Spatial matching relationship between health tourism destinations and population aging in the Yangtze River Delta Urban Agglomeration

In order to address population aging and the healthy China strategy, this paper takes three provinces and one city in the Yangtze River Delta urban agglomeration as the empirical target. Firstly, using the 2010 and 2020 census data, and taking the aging coefficient and the density of the elderly population as the basic research indexes, this paper adopts the center of gravity migration model and the standard deviation ellipse to outline the spatial evolution path of the aging coefficient and the density of the elderly population. Second, using the data of recreational tourism sites in each prefecture-level city of the Yangtze River Delta urban agglomeration from 2016 to 2022, the closest neighbor index is used to judge the overall distribution characteristics of recreational tourism sites, and the judgment results are further verified with the help of Tyson polygonal analysis. Finally, based on the cross-section data of the elderly population in 2020, the spatial distribution of recreation and tourism sites and the elderly population are matched by the geographic concentration index, the center of gravity model and the matching coefficient. The results show that (1) the degree of population aging is deepening, and the characteristics of spatial Agglomeration are obvious, but the development speed among regions is inconsistent; (2) the distribution of health tourism destinations is wide and uneven between regions, showing the characteristics of ‘overall dispersion and local concentration’; (3) the spatial distribution of health tourism destinations and the elderly population shows a certain matching relationship, but the matching degree is not high. With the Yangtze River as the boundary, the distribution of health tourism destinations is ‘high in the south and low in the north’ and assumes a ladder shape from south to north, with obvious regional differences. Overall, health tourism lags behind the development process of aging.


Introduction
Since the 21st century, with the rapid development of the economy and society and the improvement of living standards and medical and health conditions, people's attitude toward reproduction has changed, life expectancy has been extended, and the challenges of population aging have become increasingly serious. According to the seventh national census in 2020, the number of elderly people aged 65 and above in China reached 190.63 million, accounting for 13.5% of the total population. This increase shows that China has entered the middle stage of aging. While developed countries in Europe and the United States have faced the problem of population aging at later stages of economic development, China is facing this challenge at earlier stages. 'Getting old before getting rich' is related to the sustainable and healthy development of the country and is the most serious challenge facing China in the 21st century [1].
With the further improvement of the pension system, the average income of the elderly population shows a steady increase, and under the background of the government's vigorous development of social pension services, the elderly people's demand for quality of life service in old age is increasing [2]. This change in the traditional concept of aging has led to the rapid expansion of elderly consumer groups, as more and more elderly people participate in tourism activities, the travel rate of the elderly is increasing year by year, which forms a certain scale within the elderly tourism market [3]. Influenced by Covid-19, health has become a common pursuit, and the travel focus of the elderly group is also changing, they pay more attention to health, leisure and wellness [4]. Traditional sightseeing tours have not been able to meet the needs of elderly tourists, who seek a healthier tourism experience and tend to choose a type of tourism that integrates entertainment, health and knowledge growth [5]. In response to the dual trend of the aging population and the Healthy China strategy, health tourism, as a new tourism project, has not only been embraced by the elderly, but has also been widely adopted by all sectors of society. However, the degree of aging is different in different regions, thus the demand for health tourism is also different. Blind development and construction of health tourism will inevitably lead to idle and wasted resources. Therefore, it is of great theoretical and practical value to study the spatial distribution of the elderly population and health tourism destinations and analyze their matching degree and spatial characteristics; this approach can promote the development and rational layout of regional health tourism destinations and improve the well-being of the elderly population in the region.
Since the 1950s, scholars have gradually enriched the science of aging from the perspective of geography, promoted the development of the geography of aging, and enabled the study of aging to be carried out in geographical space [6]. With the acceleration of population aging, the research paths and methods on the spatial distribution of the elderly population have gradually matured, and the scale effect has attracted the attention of researchers. Scholars' research on the spatial distribution of population aging is mainly conducted from three scales: large, medium and small. On the large scale, Davies points out the uneven development of global population aging [7]. The pace of population aging in developing countries is significantly faster than that in developed countries, and the extent of aging varies widely among continents [8]. In the 1970s, scholars began to study the spatial distribution of population aging in countries and cities, and most of the results showed that population aging occurred in urban centers or remote rural areas. Moore et al showed that the aging of the Canadian population occurred mainly in more urbanized areas or in remote rural areas [9]. Heleniak et al showed that the elderly population in Russia is unevenly distributed and mainly concentrated in large cities [10]. On the small scale, Shiode et al identified the location of aging communities by empirically analyzing geospatial data from Aichi Prefecture, Japan, and showed that aging communities are mostly located in urban and sparsely populated rural areas [11].
Health care tourism integrates health care, elderly care and tourism and meets the demand for increased consumption in the tourism market while transforming the tourism industry for a new era. Understanding the spatial distribution law of health tourism can support the sustainable and healthy development of additional health tourism services and the planning of health tourism. Scholars have studied the spatial distribution of health tourism destinations at large, medium and small scales. Wang et al pointed out that the national health tourism destinations present a tripartite situation in which the Yangtze River Delta Urban Agglomeration, the Taihang Mountains Henan-Shanxi-Hebei border area and the Karst Plateau Guizhou-Chongqing-Sichuan border area represent a large-scale concentration but small dispersion of services [12]. On the medium scale, Yang et al concluded that the spatial distribution pattern of recreational tourism resources in Fujian Province is 'overall dispersed and locally concentrated' [13]. On the small scale, Xie et al showed that the distribution of recreational tourism sites in Wuhan is closely related to the location of transportation arteries and rivers and lakes [14].
An increase in the elderly population has dramatically increased the demand for social pensions [15][16][17]. In order to ensure the fair distribution of pension resources, some scholars have analyzed the spatial distribution characteristics of pension resources. Tian et al studied the spatial coordination between the distribution of the elderly population and the allocation of elderly service facilities in Changchun from two scales: the administrative district scale and the street scale [18]. Yin et al analyzed the matching relationship between elderly resources and the elderly population in China and its distribution characteristics at the provincial scale and showed that although there is a certain matching relationship between elderly resources and the elderly population, the matching degree is not high [19]. Using the spatial mismatch index, Ma et al calculated that there is a large spatial mismatch in the distribution of elderly resources between administrative districts and the elderly population in Xining [20]. The current rate of population aging is creating increasingly serious challenges, among which the provision of elderly care is prominent. A study by Ji et al showed that the spatial matching degree of elderly care resources and aging in each province has significantly improved in the context of the new era [21].
In summary, the current research results on the spatial distribution of population aging and the allocation of elderly resources are remarkable. However, previous studies have only focused on one aspect, and few scholars have studied health tourism as a special pension benefit when combined with population aging trends. Based on industrial convergence theory and coordinated development theory, this paper constructs the matching coefficient of population aging and health tourism destinations and studies the matching relationship between health tourism destinations and the spatial distribution of the elderly population in the Yangtze River Delta Urban Agglomeration under the background of aging, thus providing a theoretical reference for the coordinated and sustainable development of health tourism in the Yangtze River Delta region.
2. Research area, research methodology and data sources 2.1. Research area In December 2019, the State Council of the People's Republic of China issued the Outline of the Yangtze River Delta Regional Integrated Development Plan, which expanded the scope of the Yangtze River Delta Urban Agglomeration to include all of Shanghai, Jiangsu Province, Zhejiang Province and Anhui Province, for a total of 41 cities (figure 1). The Yangtze River Delta Urban Agglomeration has one of the strongest regional economies in China, and it is also the region with the fastest and largest aging population [22]. According to the data of the seventh national census in 2020, the population aged 65 and over in the Yangtze River Delta Urban Agglomeration is 35.5013 million, accounting for 15.2% of the total regional population, which is 1.7 percentage points higher than the national average in the same period. With the aging of the population, the needs of the elderly are also changing. The retired elderly people have abundant economic conditions and leisure time [22,23]. Tourism, especially health tourism, has become a new trend for the elderly. Therefore, this paper takes 41 cities in the Yangtze River Delta Urban Agglomeration as the research object, and explores the distribution characteristics and matching relationship between their elderly population and their health tourism destinations; these cities not only directly face practical problems but also can provide some reference for the development planning of health tourism in the whole Yangzi River Delta Urban Agglomeration.

Research method 2.2.1. Mean pointer center model
The population gravity center refers to the spatial centroid calculated by taking the population of each subregion in a certain region as the weight; which is an important index to measure the spatial distribution pattern and development characteristics of the population [24]. The gravity center of the population aging coefficient originates from the population gravity center, which is an important index used to analyze the spatial distribution and change characteristics of regional population aging [25]. Although the calculation of the population gravity center can be processed using the spatial statistics tool in Arc GIS software, the formula is still derived from the gravity decomposition and composition rule, and the calculation formula is as follows:  In the formula, x and y¯represent the barycenter coordinates of the elderly population coefficient; x i and y i represent the coordinates of the geometric center of the i prefecture-level city; and p i represents the elderly population coefficient of the i prefecture-level city.
2.2.2. Getis-Ord Gi * Getis-Ord Gi * is a method to study the distribution characteristics of local spatial clustering by identifying the cold spots and hot spots of spatial clustering of variables [26]. The calculation formula is as follows: In the formula, W ij represents the space weight matrix; if the distance between the space position j and the position i is within the critical distance d, the weight matrix W ij is 1; otherwise, it is 0; x i represents the observed value of region i; x j represents the observed value of region j.To spatially distinguish between cold spots and hot spots, the value of G d i ( ) is normalized and calculated as follows: where E G i ( )is the expected value and Var G i ( ) is the coefficient of variation. When the observed value of G i is greater than the expected value and is significant, it indicates that the value around area i is higher and belongs to the hot spot area, indicating that there is an Agglomeration feature of elements in this area. When the observed value of G i is less than the expected value, it indicates that the value around area i is low and belongs to the cold spot area, indicating that the elements are discretely distributed in this area.

Nearest neighbor index and voronoi analysis
The nearest neighbor index refers to the ratio of the average distance of the nearest neighbor points to the average distance under a random distribution and is used to reflect the proximity of point elements in the study area and to judge their spatial distribution types [27]. There are three types of spatial distribution of point elements: cohesive, random and uniform. When the nearest-neighbor index is R 1, < the distribution type of point elements is cohesive, and the smaller the index is, the higher the degree of cohesion is. When R 1, = the point feature distribution type is random; when R 1, > the point feature distribution type is uniform [25]. The calculation formula is as follows: where r 1 is the actual nearest neighbor distance; r E is the theoretical nearest neighbor distance; n is the number of research objects; and A is the area of the study area. Voronoi analysis, namely, Thiessen polygon analysis, uses the nearest neighbor index to judge the spatial distribution type of point elements, so the variation coefficient of the Thiessen polygon area is introduced to verify it. The calculation formula is as follows: where S i is the area of the i polygon, S is the average area of the polygons, and n is the number of polygons.

Gini coefficient
The Gini coefficient is an important method to describe the spatial distribution of discrete regions, and its value range is 0, 1 . ( ) When the Gini coefficient is close to 1, the distribution of geographical elements in the region is uneven, that is, absolute concentration. When the Gini coefficient is close to 0, the geographical elements are evenly distributed in the region. The calculation formula is as follows: where p i represents the proportion of the number of geographical elements in the i prefecture-level city to the total number of geographical elements in the study area; n represents the number of prefecture-level cities in the study area; and C represents the equilibrium degree of the distribution of geographical elements in the study area.

Kernel density estimation
The kernel density estimation takes each grid point as the center, searches the falling point elements according to a certain radius, counts the number of falling point elements, and calculates the density value of each grid point [26]. The calculation formula is as follows: where n represents the number of point elements; h represents the bandwidth; and x x i represents the distance from the estimated point to the sample point.

Matching coefficient
Geographic concentration is an index used to measure the degree of regional concentration of elements; it can not only mark the spatial distribution of elements but also reflect the status and role of a region in the whole region [19]. Considering the regional area, the number of the elderly population and the number of health tourism destinations, the geographical concentration of the elderly population and the geography of health tourism destinations are introduced. The calculation formula is as follows: where pop i represents the elderly population in area i; res i indicates the number of recreational tourist destinations in area i; and ter i represents the land area of region i. On the basis of the geographic concentration index, the matching coefficient between the health resort and the elderly population is constructed, and its calculation formula is as follows: When RI 1, > it shows that the Agglomeration of regional health tourism is ahead of the Agglomeration of the elderly population. When RI 1, = it shows that the distribution of regional health tourism destinations is completely matched with the distribution of the elderly population. When RI 1, < it shows that the Agglomeration of regional health tourism lags behind the Agglomeration of the elderly population.

Classification system and classification standard
The main indicators of population aging in demography are the aging coefficient, median age, longevity level, old-age population density, old-young ratio, composite aging index, etc [28]. This paper selects the aging coefficient to measure the degree of population aging and its spatial and temporal changes in the Yangtze River Delta Urban Agglomeration. According to the actual situation of population aging in the Yangtze River Delta Urban Agglomeration and the classification of population aging by Xu et al [29] and Chen et al [30], the population aging in the Yangtze River Delta Urban Agglomeration is divided into four stages. When PA 7 10,  < aging is in the initial stage; when aging is in the middle stage; when aging is in the late stage; and when PA 20, > aging is in the superaging (table 1). The coefficient of the matching degree between the health tourism destination and the elderly population is the ratio of the regional health tourism destination proportion to the elderly population proportion; it can depict the spatial matching between the regional health tourism destination and the elderly population. Based on previous studies and combined with the specific situation of the matching degree between the health tourism destination and the elderly population in the Yangtze River Delta Urban Agglomeration, the matching degree is divided into three situations (table 2): RI 0.8 < is a resource lag type, which means that the Agglomeration degree of health tourism destinations lags behind the Agglomeration degree of the elderly population, and the matching degree between health tourism destinations and the elderly population in this area is low; is the basic matching type, which means that the Agglomeration degree of health tourism destinations is about the same as that of the elderly population, and the spatial distribution of health tourism destinations and the elderly population in this area is basically matched; and RI 1.2  is the resource advanced type, which means that the Agglomeration degree of health care tourism is ahead of the Agglomeration of the elderly population, and the service advantage of health care tourism in this area is obvious [16].

Data source
The data of the elderly population involved in this paper are mainly from the statistical bulletins of the sixth national census and the seventh national census of Shanghai, Jiangsu, Zhejiang and Anhui provinces and cities, and some of the data are from the statistical yearbooks of each province and city. Descriptive statistics of Aging population are shown in table 3. Based on the strictness of the selection criteria and the authority of the institution, the data of health tourism destinations in this paper is taken from the 'China National 3. Characteristics of population aging in the Yangtze River Delta Urban Agglomeration 3.1. Spatial and temporal distribution characteristics of population aging Analysis table 4 and figure 2, we can get that, from 2010 to 2020, the degree of population aging in the Yangtze River Delta Urban Agglomeration deepened over time and changed rapidly from the middle stage of aging to the late stage of aging. The spatial distribution pattern shows a dynamic Agglomeration change, and the degree of aging expands from south to north and from west to east. Among the affected areas, Nantong City and Taizhou City are the high-value distribution areas of population aging, while the surrounding areas such as Shanghai, northern Jiangsu Province, southwestern Zhejiang Province and southwestern Anhui Province have a higher degree of aging; a relatively low degree of aging is concentrated in the southeastern part of the Yangtze River Delta Urban Agglomeration.
From a time perspective, from 2010 to 2020, the rate of population aging in the Yangtze River Delta Urban Agglomeration increased, and the overall increasing rate was faster, but there were significant differences in the degree of aging among regions. In 2010, only Zhejiang Province was in the early stage of aging, while Shanghai and Anhui Provinces were in the middle stage of aging. As a whole, Jiangsu Province was in the early and middle stages of aging and was developing rapidly toward the late stage of aging. By 2020, more than half of the cities in the Yangtze River Delta Urban Agglomeration had entered the late stage of aging. In terms of the aging category, Jiangsu Province was the first to enter the super-aging stage; in terms of development speed, the number of prefecture-level cities in Anhui Province entering the late aging stage increased from 0 to 12 in 10 years, much faster than the other three provinces and cities (table 4).
From a spatial perspective, the overall spatial distribution pattern of population aging in the Yangtze River Delta Urban Agglomeration showed a dynamic Agglomeration change from 2010 to 2020, with the degree of  aging expanding from south to north and from west to the east and forming a 'Nantong-Taizhou' double-peak aging center with uneven spatial distribution. In general, the speed of aging development was faster in northern cities, while the speed of aging development was slower in southern cities, especially some cities in Zhejiang Province. In 2010, the northern cities of the Urban Agglomeration took the lead in entering the middle stage of aging, while the southern region was still in the early stage of aging. By 2020, the northern cities still had a higher degree of aging, entering the late stage of aging, and some prefectures had entered the super-aging stage. After 10 years of development, the southern cities entered the middle stage of aging. The formation and development of the spatial pattern of aging in the Yangtze River Delta Urban Agglomeration is directly related to the gap in the economic development level within the Urban Agglomeration (figure 2). Analysis table 5 and figure 3, we can know that, from 2010 to 2020, the center of gravity of the elderly population coefficient of the Yangtze River Delta Urban Agglomeration was distributed in the direction of 'northwestsoutheast', and it always fell in Nanjing during the study period, and gradually shifted to the southeast, but did not coincide with the geometric center of gravity of the study area (118°55' 24' E, 31°29' 17' N). Specifically, in 2010, the center of gravity of the elderly population coefficient of the Yangtze River Delta Urban Agglomeration was located in the west of Nanjing Sushui district and shifted 5.32 km to the northeast of the geometric center of gravity. In 2020, the center of gravity of the elderly population coefficient shifted to the southeast of the Sushui district, with a moving distance of 1.98 km and a geometric center of gravity of 4.72 km. This progression shows that the degree of population aging in the Yangtze River Delta Urban Agglomeration is unevenly distributed, and the development process of population aging in the eastern region is faster than that in the western region. Over time, with the support of national policies, the Yangtze River Delta Urban Agglomeration have become the city cluster with the strongest comprehensive strength in China. The core cities within the region, Shanghai, Nanjing, Hangzhou, Suzhou and Ningbo, are located in the southeast of the city cluster. These cities have good economic development and a high level of urbanization, and their economic advantages indirectly influence the population aging process by affecting the life expectancy and fertility rate of the population [31]; thus, the center of gravity of the population coefficient shows a tendency to shift to the southeast.

Spatial correlation characteristics of population aging
This paper uses the cold-hot spot analysis method to identify the high-value and low-value areas of population aging and explores the relevance of the spatial distribution of the elderly population in the Yangtze River Delta Urban Agglomeration from 2010 to 2020 (table 6). The local spatial correlation results of the aging coefficient of the Yangtze River Delta Urban Agglomeration in 2010 and 2020 are divided according to the Jenks natural segment breakpoint method, and the cold spot area, the sub-cold spot area, the sub-hot spot area and the hot spot area are in turn from low to high, as shown in figure 4. From the figure, we can see that the cold spot and hot spot patterns of the aging coefficient of the Yangtze River Delta Urban Agglomeration are obviously different between the east and the west, and the development between the north and the south is unbalanced. From 2010 to 2020, the aging coefficient of the Yangtze River Delta Urban Agglomeration always takes the northeastern cities such as Taizhou and Nantong as the core hot spots and the southeastern cities such as Hefei, Suzhou, Hangzhou and Ningbo as the cold spots. The spatial distribution of sub-cold spots and sub-hot spots varies widely: in 2010, the sub-cold spots are widely distributed, and connected from south to north, and in 2020, the cold spots are reduced and scattered in the north, central and southeastern regions. In 2010, the sub-hot spots are scattered along the northeast-southwest line, and in 2020, the sub-hot spots are concentrated in the western region. The distribution of cold and hot spots of the aging coefficient in the Yangtze River Delta Urban Agglomeration is not consistent with the level of economic development. For example, economically developed cities with Shanghai as the core have not formed a high quality agglomeration of aging but as a transition zone from cold spots to hot spots.  4. Spatial distribution characteristics of health tourism destinations in yangtze river delta urban agglomeration

Spatial distribution pattern of health tourism destinations
The geographical coordinates of 272 health tourism destinations in the Yangtze River Delta Urban Agglomeration are obtained by using the Baidu Picking Coordinate System, and the spatial data are vectorized by using ArcGIS software to obtain the spatial distribution map of health tourism destinations in the Yangtze River Delta Urban Agglomeration (see figure 5). Figure 5 shows that the health tourism destinations in the Yangtze River Delta Urban Agglomeration are widely distributed, but the distribution among regions is uneven, showing the characteristics of 'overall dispersion and local concentration'. With the Yangtze River as the obvious demarcation line, to the south of the Yangtze River is a relatively concentrated area of recreational tourism,  forming a number of agglomeration centers. North of the Yangtze River, the distribution of health tourism destinations is slightly scattered, and the agglomeration characteristics are not obvious. In general, Zhejiang Province has the largest number of recreational tourism destinations, up to 121, accounting for 44.49% of the total number of regions, which is far ahead of other provinces and cities by nearly half. Next is Anhui Province, which has 75 health tourism destinations, accounting for 27.57% of the total. Jiangsu Province has 69 health tourism destinations, accounting for 25.37% of the total. The number of recreational tourism destinations in Shanghai is the lowest, only 7, accounting for 2.57% of the total number of regions.

Spatial distribution types of health tourism destinations
To explore the spatial agglomeration of health tourism destinations in the Yangtze River Delta Urban Agglomeration, the nearest distance index of 272 health tourism destinations in the study area was analyzed by using the average nearest neighbor module of the ArcGIS spatial analysis tool according to the nearest index formula. As showed in figure 6, the results show that the observed average distance is 15.93 km, the expected average distance is 20.42 km, and the nearest neighbor distance index is 0.78, which is less than the critical value of 1. The Z score is −6.94, and the P value is 0. The significance level test shows that the spatial distribution of health tourism resources in the Yangtze River Delta Urban Agglomeration has significant agglomeration. To verify the results of the nearest neighbor index, the Tyson polygon of the health resort was created, and its area was calculated. The average area of the Tyson polygon is 1324.96 km 2 , the standard deviation is 1477.41 km 2 , and the coefficient of variation CV is 112%. According to the study of Duyckaerts [32] et al, when CV 33%, < healthy tourism destinations are evenly distributed; when CV 33% 64%, < < the distribution of health tourism places is random; and when CV 64%, > the health tourism destinations show an Agglomeration distribution. The CV value of the coefficient of variation of recreation tourism sites in the Yangtze River Delta Urban Agglomeration is 112%, which is much larger than 64%, again verifying that the overall spatial distribution of recreation tourism sites in the study area has significant clustering.

Balance of the spatial distribution of health tourism destinations
The proximity index analyzes the spatial distribution types of health tourism destinations in the Yangtze River Delta Urban Agglomeration from the provincial perspective, and the Gini coefficient is introduced to further explore the distribution types of health tourism destinations within the urban agglomeration. It is calculated that H 3.41, m = H 3.71, = G 0.92, ini = and C 0.08. = According to the definition of the Gini coefficient, the closer the value is to 1, the higher the concentration of resources is. Therefore, the concentration of health tourism destinations in the Yangtze River Delta Urban Agglomeration is high, and the distribution within the Urban Agglomeration is unbalanced. To further verify the balance of the spatial distribution of health tourism destinations, the Lorenz curve of the spatial distribution of health tourism destinations in the Yangtze River Delta Urban Agglomeration is drawn ( figure 7). From the figure below, we can see that the Lorenz curve is concave, indicating that the spatial distribution of recreational tourism destinations in the urban agglomeration is not balanced. Huzhou, Jinhua, Suzhou and Hangzhou account for 1/4 of the total number of recreational tourism destinations in Urban Agglomeration, while Zhoushan City, Tongling, Ma'anshan and Huainan account for less than 5% of the total number of recreational tourism destinations. By 2022, Suzhou, Huaibei and Bengbu have no distribution of recreational tourism. This difference shows that the spatial distribution of health tourism destinations in the Yangtze River Delta Urban Agglomeration is obviously unbalanced, mainly distributed in the southern part of the urban agglomeration, especially in the southeastern part of Zhejiang Province.

Analysis of core density of recreation tourism
To visually display the spatial distribution characteristics of health tourism destinations in the Yangtze River Delta Urban Agglomeration, the density of 272 health tourism destinations is analyzed with the help of the core density analysis module in the ArcGIS spatial analysis tool, and the core density distribution map of health tourism destinations in the Yangtze River Delta Urban Agglomeration is generated ( figure 8). It can be seen from the figure that there are obvious differences in the distribution of the core density of health tourism within the  urban agglomeration, which shows the spatial distribution characteristics of 'large dispersion, small concentration' as a whole. The figure shows one high-density core area, six sub-density core areas and three zonal connecting areas, showing the distribution law of 'around the city, near the scenery'.
The high-density core area with Huzhou and Hangzhou as the core is located at the northwestern junction of Anhui Province and Zhejiang Province, which is the most dense distributed area of health tourism within the urban agglomeration. Shanghai, Suzhou, Nanjing, Anqing, Xuancheng and Jinhua are the cores of the six subdensity core areas, which are fan-shaped around the high-density core area. From north to south, the three belts are the Xuzhou-Lianyungang-Yancheng belt, Jiaxing-Shaoxing-Ningbo-Zhoushan belt and Quzhou-Lishui-Wenzhou-Taizhou belt. Among these belts, the northernmost Xuzhou-Lianyungang-Yancheng belt area has the lowest number and the lowest density of health tourism destinations, which is in line with the overall characteristics of the Yangtze River Delta Urban Agglomeration. The central Jiaxing-Shaoxing-Ningbo-Zhoushan belt area is surrounded by Hangzhou Bay in the shape of a 'C', which belongs to the medium-density area of health tourism. The southern Quzhou-Lishui-Wenzhou-Taizhou belt area is distributed in a 'V' shape as a whole, with a high proportion of mountains and hills, which is the highest forest coverage area in Zhejiang Province.

Spatial matching relationship between the elderly population and health tourism destinations in the yangtze river delta urban agglomeration
To analyze the matching relationship between health tourism destinations and the elderly population in the spatial distribution of the Yangtze River Delta Urban Agglomeration, the matching coefficient RI is introduced. The matching coefficient of the spatial distribution of health tourism destinations and the elderly population in the provinces and cities of the Yangtze River Delta Urban Agglomeration in 2020 is calculated, and it is divided into three types according to the classification criteria: resource lag type, basic matching type and resource advanced type. The results are shown in table 7 and figure 9. Table 7 shows that from the provincial perspective, the distribution of health tourism destinations and the distribution of the elderly population in the Yangtze River Delta Urban Agglomeration are extremely unbalanced. In 2020, there is only one province in Anhui Province that basically matches the spatial distribution of the elderly population, with a matching coefficient of 1.87, accounting for 1/4 of the total number of provinces in the Yangtze River Delta Urban Agglomeration, with a proportion of 28% for health tourism and 26% for the elderly population. Accordingly, the remaining three provinces in the Yangtze River Delta Urban Agglomeration do not match the spatial distribution of the elderly population. Among these provinces, there is one resource-advanced province, Zhejiang Province, with a matching coefficient of 1.84; this province has the highest spatial matching coefficient between the health tourism destination and the elderly population in the Yangtze River Delta Urban Agglomeration. In 2020, the proportion of the elderly population in Zhejiang Province was 24%, while the proportion of health tourism in the region was 44%, which is nearly half of the region. There are two resource-rich provinces, namely, Shanghai and Jiangsu, of which the matching coefficient of Shanghai is 0.23 and that of Jiangsu is 0.66. The proportion of the elderly population in Shanghai is 11%, but the proportion of health tourism destinations accounts for only 0. 03% of the whole region, and the spatial distribution of health tourism destinations and the elderly population is extremely mismatched. The proportion of the elderly population in Jiangsu Province is 39%, and the proportion of recreational tourism is 25%. Therefore, from the perspective of regional characteristics, there are obvious regional differences in the spatial matching degree between the health tourism destinations and the elderly population in the Yangtze River Delta Urban Agglomeration.
It can be seen from figure 9 that, from the perspective of the city, the spatial matching coefficient between the health tourism destinations and the elderly population in the Yangtze River Delta Urban Agglomeration is roughly bounded by the Yangtze River, which is 'high in the south and low in the north' and is distributed in a ladder shape from south to north, with obvious spatial differences. In 2020, only 10 cities basically match the Topographically, these cities are located in the North China Plain and the middle and lower reaches of the Yangtze River Plain, with a single terrain and lack of natural conditions for the development of health tourism. The development of health tourism in the region lags behind the growth rate of the elderly population.
The classification results show that the distribution of the health tourism sites and the elderly population is extremely unbalance. The proportion of the elderly population in 15 cities with advanced resources is only 26.37%, while the proportion of health tourism destinations is as high as 57.72%. In contrast, the proportion of the elderly population in 16 cities with lagging resources is 45.46%, which is close to half of the total elderly population in urban agglomerations, while the proportion of recreational tourism is only 15.07%. From the perspective of regional characteristics, the spatial matching degree of health tourism destinations and the elderly population in the Yangtze River Delta Urban Agglomeration has obvious regional differentiations. The cities with general natural resource endowment in the north of the Yangtze River and the cities with a high density of the elderly population in the estuary of the Yangtze River are mainly distributed in the resource lag area. The cities with relatively stable changes in the elderly population along the Yangtze River are mainly distributed in the resource-matching areas. The cities with excellent natural resources and a low density of the elderly population in the south of the Yangtze River are mainly distributed in the resource-advanced areas. However, whether an area is resource-advanced or resource-lagging, it is in a state of mismatch between the distribution of health tourism and the elderly population. From the perspective of regional coordinated and healthy development, the health tourism of the Yangtze River Delta Urban Agglomeration is not in an ideal state as a whole; it is not conducive to the fair enjoyment of health tourism resources and the equalization of pension services by the elderly population in the region.

Conclusions
This paper mainly studies the matching relationship between population aging and health tourism destinations and draws the following conclusions.
(1) The aging coefficient of the Yangtze River Delta Urban Agglomeration shows the spatial distribution characteristics of 'high in the north and low in the south' from 2010 to 2020; that is, the aging coefficient in the north of the Yangtze River is higher, and there is a 'Nantong-Taizhou' aging center of gravity, while the aging coefficient in the south of the Yangtze River is relatively low. The aging type has leapt from the middle to the late stage of aging. In the spatial pattern, the center of gravity of the aging coefficient of the Yangtze River Delta Urban Agglomeration generally moves from northwest to southeast, showing a spatial development pattern of 'moving from south to east'. The pattern of cold and hot spots of the aging coefficient is obviously different between the east and west, and the development of the north and south is unbalanced, demonstrating substantial changes during the 10-year period.
(2) Spatial distribution characteristics of health tourism destinations: The health tourism destinations in the Yangtze River Delta Urban Agglomeration are widely distributed, but the distribution among regions is unbalanced, showing the characteristics of 'overall dispersion, local concentration'. The spatial agglomeration characteristics are significant, but the spatial distribution within the region is uneven. The results of nuclear density analysis show that the health tourism destinations in the Yangtze River Delta Urban Agglomeration have the spatial distribution characteristics of 'large dispersion, small concentration' and the distribution law of 'surrounding the city, near the scenery'.
(3) Spatial distribution matching degree between health tourism destination and the elderly population: There is a certain matching relationship between health resorts and the elderly population in the Yangtze River Delta Urban Agglomeration, but the matching degree is not high. Overall, health tourism lags behind the development process of the aging population. The spatial matching relationship between health tourism destinations and the elderly population in the Yangtze River Delta Urban Agglomeration has obvious regional characteristics, roughly bounded by the Yangtze River, 'high in the south and low in the north', with a ladder-like distribution from south to north and obvious spatial differences.
Rational allocation of pension resources is an important research topic in the fields of population geography and geriatric sociology, etc. As a special kind of pension resources, the rational allocation and layout of recreation and leisure tourist sites also have important value for the equalization of pension services. However, this paper still has some limitations: (1) Considering the accessibility of the research data. The study based on prefecture-level cities is slightly macroscopic in scale, and may not be able to carefully and comprehensively portray the spatial distribution characteristics of population aging and recreation and tourism sites in the region and their matching relationship. If the research perspective is devolved to the county or township scale in the follow-up study, the results may be more refined. (2) This paper only uses the cross-section data of the elderly population in 2020 to analyze the matching relationship between the spatial distribution of health tourism sites and the elderly population, and the time scale is relatively small, so the time evolution study needs to be increased. If the follow-up study can expand the time scale, in-depth study of the dynamic evolution of the match between health tourism sites and the elderly population, it may reveal the reasons for the imbalance between the development of health tourism and the elderly population. (3) The study of influencing factors needs to be in-depth. The article is only based on ArcGIS spatial analysis to study the spatial distribution of the elderly population and recreation tourism sites and the spatial match between the two, and has not carried out in-depth excavation of their influencing factors. Based on the factors affecting the development process of aging and recreation tourism, how to find a path for the coordinated and sustainable development of health tourism and the elderly population, and alleviate the economic and social pressures brought by aging is still a direction worthy of research in the future.