Spatial Distribution Characteristics of Japan’s Forest Therapy Bases and Their Influencing Factors

Abstract

Forest therapy bases are essential ecological recreational locations to promote mental and physical health while at the same time allowing people to enjoy and appreciate the forest. The article took Japan, where the development of forest therapy is in a mature stage, as the research object. Using multi-data and the methodologies of Kernel Density Analysis in ArcGIS and GeoDetector, the spatial distribution characteristics of Japan’s forest therapy bases were investigated, as well as the influencing factors. The results reveal that the spatial distribution of forest therapy bases in Japan is unbalanced, with an aggregated distribution in economically developed and densely populated areas. The spatial density of natural landscape resources, Gross Domestic Product (GDP), the spatial density of population, distance from three major metropolitan areas, rail transit ridership, and spatial density of tourism resources are factors influencing the distribution of forest therapy bases in Japan. The factors interact with each other, forming the spatial distribution pattern. Among these factors, GDP has the greatest explanatory power for the spatial distribution of forest therapy bases in Japan, followed by the distance from Japan’s three major metropolitan areas and spatial density of tourism resources, while the spatial density of population, spatial density of natural landscape resources, and rail transit ridership have a relatively weaker influence on forest therapy bases in Japan. The findings provide some insight into the macroscopic layout of forest therapy bases in various regions of different countries, demonstrating that excellent transportation facilities and good natural resources are the fundamental considerations for the location of forest therapy bases and that densely populated urban areas with a strong economic foundation are key areas for the development of forest therapy bases. Additionally, to take advantage of industrial agglomeration and synergize regional development, considerations for merging with existing resources, such as national parks, forest parks, and recreation forests, should be made.

1. Introduction

According to the website of the Forest Therapy Society in Japan (https://www.fo-society.jp/en/forests.html (accessed on 20 March 2022)) [1], a “Forest Therapy Base” is an area located in a forest where relaxing effects have been observed based on scientific analysis conducted by a forest medical expert. Moreover, it is an area where nature merges with society permitting people to come together and partake in social activity surrounded by a natural environment. In other words, a forest therapy base (https://www.fo-society.jp/en/index.html# (accessed on 20 March 2022)) [2] is a vital ecological recreational place with the aim of promoting mental and physical health and improving disease prevention while at the same time allowing individuals to enjoy and appreciate the forest.

As the first element of forest therapy, a “Forest Therapy Base” is an important support in terms of theory and practice [3]. Therefore, it is necessary to conduct relevant research on forest therapy before exploring forest therapy bases. Forest therapy is a nature-based intervention that considers the specific needs of individuals and the natural and social environment in which they live (https://foresttherapyhub.com/what-is-forest-bathing-and-forest-therapy/ (accessed on 20 March 2022)) [4], which is now increasingly recognized as an effective relaxation and stress management tool [5,6,7]. Although the concept of “forest therapy” was first proposed in Germany [8], and its positive effects have been applied by European countries for many years, according to comparative research [9], East Asian countries have a high level of interest in this research, and the top three countries with influence are Japan, Korea, and China. Due to the side effects of the technology boom on urban residents, Japan, which is rich in forest resources, coined the term Shinrin-yoku, which translates in English as “Forest Bathing” [10]. This concept was developed by the Japanese Forestry Agency in 1982] [11] and means to feel and experience the rich and colorful scenery, smells, sounds, etc., formed by the natural environment of the forest through the five senses of seeing, hearing, touching, smelling, and tasting, so as to promote people’s physical and mental recovery and maintain a healthy state [12]. In 2003, the Japanese Forestry Agency put forward the concept of “Forest Therapy”. In 2004, Japan established the Forest Therapy Society, and the earliest forest therapy base was established in 2005 [13,14]. As of 2022, Japan has certified 65 forest therapy bases, forming certain industrial advantages and the industrial development is mature. Therefore, this article took 65 forest therapy bases as the research object, which was much more comprehensive and systematic.

Through correlation analysis, the concept of forest therapy is still relatively new in some countries [11]. It is found that the research on forest therapy can be divided into three aspects: Empirical research on forest therapy benefits [15,16,17,18], research on forest therapy bases [19], and research on natural experience activities [20], among which the largest proportion is the empirical research on forest therapy benefits [21]. There are numerous studies pertaining to the effects of forest therapy on human health, from physiological and psychosocial perspectives to its potential in treating specific illnesses such as hypertension and depression [11]. Forest therapy could be an effective tool in addressing stress and improving the overall health of middle-aged women [22,23], while also having the potential to be a cost-effective form of preventive medicine across all age groups and genders [11]. Therefore, research on therapy experience plays an important role in forest therapy. These experience activities usually rely on recreational facilities or recreational items set up in the forest to help people release stress and boost their spirits in the natural environment.

There are, however, relatively few studies related to forest therapy bases. According to the literature analysis [9], the main contents of studies on forest therapy bases include a general overview of Japanese forest therapy bases [24], a specific case analysis of forest therapy bases [25], and research on walks and landscapes of forest therapy bases [19]. Among the selected valid essays or articles in the literature analysis [9], the research on the self-construction and self-development status of forest therapy bases in various countries is relatively popular. It can be found that by sorting out the development history of forest therapy bases in countries with relatively mature experience models, rich experience can be provided as a reference. In addition, some scholars have explored the design methods of forest therapy bases [3], including their construction conditions, construction goals, infrastructure construction, and supporting facilities construction, through case analysis of one single established forest therapy base. A deeper understanding of the spatial distribution of multiple forest therapy bases at the national level, however, remains to be explored.

To address the aforementioned knowledge gaps, this article uses multivariate data, based on the ArcGIS platform, using Kernel Density Analysis and GeoDetector methods. Specifically, there are three main goals: (1) To study the spatial distribution of Japan’s forest therapy bases, (2) to analyze the influencing factors of Japan’s forest therapy bases, and (3) to explore the mechanism of these factors influencing the spatial distribution of forest therapy bases in Japan. Based on these objectives, this article could contribute to the theoretical advancements in the overall industrial layout of forest therapy bases and find suitable development paths.

2. Materials and Methods

2.1. Study Area

Most of the forest therapy bases in Japan were built in Japanese national parks, quasi-national parks, or prefectural natural parks, which offer excellent natural conditions. Most of these are natural forests, a few are mixed natural and artificial forests, and very few are artificial forests.

According to the local zoning of prefectures in Japan, there is one forest therapy base in Hokkaido with special resources such as Primula japonica and wetlands; one forest therapy base in Okinawa with subtropical forests; six forest therapy bases in Tohoku with forest stands mostly centered on beech forests; 14 forest therapy bases in Kanto mostly in mixed forests; and 17 forest therapy bases in Hokuriku and Koshinetsu, most of which are broad-leaved or mixed coniferous forests with Fagus crenata as the centerpiece. In the Tokai, Kansai, and Chugoku regions, there are three, four, and six forest therapy bases, respectively, mostly with mixed forests. In the Shikoku region, there are two forest therapy bases, with species such as Quercus serrata, Pinus sylvestris, and Stewartia monadelpha. In the Kyushu region, there are ten forest therapy bases, some of which are leafy forests and very few are artificial coniferous forests. Thus, on the whole, the Japanese forest therapy bases have better overall resources and more advantages than ordinary woodlands in terms of species structure, hierarchy, and diversity, and some have local characteristics.

2.2. Data Source

GDP (Gross Domestic Product) data for prefectures (2015) were obtained from the Statistics Bureau of Japan (https://www.e-stat.go.jp/ (accessed on 20 March 2022)) [26], and the data used to analyze the factors influencing the distribution of forest therapy bases were obtained from the website of the Ministry of Land, Infrastructure, Transport, and Tourism (https://nlftp.mlit.go.jp/ksj/ (accessed on 20 March 2022)) [27]. These contained data related to the region, population economy, transportation, resources, etc. Region-related data included regional data (2020) and administrative area data (2020); population economy-related data included population concentration area data (2015) and 1 km grid population data (2010); transport-related data included airport data (2018), rail station and ridership data (2019), and bus route data (2017); and resource-related data included natural park type and distribution data (2015), natural landscape resource data (2012), and tourism resource data (2014). The data type for regional data, administrative area data, and tourism resources is Shapefile. The spatial resolution of population data and rail transit ridership is 1 km and the spatial resolution of kernel density value is 10 km.

In this study, the 64 forest therapy bases (including the forest therapy path) published on the website of the Forest Therapy Society in Japan were used as the research objects (due to the suspension of activities at Yamazaki Forest Therapy Base in Hokkaido, it was not taken into account). Google Earth software was used to verify the coordinate positioning one by one, and the ArcGIS platform was used to build the geographic information database of Japan’s forest therapy bases. In order to build a database of geographic information about the distribution of forest therapy bases in Japan, the number of forest therapy bases in each region and the administrative district was first counted and spatially linked to forest therapy bases, airports, and rail transit stations, respectively. The population concentration area data were merged to screen out the densely populated areas of the three major metropolitan areas. The population data, rail transit ridership, and kernel density of tourism resources were calculated, and the spatial distribution density value of each element was determined.

2.3. Methods

The characteristics of the distribution of forest therapy bases in Japan were described in this study using spatial statistical analysis. The neighborhood analysis was utilized to demonstrate the relationship with each probable influencing factor, and GeoDetector was used to justify the correlation between spatial distribution and each potential influencing factor.

2.3.1. Kernel Density Analysis

In this study, Kernel Density Analysis [28] was used to calculate and visualize the distribution density of Japan’s forest therapy bases, and their spatial density characteristics were explored with the calculation formula in Equation (1):

$$f\left(x\right)=\frac{1}{n\times h}{{\displaystyle \sum }}_{i=1}^{n}k\left(\frac{x_i-x}{h}\right),$$

(1)

where f(x) is the kernel density estimate; n is the number of Japan’s forest therapy bases; x_i is the number of independently and identically distributed Japan’s forest therapy base observations; x is the mean value; k is the kernel function; and h is the bandwidth.

2.3.2. GeoDetector

Based on the spatial averaging technique and set theory, GeoDetector [29] is a spatial analysis method to detect the consistency of the spatial distribution of variables being explained and their explanatory factors, and the GeoDetector principle ensures that it is immune to multivariate covariates, being widely used in the fields of the regional economy, regional planning, and tourism [29,30]. In this study, the influencing factors and influences that affect the spatial distribution pattern of forest therapy base density in Japan were explored with the help of GeoDetector, as calculated in Equation (2) [31]:

$$q_{X,Y}=1-\frac{1}{N{\sigma }^2}{\displaystyle \sum _{h=1}^L}{N}_h{\sigma }_h{}^2=1-\frac{SSW}{SST}$$

(2)

$$SSW={\displaystyle \sum _{h=1}^L}{N}_h{\sigma }_h{}^2,SST=N{\sigma }^2$$

(3)

where X is the influencing factor; Y is the spatial distribution of Japan’s forest therapy bases; q_X,Y is the explanatory power of X to Y; L is the strata number of the influencing factor X. N_h and ${\sigma }_h{}^2$ are the number of elements and variance of Strata h, respectively, and N and ${\sigma }^2$ are the number of elements and variance of the entire study area, respectively. SSW and SST represent the Within Sum of Squares and the Total Sum of Squares. q∈[0,1], and the larger the value of q, the stronger the explanatory power of the influencing factor X to the spatial distribution of Japan’s forest therapy bases, and vice versa.

3. Results

3.1. Spatial Distribution Characteristics of Japan’s Forest Therapy Bases

According to the website of the Forest Therapy Society in Japan, there are 64 forest therapy bases in Japan (as of September 2021) (Figure 1), ranging from Kunitou Village in Okinawa in the south to Tsubetsu Town in Hokkaido in the north, with the Kanto, Hokuriku-Koshinetsu, and Kyushu regions having the highest number of forest therapy bases and the Hokkaido region having the lowest. Of the 47 prefectures in Japan, Nagano Prefecture (10) in Hokuriku-Koshinetsu has the highest number of forest therapy bases, accounting for 58.8% of the total number of forest therapy bases in the region, followed by Kanagawa Prefecture (5) and Gunma Prefecture (4) in the Kanto region, together accounting for 64.2% of the total number of forest therapy bases in the region.

The Kernel Density Analysis was used to calculate the density values of Japan’s forest therapy bases nationwide, and the results (Figure 2) show that there is one high-density core area located in central Japan, extending from the Tokyo metropolitan area to the northwest along the Hokuriku-Koshinetsu region. Meanwhile, there are two micro-aggregations located in the Kyushu region in southwestern Japan, close to the two largest cities in Kyushu (Fukuoka and Miyazaki). In terms of the ratio of the area to the number of bases, the Kanto and Hokuriku-Koshinetsu regions are the most densely distributed, followed by Kyushu, while Hokkaido and Tohoku are the most sparsely distributed.

3.2. Influencing Factors of the Spatial Distribution of Forest Therapy Bases in Japan

3.2.1. Selection of Influencing Factors

Forest therapy requires a forest environment, service facilities, and socio-economic aspects [32]. This study combined landscape planning experience with related research results [33,34,35,36] and selected ten factors across four dimensions: Natural resources, population economy, transportation resources, and tourism resources for analysis (

Table 1. Influencing factors of the spatial distribution of Japan’s forest therapy bases.

Indicator Dimensions	Detection Factors	Indicator Explanation	Spearman Correlation	Sig. (2-Tailed)
Natural resources	Spatial density of natural landscape resources	Kernel density value calculated based on the spatial distribution of natural landscape resources, such as mountain landscape, volcanic landscape, land landscape, limestone landscape, lake and marsh landscape, coastal landscape, river landscape, water landscape, etc.	0.553 **	0
	Natural resource grade	Forests are classified as World Natural Heritage, National Park, Quasi-National Park, Prefectural Natural Park, and other forests in descending order of rank.	0.142	0.264
Population economy	Gross domestic product	Gross domestic product of the prefectures in which the bases are located	0.361 **	0.003
	Population spatial density	Kernel density value calculated based on the spatial distribution of population numbers	0.347 **	0.005
	Distance from three major metropolitan areas	Euclidean distance between the base and the nearest densely populated area of the metropolitan area	0.669 **	0
	Distance from major cities	Distance of the base from the city center of the nearest top 20 cities	−0.069	0.59
Transportation resources	Distance from rail transit stations	Distance of the base from the nearest rail station	−0.163	0.199
	Density of bus routes	The total length of bus routes within 5km of the base	0.242	0.054
	Rail transit ridership	Kernel density value based on the spatial distribution of the average daily passenger flow (person/day) at each rail station	0.331 **	0.007
Tourism resources	Spatial density of tourism resources	Kernel density value calculated based on the spatial distribution of tourism resources, such as parks, temples, places of interest, etc.	0.539 **	0

Note: ** Significant correlation at 0.01 level (2-tailed).

). The natural resources factor includes two factors: Spatial density of natural landscape resources and natural resource grade, of which information on natural landscape resources is extracted from the Third Basic Survey on Natural Environment Protection (Natural Environment Information Map) of the Ministry of the Environment, including 10 types of volcanoes, limestone, rivers, lakes and marshes, and coasts. Based on how natural parks are categorized under the Japanese Natural Park Law, the ranking of natural resources ranks World Natural Heritage as the highest, followed by national parks, quasi-national parks, prefectural natural parks, and other forests at the bottom. The population economy factors include GDP, population spatial density, distance from three major metropolitan areas, and distance from the nearest major city (top 20 cities in Japan), where the population spatial density value is obtained by calculating the kernel density based on the population size and spatial distribution. The transportation resources factor is based on the high proportion of rail and bus travel in Japan and includes three factors: Distance to the nearest rail station, the total length of bus routes within 5 km, and rail transit ridership, where the rail transit ridership value is obtained by calculating the Kernel density value based on the spatial distribution of the average daily passenger flow (person/day) at each rail station. The tourism resources factor includes the spatial density factor of tourism resources, which is the sum of the tourism resources with an assessment grade of A or higher listed in the Tourism Resource Register created by the Tourism Resource Assessment Committee, which was established by the Japan Transportation Public Corporation, and the information on tourist spots in each prefecture (province) recorded in the Tourist Spots List.

3.2.2. Analysis of Influencing Factors

The Spearman correlation test was conducted between the spatial distribution of Japan’s forest therapy bases and the influencing factors (Table 1). The results show that the correlation coefficients of six factors in four dimensions (spatial density of natural landscape resources, GDP, spatial density of population, distance from three major metropolitan areas, rail transit ridership, and spatial density of tourism resources) pass the confidence level of 0.01, and the correlation coefficients are all greater than 0.3, which are reasonable influencing factors.

3.2.3. Strength Analysis of Influencing Factors

The six influencing factors in the four dimensions that were significantly correlated were categorized, and the determining power q values of each factor on the spatial distribution of forest therapy bases in Japan were calculated separately using a GeoDetector. The results (

Table 2. Geo-detection analysis of factors influencing the spatial distribution of forest therapy bases in Japan.

Indicator Dimensions	Detection Factors	Determining Power q
Natural Resources	Spatial density of natural landscape resources	0.28
Population economy	Gross domestic product	0.88
	Spatial density of population	0.31
	Distance from the three major metropolitan areas	0.63
Transportation resources	Rail transit ridership	0.18
Tourism Resources	Spatial density of tourism resources	0.38

) show that GDP (0.88) has the greatest explanatory power on the spatial distribution of Japan’s forest therapy bases, indicating that they have a strong socio-economic dependence. The distance from three major metropolitan areas (0.63) and spatial density of tourism resources (0.38) have greater explanatory power on the spatial distribution of Japan’s forest therapy bases, indicating that they show obvious centripetal characteristics toward the three metropolitan areas, i.e., the further away from the densely populated areas of the metropolitan areas, the fewer the number of forest therapy bases, and their dependence on surrounding tourism resources.

3.2.4. Interaction Detection of Influencing Factors

The interaction detector in GeoDetector was used to explore the influence degree of factors in pairs on the spatial distribution of forest therapy bases in Japan. The results (

Table 3. Interaction of factors influencing the spatial distribution of forest therapy bases in Japan.

	Spatial Density of Natural Landscape Resources X1	Gross Domestic Product X2	Spatial Density of Population X3	Distance from Three Major Metropolitan Areas X4	Rail Transit Ridership X5	Spatial Density of Tourism Resources X6
Spatial density of natural landscape resources X1	0.2767
Gross domestic product X2	0.9246	0.8752
Spatial density of population X3	0.7794	0.8971	0.3077
Distance from three major metropolitan areas X4	0.8172	0.9111	0.7804	0.6304
Rail transit ridership X5	0.7728	0.8970	0.3972	0.6773	0.1758
Spatial density of tourism resources X6	0.7495	0.9840	0.6613	0.8974	0.6405	0.3798

Note: X5 ∩ X6, X1 ∩ X3, X1 ∩ X5, X1 ∩ X6 interactions result in non-linear enhancement, and other factor interactions result in two-way enhancement.

) show that there are two types of interaction between the influencing factors, two-factor enhanced and non-linear enhanced, and there are no mutually independent factors. The spatial density of natural landscape resources interacts with the spatial density of population, rail transit ridership, and spatial density of tourism resources, and the spatial density of rail transit ridership interacts with the spatial density of tourism resources, showing a non-linear enhancement, while the other factors interact with a two-factor enhancement. The combined explanatory power of the non-linearly enhanced factors is significantly higher than the explanatory power of each factor individually, indicating that forest therapy bases tend to be located in areas with good natural resources, population economy, and infrastructure conditions. The spatial density interaction between GDP and tourism resources has an effect of 0.98, indicating that the interaction between these two factors has the highest consistency with the spatial distribution of forest therapy bases in Japan.

4. Discussion

4.1. Mechanism of Influencing Factors on the Spatial Distribution of Japan’s Forest Therapy Bases

Forest therapy originated in Germany and is popular in developed countries and regions such as Japan, Korea, the United States, and Europe. After hundreds of years of development, a variety of development models featuring forest medicine, forest therapy, and forest bathing have been formed, and a certain scale of the industry has been established, generating huge comprehensive benefits [37]. At present, international forest healthcare is developing in the direction of base, standardization, and systematization. Among the existing studies, there are few studies on the spatial structure of forest therapy bases from the perspective of geography. The use of quantitative data to explore its spatial evolution pattern and influencing mechanism is still rare, and the systematicity and depth of its research need to be strengthened. Based on the previous research and analysis of the spatial distribution characteristics and influencing factors of forest therapy bases in Japan, the authors explored the mechanism of influencing factors of their spatial distribution and provide references for future research on the site selection considerations of international forest therapy bases and even the overall industrial layout, with some important implications.

4.1.1. Natural Resources

The analysis of the natural resource grade (Figure 3) within 5 km of the Japanese forest therapy bases shows that the explanatory power of a single natural landscape resource for the spatial differentiation of the bases is weak, while the explanatory power increases significantly when it is superimposed with population economy and infrastructure factors, indicating that the bases are selected with a combination of natural and social conditions in mind and that social factors such as population economy factors and infrastructure factors are even more important than natural resources.

The website of Japanese forest therapy bases usually features the natural resources of the surrounding volcanoes, lakes, and highlands. Natural resources are the basis of the construction of forest therapy bases, and good natural resources can provide a healthy environmental atmosphere and meet the strict biochemical inspection requirements for forest therapy base certification. Therefore, the bases generally have good natural resources, making their spatial distribution insensitive to differences in natural resource grade.

Due to the lack of systematic research on multiple forest therapy bases on a national scale, the authors compared classic cases from Germany, Korea, and the United States. In Germany, where forest healthcare is a national policy, natural resources are also the basis of its development. The black forest in Baden-Württemberg in southwest Germany and the spa resort of Bad Krozingen, 15 km south of Freiburg, have both developed into modern spa resorts thanks to the discovery of hot springs. The northern part of the black forest, which is made up of a large number of pine and fir trees, has seen a rapid development of recreational forests with healing functions because of its unique tree resources. In Korea, forest therapy is called a recreation forest. Some of the famous forests are Sanin and Ususan, which are located in the valleys between the peaks of the mountains, where visitors can enjoy the beautiful natural scenery. The United States is one of the first countries in the world to develop health and wellness tourism. Tucson Canyon Ranch Resort, known as “America’s first health resort”, is a “green space in the desert” with excellent natural resources that are distinctly different from those in the surrounding area [37]. To sum up, the research results of this article are generally consistent with the situations of classic mature cases in various other countries. Therefore, to a large extent, it has been proven that good natural resources are a necessary condition for the development of the forest therapy industry.

4.1.2. Population Economy

Statistics from the Statistics Bureau of Japan (https://www.stat.go.jp/ (accessed on 20 March 2022)) [38] show that Japan’s GDP in 2019 was approximately ¥5.36 trillion, with a population of approximately 126 million. Each prefecture shows a highly significant positive correlation between GDP and total population, with a correlation coefficient of 0.997. Japan’s population, businesses, universities, and other educational institutions are concentrated in three major metropolitan areas (Tokyo, Nagoya, and Osaka), and 49% of the entire population of Japan is concentrated within 50 km of the city centers of the three major metropolitan areas.

According to the results of GeoDetector, the explanatory power of Gross Domestic Product and the spatial density of the population on the spatial distribution of forest therapy bases in Japan differed significantly. Overlay analysis of the forest therapy base locations with factors of GDP (Figure 4) and the spatial density of population in Japan (Figure 5) of all 36 prefectures where forest therapy bases exist shows that forest therapy bases are mostly located in the periphery of economically developed and densely populated areas. Densely populated areas are areas with high population density within urban towns and village areas obtained from census data. The densely populated areas of the three metropolitan areas were selected to construct 30 km, 60 km, and 90 km (i.e., 0.5 h, 1 h, and 1.5 h by motor vehicle) multi-ring buffer zones, and it was found that 25 forest therapy bases were located within 30 km of the densely populated areas of the three metropolitan areas (39.0%), 31 were located within 60 km of the densely populated areas (48.4%), and 36 were located (56.3%) within 90 km of population centers. It can be seen that the spatial distribution of forest therapy bases shows, to some extent, a tendency of spreading along the three major metropolitan areas to the periphery and shows a pattern of decreasing with increasing spatial distance. In this way, they can enjoy the resources of the population, economy, and traffic spillover from the big cities, and at the same time, take advantage of the conditions of the small local population, low construction intensity, and good natural environment, which integrates the edge effect and complementary effect of resources and is in line with the rule of outbound travel of metropolitan residents.

4.1.3. Transportation Resources

According to the results of the survey on transportation characteristics in Japan in Heisei 27 (2015), the proportion of rail transit and buses among the weekday and holiday travel modes of residents in the three major metropolitan areas has been increasing year by year, and the proportion of cars has been decreasing year by year. As the basis for effectively conveying logistics and passenger flow, the road network is also an important support tool for the development of forest areas. A 5 km radius (1 h walking or 0.5 h riding) of rail stations and bus routes throughout Japan was selected to map the buffer zone and was superimposed on the location of forest therapy bases for analysis. The results show that 61 of the 64 forest therapy bases across Japan have at least one mode of public transport coverage, including 1 h of walking by rail and 1 h of walking by bus, and 30 have two modes of public transport coverage, which indicates that the majority of forest therapy bases have easy access to transport, and therefore the spatial distribution of bases is not sensitive to specific differences in transport resources. The areas with relatively complete transportation resources make the forest therapy bases more accessible, which provides good conditions for external contacts and tourist attraction, and can, to a certain extent, reduce the expenditure of the forest therapy base.

The amount of rail transit ridership is a good indicator of the comprehensive use of local transport resources. Overlay analysis of the location of forest therapy bases and the spatial density of rail transit ridership shows that forest therapy bases are not concentrated in areas with dense rail transit passengers but are mostly distributed in areas with moderate rail transit ridership and their periphery (Figure 6). These areas are relatively convenient in terms of transportation, which indicates that lower development is more likely to protect the ecological environment and natural conditions.

4.1.4. Tourism Resources

Based on the nature of forest therapy activities, a radius of 5 km (1 h walking or 0.5 h cycling) was chosen, with 64 forest therapy bases as the center, and the number of regional tourism resources within this range was counted. It was found that the forest therapy bases showed the characteristics of agglomeration with other tourism resources to a certain extent (Figure 7). This is conducive for the forest therapy bases and other tourism resources through the spatial agglomeration effect to increase the flow of visitors, share infrastructure costs, and employ the advantages of an industrial cluster with shared resources, shared markets, and shared brands, in order to achieve the coordinated development of related industries in the region.

Indeed, forests are one of the main environments that offer recreational and nature tourism possibilities, and scholars researching the well-being effects of forests have made significant contributions to the tourism research agenda [20]. In recent years, the social functions of forests, particularly recreation, have also been part of the political agenda in many countries. In general, areas with a relatively good endowment of tourism resources, better tourism support facilities, and accessibility are more suitable for the development of the tourism industry. The results of the study show a strong symbiotic relationship between forest therapy bases and areas with better tourism resources, which is also consistent with our previous findings that areas with better tourism resources are generally better served by a more convenient transport network. The development of tourism in forest therapy bases relies on the capital and information flow brought by areas with better tourism resources to continuously expand its unique industrial advantages, and ultimately, mutually benefit from other tourism resources.

4.2. Limitations

Most studies found a positive relationship between urban forest cover and income, but there were some exceptions, and the magnitude of the relationship differs across studies [39]. When considering only more urbanized areas, the spatial correlations are positive, indicating an uneven distribution of urban forests that favors wealthier neighborhoods [40]. In an analysis of the spatial distribution and influencing factors of forest recreation tourism resources in Xinjiang, China, based on forest parks [33], it was found that in the Xinjiang region of China, the distribution of forest parks has a low overlap with population density, transportation accessibility, and economic development level, and high overlap with the water resources distribution. This is likely due to the large area of Xinjiang, the uneven distribution of forest resources, and its poor infrastructure, with accessibility being one of the main factors limiting its development. Our research is based on the more developed regions of Japan, and therefore our findings are likely to be more applicable to more developed countries or regions. When providing references for other countries and regions, the population size, natural resources, and national conditions should be taken into account to find a sustainable development pathway.

This article studies the influencing factors of the distribution of forest therapy bases. The distribution is not only affected by its surrounding environment, transportation resources, population economy, etc., but also related to the forest resources of the forest therapy base itself. Both are important factors affecting the spatial distribution of forest therapy bases in Japan. The former focuses more on the external environment, while the latter focuses more on internal factors. Regarding the efficacy of forest therapy bases, a large number of experimental verifications are needed. This involves the specific data of each forest therapy base, which is relatively complex and difficult to obtain, so we do not discuss it here. To some extent, it is also a limitation of this article, which will be discussed in subsequent research.

In addition, similar studies have focused on the spatial-temporal distribution characteristics and evolution mechanism of their research object [41,42], which suggests to the authors that the spatial distance of travel has been compressed over time as the economic level has increased and the transportation network has become progressively more convenient. It is possible that the explanatory power of our detection factors for the spatial distribution of forest therapy bases in Japan is gradually decreasing. As people’s demand for the ecological, healthcare, and cultural functions of forests is becoming more and more vigorous and urgent, forest therapy bases with their excellent natural environment have rightly become the first choice for many urbanites to escape the hustle and bustle of the city and pursue wild nature. Perhaps people can give up certain transportation and economic conveniences in order to seek the unique natural resources and recreation functions of forest therapy bases.

5. Conclusions and Implications

5.1. Conclusions

Based on multiple data, this study constructed a geographic information database of elements related to the distribution of Japan’s forest therapy bases. Based on the analysis of the spatial distribution characteristics of Japan’s forest therapy bases using Kernel Density Analysis and the GeoDetector method, the influencing factors of their spatial distribution were further explored, with the following conclusions:

(1) The spatial distribution of Japan’s forest therapy bases is uneven, showing the characteristics of aggregation in the periphery of economically developed and densely populated areas, with a high-density core area extending from the Tokyo metropolitan area along the northwest to the Hokuriku-Koshinetsu area.

(2) The factors influencing the distribution of forest therapy bases in Japan are the spatial density of natural landscape resources, GDP, the spatial density of population, distance from three major metropolitan areas, rail transit ridership, and the spatial density of tourism resources. These factors interact to influence the spatial pattern of forest therapy bases in Japan, while the spatial density of forest therapy bases is not significantly correlated with the distance from major cities, distance from rail transit stations, bus route density, or natural resource grade.

(3) Gross Domestic Product has the greatest explanatory power for the spatial distribution of forest therapy bases in Japan, followed by the distance from three major metropolitan areas and the spatial density of tourism resources, while the spatial density of the population, spatial density of natural landscape resources, and rail transit ridership have relatively weaker effects on Japan’s forest therapy bases. When exploring the interaction of each influencing factor, the spatial density of natural landscape resources interacts with the spatial density of population, rail transit ridership, and spatial density of tourism resources respectively, showing non-linear enhancement. The interaction between rail transit ridership and spatial density of tourism resources shows a non-linear enhancement, and the interaction of other factors shows a two-way enhancement. The influence of interaction between GDP and spatial density of tourism resources reaches 0.98.

In addition, due to the limited sample size and slow growth of forest therapy bases in Japan, the influence of site selection contingency cannot be completely excluded when quantitatively assessing the factors influencing their spatial distribution using GeoDetector. The research scale of this study is macroscopic and may be subject to some errors due to the statistical caliber and granularity of the data.

5.2. Implications

Nowadays, “sub-health” is becoming more and more common in urban areas, and people are more and more concerned about their quality of life and physical and mental health. Forest healthcare is a way to experience the natural environment of the forest through the five senses and promote physical and mental recovery of the body and the mind and maintain a healthy state. Its construction and development will have broad market prospects and good socio-economic benefits. Despite the differences in terms of population size, natural resources, and national conditions in different countries or regions, Japan’s forest therapy bases have certain inspirations and references in terms of site-selection considerations.

First of all, a sound transport infrastructure and good natural resource conditions are common features of Japan’s forest therapy bases, which should be considered as the basis for forest therapy bases in other countries or regions. Secondly, natural resources and population economy factors should be taken into account, focusing on densely populated urban clusters with a good economic base and a high proportion of tertiary industries, which can provide a large number of visitors for forest therapy bases. Thirdly, it is necessary to consider the spatial agglomeration effect of forest therapy bases and other tourism resources, make full use of existing resources, and focus on the combination with existing resources such as national parks, forest parks, and recreation forests, giving full play to the advantages of industrial agglomeration and enriching the tour activities. Finally, the development of forest therapy bases in each region also requires scientific coordination and orderly promotion, strengthening certification mechanisms, improving support services such as forest therapist training, forest retreat curriculum development, accommodation, and food management, and ensuring systematic and scientific development of forest retreat bases. In addition, it should also combine its own social, economic, and cultural characteristics, coordinate regional development and industrial interaction, formulate reasonable industrial development plans and find a sustainable development path that suits itself.