Land Use Transition and Effects on Ecosystem Services in Water-Rich Cities under Rapid Urbanization: A Case Study of Wuhan City, China

Abstract

Ecosystems can provide ecosystems service for human society. Exploring the impact of land use transition of water-rich cities on ecosystem service can obtain a large amount of scientific data, and its findings have certain significance for building a sustainable city land use transition pattern. This study takes Wuhan, a water-rich city, as the study area, combines remote sensing image data and macro-economic data in the region from 2000 to 2020, and uses spatial analysis methods, the equivalent factor calculation method, and hot spot analysis to portray the spatial-temporal patterns of land use transition in Wuhan and its effect on ecosystem service values. The results show that: (1) farmland, water, and built-up land are the main land use types in Wuhan, and the continuous expansion of built-up land area is an important factor in the decrease of farmland, wetland, and grassland areas; (2) The change in ecosystem service values in Wuhan is influenced by the fluctuation of water area, and the overall ecosystem service values in Wuhan increased from CNY (Chinese Yuan) 56.498 billion (USD 8.47 billion) in 2000 to CNY 56.749 billion (USD 8.508 billion) in 2020, with an increase of CNY 251.3 million (USD 37.676 million) between 2000 and 2020. Among them, the ecosystem services values of water increased by CNY 1.223 billion (USD 183.358 million), and the ecosystem service values of its assumed hydrological regulation function also increased by CNY 995.7 million (USD149.28 million) during this period; (3) In the period 2000–2020, the hotspot areas for the value supply of ecosystem services were mainly distributed in Jiangxia, Hongshan, Hannan, Xinzhou, Huangpi and some areas of Caidian, which are covered with a large proportion of water, while the cold spots are mainly distributed in Jiang’an, Jianghan, Qiaokou, and Qingshan districts, which have more built-up land, and Huangpi, Caidian, Jiangxia, and Xinzhou districts, which have more farmland. Sustainable land planning solutions should take into account typical regional land use transition patterns and incorporate them into smart land planning practices. This study can provide key information for smart land planning.

1. Introduction

Since the Reform and Opening-up, the level of China’s economic development has been steadily improving and the urbanization process has been promoted. According to statistical data, by the end of 2021, China’s urbanization level had reached 64.72% [1]. At present, the level of urbanization rate in China has been continuously increasing at a level of more than 1% for many years, indicating that China’s urbanization has entered a stage of rapid development and that the degree and depth have reached a high level. While rapid urbanization brings about rapid economic and social development, it also profoundly affects urban land use and the natural environment [2]. The urban land use pattern has changed in the continuous development of urbanization, and the natural land space has been squeezed by the expansion of urban land [3,4], which has exerted certain pressures on the ecosystem, disrupted the material transfer between ecosystems [5], and negatively affected the ecosystem services [2].

Land use transition is a process of morphological transition of land use in a certain period driven by economic and social changes. Its transition process is divided into two dimensions, including changes in quantity and structure as well as changes in quality and property rights [6]. Land use transition causes certain ecological changes along with land use changes [7]. Ecological changes include changes in elements such as climate [8,9] and water [10] as well as changes in environmental quality [11,12]. Scholars usually measure the ecological effects of land use from two perspectives. One is to adopt the method of biomass measurement, and select single indicators such as vegetation net primary productivity [13], normalized vegetation index [14], or comprehensive measurement indicators such as remote sensing ecological distance index [15], remote sensing ecological index [16] for measurement. The second is to use quantitative analysis means of land use change, including ecosystem service value [17,18], ecological environment quality index [19], and landscape ecological risk index [20]. The above studies fully illustrate the role of land use change in causing ecosystem change. However, changes in the structure and morphology of ecosystems will also lead to changes in ecosystem services, suggesting an interactive relationship between land use transition and ecosystem services.

Ecosystem services are the benefits that humans derive directly or indirectly from ecosystems [21,22]; they provide conditions for improved human well-being as well as sustainable development [23,24]. In 1997, Costanza et al. estimated the global ecosystem services values (ESV) based on equilibrium values theory as well as effect values theory by classifying ecosystem services into 17 types [22]. In 2003, in the Millennium Ecosystem Assessment (CA) project, ecosystem services were divided into four categories: support, supply, regulation and culture, which are considered to have a fundamental role in human survival and development [24]. Chinese authors Xie et al., constructed an ecological service values per unit area equivalence scale for terrestrial ecosystems in China based on Costanza’s study, and continuously improved the ecological service value assessment method based on expert knowledge as well as statistical information [25,26,27]. A scientific method of ESV evaluation can provide a reference for maintaining the reasonable stability of ecosystems [28], so that ecosystem service provisions can be maintained at a certain level.

At present, the research on ecosystem services has been intensified, and the changes in ecosystem services caused by land use transition have also attracted more scholars’ attention. Existing studies have formed diverse analytical ideas based on different research scales, research objects, and research methods. In terms of research scales, a large number of scholars have conducted numerous studies based on global [29], intercontinental [30], national [31], regional [32,33,34], provincial [35], municipal [36], and county [37] scales. As for the research objects, scholars mostly analyze them based on ecological lands such as waters [38], wetlands [39], and forests [40,41], which vary more significantly due to higher ESVs coefficients per unit area. In terms of research methods, scholars have mostly conducted studies based on spatial analysis methods [42,43] as well as statistical analysis [44] to grasp the inner logic of land use transition and ecosystem service changes. In addition, some scholars innovatively cast their perspectives on the changes in ESVs due to future land use change to provide a basis for ecosystem safeguarding. For example, Zheng et al., 2021 took Tapieh Mountains Geopark as the research object and simulated the change of regional ESVs under three scenarios: priority of tourism development, core area protection, and strict ecological protection [45].

Wuhan, as one of the important central cities in the central region, has a long history of economic development, and its urbanization rate has grown in this context, from 77.07% in 2010 to 84.31% in 2020 (from the Wuhan Municipal Bureau of Statistics.). At the same time, according to 2021 data, as a water-rich city, Wuhan has a water area of 2217.6 km², accounting for 26.1% of the total area of the city (from Wuhan Municipal People’s Government.). As an important supply of ecosystem services, the water provides higher services per unit area compared to other land types. Thus, in the context of rapid urbanization, along with the dramatic land use transition, there are also significant changes in ecosystem services in Wuhan. To quantify the ecosystem services, this study proposes to analyze the following methods: (1) To investigate the characteristics of land use transition in Wuhan during 2000–2020 using the land use transition matrix and spatial analysis; (2) Exploring the change of ESVs in Wuhan from 2000 to 2020 based on the equivalence factor method; (3) To investigate the spatial distribution of the change in ESVs based on the detailed processing of each raster cell, combined with the hot spot analysis method. (4) To clarify the intrinsic relationship between land use transition and ecosystem service changes in Wuhan. The quantitative analysis of land use transition status and its ecological service effects in Wuhan is of significance. First, this study takes a water-rich city as the study area, which can provide key information for exploring the characteristics of ecosystem service changes in water-rich cities or regions and quantify the influence of water on regional ecosystems. It can provide a reference for other water-rich cities; Secondly, this study reflects the changes of ecological environment conditions in Wuhan based on the provision of ecosystem services, and then provide a reference for the rational land use planning in Wuhan, so that urban land use planning can play an important role in ecological protection and governance and improve the ecological environment condition to some extent.

2. Data and Methodology

2.1. Data

Wuhan, the capital of Hubei province, is located in the central region of China, with a land area of 8569 km²; the topography is mainly plain, while a small number of low hills are distributed with extensive farmland and rich natural resources. Wuhan’s water is widely dispersed, with an overall water area of 2217.6 km², accounting for 26.1% of the city’s total area. Wuhan is known as the “City of 100 Lakes”, and there are 166 lakes in the city. According to the data of the 2021 Wuhan Water Resources Bulletin, the total area of lakes in Wuhan is 867.07 km², of which the area of Tangxun Lake accounts for 47.6 km², which is the largest urban lake in China [46]. These superior water resource conditions make Wuhan a typical water-rich city. A map of Wuhan is shown in Figure 1.

Wuhan is known as the “thoroughfare of nine provinces” and is located at the intersection of the Yangtze River golden waterway and the Beijing-Guangzhou railroad trunk line, with convenient water and land transportation conditions. Riding on the wave of Reform and Opening-up, Wuhan has witnessed remarkable development in the past three decades. In the first half of 2019 alone, Wuhan’s foreign direct investment was $1.204 billion, up 66.5% year-on-year, ranking second among 19 sub-provincial cities and above nationwide in terms of growth rate. During this period, Wuhan has added 16 Fortune 500 companies, bringing the total to 282.

The land use data used in this study are obtained from the Resource and Environmental Science and Data Center (https://www.resdc.cn/, accessed on 5 May 2022), and the data of the study area for the three periods of 2000, 2010 and 2020 were obtained through the mosaic in ArcGIS 10.2. Based on the LULC classification and Xie’s study [27], the land use types are classified into seven categories: farmland, forest, grassland, water, wetland, built-up land, and bare land. The socio-economic data used include average crop yields, average output values per unit of the crop, and crop prices from the “Wuhan Statistical Yearbook” from2019–2021, “the National Compilation of Agricultural Costs and Returns”, and the Hubei Provincial Grain Bureau.

2.2. Methodology

2.2.1. Land Use Transition Matrix

The land use transition matrix is derived from the quantitative description of system state and state shift in system analysis, which can reflect the changes of land use types in two specific years in the form of a matrix(

Table 1. General form of the land use transition matrix.

T2						$P_{i+}$	Reduce
		${\mathit{A}}_1$	${\mathit{A}}_2$	…	${\mathit{A}}_{\mathit{n}}$
T1	$A_1$	$P_{11}$	$P_{12}$	…	$P_{1n}$	$P_{1+}$	$P_{1+}-P_{11}$
	$A_2$	$P_{21}$	$P_{22}$	…	$P_{2n}$	$P_{2+}$	$P_{2+}-P_{22}$
	⋮	⋮	⋮	⋮	⋮	⋮	⋮
	$A_n$	$P_{n1}$	$P_{n2}$	…	$P_{nn}$	$P_{n+}$	$P_{n+}-P_{nn}$
	$P_{+j}$	$P_{+1}$	$P_{+2}$	…	$P_{+n}$	1
	Increase	$P_{+1}-P_{11}$	$P_{+2}-P_{22}$	…	$P_{+n}-P_{nn}$

), and can better reflect the changing pattern of land use in each period [47], and at the same time combine with the spatial analysis function of GIS to reflect the spatial shift distribution of land use in the region more intuitively. In this study, we analyze the changes in land use patterns from 2000 to 2020 by using the land use transition matrix. The land use transition matrix quantitatively describes the land use pattern at two time points, where rows denote land use types at time point T1 and columns denote land use types at time point T2. $P_{ij}$ denotes the percentage of area converted from land use type i to land use type j to the total land area between T1 and T2; $P_{ii}$ denotes the percentage of area where land use types remain unchanged between T1 and T2; $P_{i+}$ denotes the percentage of total area of land type i at T1 total area percentage, and $P_{+j}$ denotes the total area percentage of land type j at T2; $P_{i+}-P_{ii}$ is the percentage decrease of land type i area between T1 and T2; $P_{+j}-P_{jj}$ is the percentage increase of land use type j area between T1 and T2.

2.2.2. Ecosystem Service Values Assessment

Considering the situation of Wuhan, this study classifies the land use types into seven categories: farmland, forest, grassland, water, wetland, built-up land, and bare land based on the relevant research content of LULC, and adjusts and consolidates the secondary classification of each category. The specific scope of each category is shown in

Table 2. Criteria for secondary land classification.

Primary Land Classification	Secondary Land Classification
Farmland	paddy field, dry land, vegetable field
Forest	deciduous broad-leaved forest, evergreen broad-leaved forest, deciduous coniferous forest, evergreen coniferous forest, mixed forest, shrub land
Grassland	grassland, meadow, savanna, desert grassland, urban artificial grassland
Water	river, lake, reservoir, pits, pond
Wetland	swamp, river flood wetland, forest/shrub wetland, peat swamp, mangrove, salt marsh
Built-up land	residential land, industrial and mining land, transportation facilities land
Bare land	desert, sandy land, gravel land, bare rock, saline alkali land

.

One standard unit of agroecosystem ecological service values equivalence factor (hereafter referred to as standard equivalence) is the economic values of natural food production per year for 1 hm² of farmland with a national average production. Taking this equivalence as a reference and combining it with expert knowledge can determine the equivalence factor of other ecosystem services, which is useful because it can quantify the potential contribution capacity of different types of ecosystems to ecological service functions. In this study, based on the ESV per unit area equivalence table [26] of Xie, and combined with the situation of Wuhan, the grain yield correction method is used to derive the ESV coefficient table of Wuhan. For the selection of coefficients for each category, the ESV coefficients of farmland, forest, and grassland are obtained by averaging each secondary category according to the actual situation in Wuhan. Since there is no glacial snow land in Wuhan, the ESV coefficient of the water system is used to represent the water area in this study. As for the economic value of the equivalent factor, this study refers to the method of Xie et al., and considers that the economic value of the equivalent factor of ESV is equal to 1/7 of the annual economic value of 1 hm² average yield of farmland [25]. The grain yield value of the farmland ecosystem is mainly calculated based on the main grain products such as early-season rice, middle-season rice, late rice, wheat, and corn.

To ensure the accuracy of the experimental results, this study adjusts the differences in ecosystem values between Wuhan and the national regional scale by grain yield correction as follows. By collecting the average 1 hm² yields of early-season rice, middle-season rice, late rice, wheat, and corn in Wuhan from 2018 to 2020 and the average unit price of grain in that year, the values of a single ecosystem equivalence in Wuhan are 2186.54 CNY/hm² after combining equation (1) and then averaging and combining them with the study of Xie et al., [26]; this is used to construct the Wuhan unit area ESV coefficient table (

Table 3. The ESVs coefficient per unit area in Wuhan (CNY·hm⁻²·a⁻¹).

		Farmland	Forest	Grassland	Wetland	Water	Bare Land
Supply	Food production	2416.13	552.10	510.19	1115.14	1749.23	0.00
	Production of material	535.70	36,985.38	750.71	1093.27	502.90	0.00
	Water supply	−2853.44	655.96	415.44	5663.15	18126.44	0.00
Regulation	Gas conditioning	1946.02	4170.83	2638.43	4154.43	1683.64	43.73
	Climate regulation	1016.74	12,479.70	6975.07	7871.56	5007.18	0.00
	Clean the situation	295.18	3656.99	2303.16	7871.56	12,135.31	218.65
	Hydrological regulation	3268.88	8166.74	5109.22	52,979.94	223,552.18	65.60
Support	Soil conservation	1137.00	5078.25	3214.22	5050.91	2033.49	43.73
	Maintain nutrient cycle	338.91	388.11	247.81	393.58	153.06	0.00
	Biodiversity	371.71	4624.54	2922.68	17,208.10	5575.69	43.73
Cultural	Aesthetic landscape	163.99	2028.02	1290.06	10,342.35	4132.57	21.87

).

$$E_a=\frac{1}{7}{\displaystyle \sum }_{i=1}^n\frac{q_i{p}_i{m}_i}{M}$$

(1)

where, $E_a$ denotes the values of 1 hm² ecosystem service equivalence factor in yuan/hm²; i is the type of crop; n is the total type of crop; $q_i$ is the average yield per hectare of the i crop; $p_i$ is the average price of the i crop; $m_i$ is the planted area of the i crop; and $M$ is the total planted area of the n types of crops.

According to the revised table of the ESV coefficient per unit area of Wuhan, and combined with Formula (2) and (3), the ESV is calculated [27,36]:

$$ESV={\displaystyle \sum }_{i=1}^n{A}_i\times V{C}_i$$

(2)

$$ES{V}_f={\displaystyle \sum }_{i=1}^n{A}_i\times V{C}_{if}$$

(3)

where, ESV is the total ESVs of the study area; $ES{V}_f$ is the values of the f ecosystem service function; $A_i$ is the area of the i land use type; $V{C}_i$ is the ESVs coefficient of the i land use type; $V{C}_{if}$ is the values coefficient of the f ecosystem service function of the i land use type.

2.2.3. Hot Spot Analysis

Hot spot analysis (Getis-Ord Gi*) is often used to study the clustering of specific elements, and then derive the cold spots and hot spots of their distribution to measure the spatial correlation of regional elements within a certain range [48]. Based on a 30 m × 30 m cell for processing, the cold and hot spot analysis of ESVs change in Wuhan can yield high values areas and low values areas of ESVs change. Getis-Ord Gi* local statistics are shown in Equation (4).

$$G_i^{*}=\frac{{{\displaystyle \sum }}_{j=1}^n{w}_{i,j}{x}_j-\overline{X}{{\displaystyle \sum }}_{j=1}^n{w}_{i,j}}{S\sqrt{\frac{\left[n{{\displaystyle \sum }}_{j=1}^n{w}_{i,j}^2-{\left({{\displaystyle \sum }}_{j=1}^n{w}_{i,j}\right)}^2\right]}{n-1}}}$$

(4)

where,$x_j$ is the attribute values of element j; $w_{i,j}$ is the spatial weight between elements i and j; n is the total number of elements. The higher the values of $G_i^{*}$, the higher the concentration of hot spots in the region, and vice versa. To become a statistically significant hot spot, this region should have a high value and be surrounded by other regions with high values. Similarly, the statistically significant cold spot region not only has a low value, but the surrounding regions also have low values.

3. Results

3.1. Land Use Transition in the Context of Rapid Urbanization

The land use data of Wuhan in the periods of 2000, 2010, and 2020 are reclassified to obtain the land use change map. From Figure 2, it can be seen that the main types of land use in Wuhan during 2000–2020 are farmland, water, forest, and built-up land. Among them, farmland is mainly distributed in Huangpi District, Xinzhou District, and Jiangxia District. However, during this period, the amount of farmland in Wuhan has been decreasing in general. The built-up land is distributed in the central area of Wuhan and decreases from the central area to other areas, with the overall area expanding. Water is mainly distributed along the Yangtze River, among which lakes such as the East Lake, Tangxun Lake, and Liangzi Lake are more obviously distributed. The rest of the land types are less distributed, except for patches of forest in the northwestern part of Wuhan, while forest, grassland and bare land in the rest of the city are scattered.

The data of land use changes in the three periods of 2000, 2010, and 2020 are counted to obtain the figure of land use type area changes in Wuhan. As can be seen from Figure 3, the overall land use changes in Wuhan during the two decades are mainly characterized by the decrease in farmland and the increase in built-up land. During this period, the farmland area decreased by 456.94 km², and the change was more drastic in the first period and increased in the later period. The built-up land has increased the most in these 20 years, with an increase of 481.11 km², of which the increase in the first ten years was 394.63 km², while the increase in the area of built-up land between 2010 and 2020 was relatively small. In addition to the built-up land, the area of water and forest also increased. The rest of the land types have decreased to different degrees, but the decrease is relatively small. Farmland, water, and built-up land are the main land use types in Wuhan, among which the proportion of farmland is the largest, decreasing from 61.09% in 2000 to 55.77% in 2020. This is followed by the water area, whose proportion fluctuates during the study period, increasing from 18.23% in 2000 to 18.75% in 2020. The proportion of built-up land keeps increasing, from 7.69% in 2000 to 13.3% in 2020.

To quantify and visualize the land use transition in Wuhan, this study presents this content through the land use transition matrix. This study makes some adjustments based on the general form of the land transition matrix. As shown in

Table 4. Land use transition matrix during 2000–2020 in Wuhan (km²).

		2010
	Land Use Type	Farmland	Forest	Grassland	Wetland	Water	Built-Up Land	Bare Land	Total
2000	Farmland	4665.09	23.13	15.02	51.04	149.50	334.17	0.57	5238.52
	Forest	15.29	741.74	1.29	0.20	3.22	19.73	0.08	781.55
	Grassland	3.82	11.47	56.49	2.06	4.57	10.92	0.19	89.53
	wetland	12.25	0.07	0.26	109.89	99.97	9.46	0.05	231.95
	Water	48.60	2.33	3.36	90.85	1362.86	55.31	0.17	1563.47
	Built-up land	15.66	2.48	1.11	1.68	17.43	620.68	0.06	659.10
	Bare land	0.71	0.33	0.10	0.00	0.14	3.46	6.01	10.76
	Total	4761.43	781.55	77.62	255.73	1637.69	1053.74	7.14	8574.88
		2020
	Land Use Type	Farmland	Forest	Grassland	Wetland	Water	Built-up land	Bare land	Total
2010	Farmland	4376.01	28.37	1.45	13.94	65.06	275.95	0.65	4761.43
	Forest	25.05	735.35	1.74	0.20	7.61	11.21	0.38	781.55
	Grassland	14.01	1.74	54.50	0.18	1.89	4.63	0.67	77.62
	wetland	49.72	0.35	0.81	116.16	81.13	7.51	0.03	255.73
	Water	120.99	3.85	2.93	46.25	1427.41	36.15	0.10	1637.69
	Built-up land	195.74	15.57	4.16	7.34	24.52	804.09	2.31	1053.74
	Bare land	0.35	0.06	0.02	0.05	0.39	0.66	5.60	7.14
	Total	4781.88	785.28	65.61	184.13	1608.01	1140.21	9.75	8574.88

, based on the ecosystem classification described in the previous section, the rows of the land transition matrix in this study represent the land use status in the previous period and the columns represent the land use status in the next period, both in km².

Based on the analysis of the land use transition matrix for the two periods of 2000–2010 and 2010–2020, the results show that: (1) during the period of 2000–2010, the land use conversion in Wuhan is mainly characterized by the transfer of farmland to built-up land and water. During this period, 573.43 km² of farmland was transferred out, of which 334.17 km² was converted into built-up land and 149.5 km² was converted into water. At the same time, 48.6 km² of water and 15.66 km² of built-up land were converted into farmland. Farmland, in general, showed a greater transfer out than in, with a decrease in the total area. In addition, grassland and bare land also showed the characteristics of transfer out more than transfer in. In contrast, forest, wetland, water and built-up land showed a trend of greater transfer in than transfer out; and (2) During the period of 2010–2020, the overall land use conversion in Wuhan was obvious, featuring the partial conversion of farmland into other land types and the conversion of a considerable number of various types of land into built-up land. During this period, farmland was partially converted out, with an area of 385.42 km², of which most of it was converted into built-up land and water area, with an amount of 275.95 km² and 65.06 km², respectively. At the same time, 195.74 km² and 120.99 km² of built-up land and water were transferred to farmland, respectively. The conversion of built-up land was also significant during this period, with a larger area of 249.64 km² of built-up land transferred out and 336.12 km² transferred in as built-up land, mostly from farmland and to a lesser extent from water, grassland, forest, wetland and bare land. At the same time, the water was also partially transformed, and the transferred out was larger than the transferred in, and the main object transferred out was farmland, and the source transferred in was wetland. It can be seen that during the study period, along with the continuous urbanization of Wuhan, the urban built-up land in Wuhan has been greatly expanded, resulting in a large amount of encroachment on farmland, water area, and forest. The demand for built-up land for urban production and living has led to a continuous reduction of ecological land, and therefore attention needs to be devoted to the conservation of ecological land. The conversion process between farmland, forest land, and water area also reflects Wuhan’s response to the policy of returning farmland to forest and lakes, and thus the conversion of farmland to forest and water.

3.2. Ecosystem Service Changes Driven by Land Use Transition

With the land use transition in Wuhan, the ESVs have been changing. According to

Table 5. Changes in ESVs by category in Wuhan during 2000–2020 (CNY).

Ecosystem	ESV (100 Billion)			Change of ESV (100 Billion)
	2000	2010	2020	2000~2010	2010~2020	2000~2020
Farmland	45.244	41.124	41.300	−4.121	0.177	−3.944
Forest	61.576	61.576	61.870	0.000	0.294	0.294
Grassland	2.361	2.047	1.731	−0.314	−0.317	−0.631
wetland	26.383	29.087	20.943	2.704	−8.144	−5.440
Water	429.411	449.794	441.644	20.383	−8.150	12.233
Bare land	0.005	0.003	0.004	−0.002	0.001	0.000
Total	564.980	583.631	567.492	18.651	−16.139	2.513

, the overall ESVs continued to fluctuate during the study period, with different degrees of change for different land types, with water being the main type of increase and farmland and wetland being the main types of loss. During the period 2000–2010, the ESVs of the study area increased by CNY 1865.1 million, with water contributing the largest increase of CNY at 2038.3 million. The farmland ESVs decreased in this period, with a loss of CNY 412.1 million. In the period 2010–2020, the overall ESVs changed from an increase in the previous period to a deficit, with some land types showing a large deficit. Among them, the downward trend of water ESVs is obvious, with a loss of values of CNY 815 million. In addition the wetland’s ESVs also changed from profit to loss, decreasing by CNY 814.4 million. Furthermore, the grassland’ ESVs also decreased somewhat. And the farmland’ ESVs changed from a decrease in the previous period to an increase in this period, with increased values of CNY equaled 17.7 million. The rest of the land types also have some degree of increase. From the whole period of 2000–2020, ESVs in Wuhan increased by CNY 251.3 million, and most of the land types had a decrease in ESVs, with the largest decrease in the ESVs of wetland, and only water and forest increased. The urbanization of Wuhan has put significant pressure on the ecological environment, and the transition of land use due to population and spatial urbanization has affected the ability of ecosystems to provide services, so the overall ESVs continue to fluctuate, but benefit from the sustained high level of ecosystem service provision in water, the overall ecosystem service remains at a certain level. The increase in the ESVs of water also reflects the effectiveness of conservation measures for Wuhan’s water, which is the main ecosystem providing services, and the large area of water in Wuhan, whose conservation policies deeply affect the overall ESVs of the city.

Based on the coefficients of each ecosystem function to calculate its ESVs, the table of changes in the structure of ESVs in Wuhan (

Table 6. Changes in the structure of ESVs in Wuhan during 2000–2020.

Primary Classification	Secondary Classification	ESV (CNY 100 Million)			Change of ESV (CNY 100 Million)
		2000	2010	2020	2000~2010	2010~2020	2000~2020
Supply	Food production	16.128	15.125	15.039	−1.002	−0.086	−1.089
	Production of material	32.819	32.618	32.665	−0.201	0.047	−0.154
	Water supply	15.256	18.092	17.088	2.836	−1.004	1.832
Regulation	Gas conditioning	17.287	16.550	16.227	−0.736	−0.324	−1.060
	Climate regulation	25.359	25.349	24.621	−0.009	−0.729	−0.738
	Clean the situation	25.412	26.331	25.400	0.919	−0.931	−0.012
	Hydrological regulation	385.772	402.002	391.611	16.230	−10.391	5.839
Support	Soil conservation	14.564	14.254	13.836	−0.310	−0.418	−0.728
	Maintain nutrient cycle	2.432	2.288	2.260	−0.144	−0.027	−0.171
	Biodiversity	18.533	19.143	17.736	0.611	−1.408	−0.797
Cultural	Aesthetic landscape	11.420	11.879	11.011	0.459	−0.868	−0.409

) is obtained, and the results show that the overall ESVs in Wuhan continued to fluctuate, and the ESVs of each ecosystem function, except water supply and hydrological regulation function, have different degrees of deficit. Overall, the most dominant ecosystem function in the study area is the hydrological regulation function, whose ESVs in 2020 were CNY 39.161 billion, occupying the highest proportion of the total study area. During the period 2000–2010, the ESVs of hydrological regulation function and water supply increased more with CNY 1.623 billion and CNY 283.6 million, respectively, while the ecosystem services provided by purification, biodiversity and aesthetic landscape functions also increased to some extent, and the ESVs of all other functions decreased to different degrees. During the period 2010–2020, the ESVs of all ecosystem functions except the raw material production function showed a decreasing trend. The hydrological regulation function, however, changed from a large increase in the previous period to a deficit, with a deficit value of CNY 1039.1 million, which was the main reason for the decrease in ESVs in the study area. From the whole period of 2000–2020, the hydrological regulating function profoundly influenced the overall ESVs level in the study area, and its deficit amount also directly explained the change in ESVs in the study area. The lakes and rivers in Wuhan occupy a large area and the water bears a large ecological service function, which has a direct impact on the quality of the ecological environment, while the land use transition in the study area has been a pressure on the protection of the water, resulting in a decrease in the level of the service function that the water can bear.

3.3. Changes of ESVs Spatial Pattern Driven by Land Use Transition

Based on the 30 m × 30 m unit land use transition status of the study area for processing, combined with the fluctuation of ESVs in each region, the hot spot analysis map of ESVs in Wuhan city was drawn based on the data of ESVs during 2000–2020 (Figure 4), and the results showed that during the twenty years from 2000 to 2020, the hot spot areas of ESVs distribution are mainly in areas with the extensive distribution of rivers and lakes, such as Jiangxia, Hongshan, Hannan, Xinzhou, Huangpi, and some parts of the Caidian district, while there are also larger areas along the Yangtze River. There are more concentrated and contiguous hotspots in these areas, which are the main areas of ESVs supply. In Jiang’an, Jianghan, Qiaokou, and Qingshan districts, where the population is large, economic activities are frequent, and the degree of land use is high, there are large areas of cold spots, indicating that human production and living activities affect the structure and pattern of land use while causing fluctuations in ESVs. Meanwhile, in Huangpi, Caidian, Jiangxia, and Xinzhou District, which have more farmland, there are also contiguous cold spots because the ecosystem services provided per unit of farmland are lower compared to other land types. During the period 2000–2020, the overall hot spot areas in Wuhan are decreasing at the 95% confidence level, while the 99% and 90% confidence levels have not significantly changed. In terms of the cold spot areas, the overall trend of change in the 99% confidence level areas is not obvious and has been maintained at a high level. The main reason for the change in the hot spot areas is the decreasing area of ecological lands such as water, forests, and wetlands, which are being squeezed by human activities and are in a state of shrinking ecosystem service functions but are still at a high level overall. A large number of lakes and rivers in the city provide a considerable amount of ecosystem services to Wuhan, but in recent years, various types of ecological land are shrinking, making the protection of the ecological environment challenging.

4. Discussion

This study takes the rapid urbanization as the background, selects the water-rich city of Wuhan as the research object, locates the research time in the last two decades when the land use transition in Wuhan was dramatically changed, explores the spatial-temporal patterns of land use transition by combining the land use transition matrix, and measures ESVs by using the equivalence factor method as a basis to investigate the interaction between the two in depth. Currently, most studies on land use change and its interaction with ESV have focused only on a provincial [49,50], regional [51,52] or municipal [36] ESV change, but less on a region with distinctive land use characteristics to explore the intrinsic relationship between land use transition and ESV. The relationship between land use transition and ESV has rarely been explored in a region with distinct land use characteristics. Under the context of rapid urbanization, population growth and land use change have a significant impact on the ecological environment [53,54]. Among them, land use is considered to be an important factor affecting the supply of ecosystem services. The specific land use pattern of a region is the result of the interaction of socio-economic factors [55]. Urban transportation planning will affect urban development patterns [56], lead to changes in land use patterns, and then affect the supply of ecosystem services. The process of using the research method in this study is more convenient and easier to operate, and the results are easier to compare, while the evaluation results based on the physical quantity method are used as a reference to reduce the errors caused by subjectivity. In contrast, ecosystems are more complex internally, with large differences in the size and type of different service functions. Therefore, to ensure the accuracy of the study, ecosystem types and service function categories should be distinguished to the greatest degree of fineness. The research for some types of ecosystem service functions is relatively lacking, and only some secondary ecosystem classifications can be reasonably adjusted and combined, and the impact of this adjustment method on the research results needs to be further explored. Overall, this method improves the previous equivalence factor method, refines the secondary classification of ecosystems, and reduces the subjectivity of evaluation, making the evaluation results more accurate. At the same time, the method is suitable for the study of larger scale areas, its application requires a small amount of data, the process is convenient, the assessment is comprehensive, and the results are easy to compare, making it a more scientific and mature evaluation method. This study focuses on Wuhan, a water-rich city in a period of developmental change, and explores its land-use transition patterns and changes in ESVs. Water is an important part of the ecosystem and is responsible for providing a large number of ecosystem services. It provides the most ecosystem services per unit area among all land use types and has an important impact on the ecosystem, so the study of ecosystem services in water-rich cities is representative. On the one hand, this study places the study in the context of rapid urbanization, so that the relationship between the study results and the study context can be clearly shown, which is more conducive to clarifying the relationship between the study subjects. On the other hand, taking Wuhan as an example, it is helpful to explore the evolution of ecosystem services in cities with significant land use characteristics such as water-rich cities. The study will help to explore the evolution of ecosystem services in water-rich cities, a category of cities with significant land use characteristics. At the same time, the research results based on the interaction between land use transition and ecosystem services can provide a theoretical reference for exploring the rational planning of urban land use patterns so that urban land use can be developed sustainably and can achieve good ecological conservation effects.

As one of the mega-cities in central China, Wuhan is developing at a high rate, and its economic and social development is remarkable. The economic development and the influx of population have contributed to a strong demand for land for construction, which has inevitably influenced the land use pattern in Wuhan. In the past 10 years, economic development was the main focus of urban development, and as a result, farmland, water, forest and grassland were often in a disadvantaged position in competition with built-up land, which led to a significant decrease in farmland and water area while built-up land continued to expand in Wuhan. The reduction of farmland, forest, wetland and water to a certain extent led to the deterioration of the ecological environment of the city and the reduction of ecological resources for leisure and recreation of the residents, resulting in the decline of the quality of the urban habitat, which hurts the urban environment and sustainable development. In the new era, Wuhan’s land use transition model should strictly control the amount of built-up land, curb its disorderly expansion, and ensure the amount of farmland, water, and other ecological lands.

Although the overall ESVs in Wuhan have increased, it is mainly because water can provide a large proportion of ecosystem services. The ecosystem service function that can be provided by the water per unit area is higher than other land use types. From 2000 to 2010, a large amount of farmland in Wuhan was converted to built-up land and water areas, resulting in a decrease in farmland area and an increase in water area during this period. After 2010, the water area and wetland area in Wuhan decreased. However, the proportion of Wuhan’s water area to its total land area has increased, which is similar to the research result of Liu et al. [57] and Huang et al. [58]. The water area in Wuhan is large and has increased in the past 20 years, so the ESVs of the water area account for the highest proportion and has a great impact, which may be the main reason for the increase of the overall ESVs in Wuhan. However, the water is often at a disadvantage in urban development, and the phenomenon of lake enclosure and development has intensified the squeeze on water space, creating ecosystem service supply challenges. The ESVs of farmland and forest also have some deficit, but the value of the deficit is not large. As for the hot spot areas of ESVs in the study area, Jiangxia, Hongshan, Hannan, Xinzhou, Huangpi, and some parts of the Caidian districts, where extensive water exists, have large concentrated contiguous hot spot areas, while the distribution of hot spot areas shows a shrinking trend due to the reduction of water area. The cold spots are mainly located in the Jiang’an, Jianghan, Qiaokou, and Qingshan districts, which have a high degree of land use development, and in Huangpi, Caidian, Jiangxia, and Xinzhou districts, which have a large amount of farmland. Thus, there is an interactive relationship between land use transition and ESVs change. Given the change of ESVs in the hot spot area, the degree of land use transition in the area should be strictly controlled and the land use transition mode should be optimized so that the area can provide a certain amount of ecosystem service functions while ensuring economic development.

5. Conclusions

Based on remote sensing image data and macroeconomic data, this study explored the land use transition pattern and analyzed its impact on ESVs during 2000–2020, taking Wuhan, a city with significant water-rich characteristics in the context of rapid urbanization, as the research object, and came to the following conclusions:

(1) Farmland, water, and built-up land are the main land use types in Wuhan. The land transition in Wuhan during the rapid urbanization is characterized by decreasing farmland, grassland, and wetland due to the expansion of built-up land area, and the threat of farmland to forest and water space. In the period 2000–2020, with the increasing demand for built-up land in Wuhan, the ecological land of all kinds was decreasing, and the space of farmland and water area was squeezed by built-up land, thus there was a large amount of built-up land encroaching on ecological land, and the total area occupied during this period is 769.18 km². On the other hand, there is a clear trend of mutual transition of farmland and water area during the study period. The area of farmland converted to water and the area transferred from water to farmland were 214.57 km² and 169.59 km², respectively.

(2) The change in ESVs in Wuhan was influenced by the fluctuation of water area, and the overall values of ecosystem services in Wuhan increased from CNY 56.498 billion (USD 8.47 billion) in 2000 to CNY 56.749 billion (USD 8.508 billion) in 2020, with an increase of CNY 251.3 million (USD 37.676 million). Among them, the expansion of the water area led to an increase of CNY 1.223 billion (USD 183.358 million) in ESVs, which is the main influencing factor for the change of ESVs in the study area. Meanwhile, the ESVs of the hydrological regulation function undertaken by the water also increased by CNY 583.9 million (USD 87.541 million) during this period.

(3) Most of the hot spots of the ESVs supply in Wuhan are in areas with a large water area, while there are a large number of cold spots of ESVs supply in areas with more built-up land and farmland distribution. From 2010 to 2020, the hot spots of ecosystem service supply are mainly located in the Jiangxia, Hongshan, Hannan, Xinzhou, Huangpi, and Caidian districts, which have a wide distribution of water area, while the cold spots are mainly located in the Jiang’an, Jianghan, Qiaokou, and Qingshan districts, which have a high degree of land exploitation and frequent production activities, as well as Huangpi, Caidian, Jiangxia and part of the Xinzhou districts, which have a large amount of farmland.

In general, farmland and grassland space in Wuhan has been squeezed by the expansion of built-up land in the context of rapid urbanization. The important role of water areas for ecosystem service provisioning and the increase in water area has led to an increase in overall ecosystem services in Wuhan. At the same time, most of the hot spots of ecosystem service supply are located in the water area, which proves the important contribution of water area to ecosystem service supply. The current situation of the supply of ecosystem services in Wuhan shows that various ecological measures in the region have achieved initial results. In the future, it is still necessary to coordinate the protection of ecological land, such as water areas, and control the expansion of construction land, ensuring the provision of ecosystem services.