, while low in western and central-southern Hubei. This implies that RLCPs can improve ESV, but the improvement effect has significant regional differences;

(3)

The amount of ESV increase is positively corrected with GDP and construction scale, but is negatively correlated with investment scale and per capita income of rural residents. The rate of ESV increase is negatively correlated with the proportion of cultivated land and the change in the land use diversification index, but positively correlated with the change in the land use evenness index. However, their driving effects have significant spatial heterogeneity. Therefore, governments should carry out differentiated RLCPs according to regional natural geographical conditions and socioeconomic development levels in order to protect the ecological environment.