The Transition of Forest Cover and Its Cultural Values in Shrine/Temple Forests in the Mountainous and Foothill Areas of Kyoto City: A Study Based on Topographic Maps and Aerial Photos

Abstract

There is growing interest in the diverse roles of forests in addressing climate change and biodiversity goals. Recent studies have indicated a disregard for the cultural values of forests that have been formed in close association with human activities. This may potentially lead to the loss of cultural characteristics, traditional forest knowledge, and biodiversity. This study explores historical forest dynamics and their unique cultural values from the end of the 19th century in 15 shrine/temple forests located in the mountainous and foothill areas of Kyoto city. Using geographical information systems (GIS) based on topographic maps and aerial photographs, this study investigates the forest composition in the 1890s, 1980s, and from 2010. The results indicate that approximately half of the targeted shrine/temple forests were composed of low Pinus densiflora forests and coppice forests in the 1890s. Between the 1890s and 1910s, coniferous forests were planted in these areas with the intention of land conservation and timber production. This distinctive forest cover became a typical characteristic for shrine/temple forests until the 1980s. However, from the 1980s, a decrease in the cultural value of shrine/temple forests was observed due to the lack of human activities in these forests. As a result, the distinction between shrine/temple forests and the surrounding forests has become blurred. This could potentially cause the homogenization of cultural characteristics. This study aims to inform readers of the cultural value associated with the historical landscape and biodiversity found in shrine/temple forests.

1. Introduction

1.1. Research Background

There is a growing interest in the diverse roles of forests in addressing climate change and biodiversity goals. The United Nations declared 2021 to 2030 to be the “Decade on Ecosystem Restoration”, with focused efforts on promoting forest restoration [1]. Apart from the increasing expansion of forested areas, there is an urgent need to enhance the natural characteristics of forests, based on the assessment of forest ecosystem services [2]. In this context, old plantation forests that are not utilized would, over time, deliver environmental benefits more effectively if they were restored to natural forest or semi-natural forests [3]. On the other hand, there is also the practice of maintaining the characteristics of forests that have been shaped by ongoing human activities through historical evidence. An example of such a practice can be found in Italy’s coppice woodlands [4]. This Italian approach argues that forest management practices must consider the historical context to avoid endangering the cultural value of these forests [4]. According to the theories related to ecosystem services, cultural values have been associated with the “aesthetic, artistic, educational, spiritual and/or scientific values of ecosystems” [5]. However, these assessments are from the perspective of the recipients of these services, and they may not encompass the cultural values influenced by human actions such as forest management policies and operational activities [2]. Traditional forest management practices applied for centuries have significantly influenced forest characteristics (in terms of extension, density, species composition, and vertical and horizontal structure). Even if traditional forest management practices are seldom practiced today, the influence of past management practices continues to shape the current characteristics of forests [4]. Therefore, to neglect the cultural aspects of forests carries the risk of the future homogenization of forest types, and the loss of cultural significance, traditional forest knowledge, and biodiversity on a landscape scale [4].

In Japan, the Satoyama Initiative [6] promotes conservation efforts to establish land use systems that respect traditional knowledge while fostering a harmonious relationship between nature and human activities. These efforts have also involved the assessment and management of ecosystem services, including the diversity of secondary natural environments [7].

Moreover, “shrine/temple forests” in Japan are forests that are owned or utilized by shrines or temples and formed in close association with human activities [8]. Originally, these forests were held for the purpose of maintaining the religious ambience and management of shrines and temples. They served as places to foster religious solemnity, conduct rituals and practices, and functioned as disaster prevention areas [9,10]. In addition, shrine/temple forests have historically served as a source of income for shrines and temples through the production of timber and non-timber forest products (NTFPs) [9,10]. Therefore, shrine/temple forests can be described as green environments that exhibit various functions that align with the needs and requirements of shrines or temples, including landscaping, land conservation, timber production, and NTFP production [8].

Recent studies on shrine/temple forests have presented varying perspectives regarding their historical characterization. Miyawaki (1970) [11] considers these forests to be primeval and left undisturbed by human intervention. In contrast, other researchers have indicated that until the Meiji period (mid-19th century), coniferous forests, managed for the extraction of forest resources, constituted the core of shrine/temple forests [12,13,14]. Moreover, the current deciduous broad-leaved forests are suggested to comprise secondary forests that emerged after the Meiji period (mid-19th century) [12,13,14]. Such research is instrumental in guiding the discourse and management strategies that consider the multifaceted historical significance of these forests.

Due to the differences in perception regarding the historical significance of shrine/temple forests, little research has focused on shrine/temple forests and their cultural values, influenced by human activities [15]. Therefore, this study utilizes the discourse propounding that shrine/temple forests in Japan are forests formed under the influence of active human activities.

1.2. The Shrine/Temple Forests

The shrine/temple forests observed today have a historical background, and have been influenced by events that occurred after the Meiji Restoration (1868). Primarily, they were influenced by the “Land Tax Reform” in 1875, whereby most of the shrine/temple forests became nationalized [9,10]. However, the “National Forests Act” in 1898 and the “Disposal of Former Shrine and Temple Reserve Forest” in 1947 resulted in the return of certain areas to the original shrines and temples [15]. As a result, through these historical processes, the boundaries and extent of the current shrine/temple forests were determined.

Between the 1890s and 1920s, the management system for shrine/temple forests began to take shape within the Japanese government. Scholars such as Honda Seiroku, Hongo Takanori, and Uehara Keiji developed the concept of “Shaji Fuchirin Ron (Management Theory of Shrine/Temple Forest)”, from the perspectives of forestry and landscape architecture [16]. This concept advocated for the ideal aesthetic qualities of shrine/temple forests, emphasizing the cultivation of religious ambience through the planting of evergreen coniferous trees, such as Cryptomeria japonica and Chamaecyparis obtusa. These trees are valued as construction materials for shrine/temple buildings and as general timber. However, they recognized that maintaining these coniferous trees in urban areas with a high population density and ongoing industrialization would become unfeasible. As an alternative, a proposal for maintaining shrine/temple forests could be the natural regeneration of evergreen broad-leaved trees such as Cyclobalanopsis, Castanopsis, and Cinnamomum camphora. Promoting the preservation of evergreen broad-leaved forests marked a significant milestone in the concept of shrine/temple forest landscape [17].

1.3. The Historical Landscape and Forest Dynamics in Kyoto City

Kyoto was the imperial capital of Japan from its foundation until the middle of the 19th century. As the center of Japanese culture for more than 1000 years, Kyoto illustrates the development of Japanese wooden architecture, particularly shrine/temple architecture [18]. In Kyoto city, shrine/temple forests have been recognized as essential elements that contribute to the historical landscape of shrine/temple historical sites and architecture [18]. Since the Meiji period (1868–1912), efforts have been made to preserve forest landscapes as tourist resources and local scenery, focusing on nationalized shrine/temple forests in the suburbs of Kyoto city such as Higashiyama and Arashiyama [19,20,21]. Ogura (1992) [22] and Nakajima (1996) [23] conducted detailed examinations of the vegetation surrounding Kyoto city during the mid-Meiji period (1890s) by using a topographical map with a scale of 1/20,000, surveyed by the Land Survey Department of Japan’s Army in 1889 (Annotation A1 in Appendix A). These studies found that around the urban areas of Kyoto, there were widespread Pinus densiflora forests, with a height of less than 2.7 meters (m). On the other hand, the background of shrines and temples consisted of Pinus densiflora forests greater than or equivalent to 5.4 m, while the more remote areas in the mountainous northern part of Kyoto consisted mainly of Cryptomeria japonica forests. These studies highlighted that these forests constituted distinctive landscapes associated with shrine/temple forests in the 1890s.

During the Meiji period (1868–1912), forest landscapes were generally preserved; however, the excessive logging during World War II destroyed these forest landscapes [24]. However, also during World War II, shrine/temple forests were protected from this excessive logging due to their religious significance [24]. This exemption from excessive logging provided the forest stands with continuity for over 75 years, which was revealed in the 1980s shrine/temple forest survey [25]. Until the 1980s, secondary forests primarily composed of Pinus densiflora continued to play a central role in shaping the scenic beauty of the ancient capital, Kyoto city [26]. Pinus densiflora forests surrounding the city of Kyoto have been utilized throughout history as a source of construction materials, fuelwood, and food (such as mushrooms), and have promoted maintenance practices like underbrush clearing and leaf raking [27]. However, with the decline in its utilization due to the fossil fuel revolution, coupled with the outbreak of pine wilt disease, Pinus densiflora has experienced a rapid decline in its numbers since the 1980s [28]. Pine wilt disease (PWD) is caused by Bursaphelenchus xylophilus, the pine wood nematode (PWN) [29,30]. It has caused serious forest issues primarily in East Asia [31] since the 1970s, with serious outbreaks in China, Korea and Taiwan [32,33,34]. Across Japan, except for certain areas of Hokkaido, it became a nationwide outbreak [31]. The PWD rapidly became prevalent in Kyoto [28] during the 1980s, resulting in a change in the forest landscapes surrounding Kyoto city. By the late 1990s, forests consisting of Pinus densiflora were being naturally replaced by maturing understory trees, and the natural regeneration of Pinus densiflora forests became unfeasible in areas that ceased the thinning and mowing of forest undergrowth [28,35]. Eventually, Pinus densiflora forests were replaced by broad-leaved forests mainly with Quercus variabilis and Quercus serrata [35,36]. Currently, the distribution of Castanopsis cuspidata is expanding [36]. Castanopsis cuspidata is an evergreen broad-leaved tree and is known as a climax species in most areas of Japan, including Kyoto city [37]. In the temperate regions of Japan, there have been reports suggesting that secondary forests have a higher biodiversity compared to climax forests [36,38,39]. These changes have also affected shrine/temple forests, and consequently, resulted in a growing need for the conservation and restoration of historical forest landscapes [24,40].

Conservation efforts have been undertaken to preserve the historical landscape centered around Pinus densiflora forests, encompassing the entire forest area surrounding Kyoto city [24,35]. However, approximately 42% of shrine/temple forests in Kyoto city are occupied by planted coniferous forests of Cryptomeria japonica or Chamaecyparis obtusa [41]. There are no specific policies that encompass the preservation of planted coniferous forests of Cryptomeria japonica or Chamaecyparis obtusa [24,40]. Although shrine/temple forests have been managed for various purposes, including the preservation of religious solemnity and as a source of timber resources for shrines and temples [9,10], human activities in shrine/temple forests are different when compared to non-shrine/temple forests in Kyoto city. As a result, there is a need to develop explicit policies that target the conservation and restoration of shrine/temple forests whilst considering the cultural value of the forest landscape.

1.4. Research Questions

Previous studies on shrine/temple forests in Kyoto city have primarily focused on (i) the landscape conservation policies implemented during the Meiji period (1868–1912) [42,43], (ii) the ecological value of forests [44,45], (iii) on examining the vegetation and resource utilization of a specific shrine during the Edo period (1603–1868) [13,14]. For example, studies conducted by Ogura (1992) [22] and Ogura (2012) [46] regarding the vegetation of shrine/temple forests in Kyoto city focused only on the Meiji period (1868–1912). However, investigations into the changes in the forest cover of shrine/temple forests and the reasoning behind planting coniferous forests of Cryptomeria japonica or Chamaecyparis obtusa since end of the 19th century are lacking. Furthermore, the cultural values implicated in each of these processes have not been clarified.

Therefore, in this study, the following two research questions were established to investigate the conservation of cultural values in shrine/temple forests in Kyoto city.

Q1. How has the forest cover of shrine/temple forests in Kyoto city changed since the mid-Meiji period (1890s)?

Q2. What is the history of the cultural values held by shrine/temple forests in Kyoto city since the mid-Meiji period (1890s)?

2. Materials and Methods

2.1. Study Sites

This research focuses on 15 shrine/temple forests located within Kyoto city (Figure 1). These 15 sites were selected based on the “Survey for the Identification of Outstanding Shrine/Temple Forests in Kyoto Prefecture” [25] (Annotation A2 in Appendix A), conducted by the Kyoto Prefectural Government in 1988–1989 (hereinafter referred to as the “Kyoto Prefectural Shrine/Temple Forest Survey”). In this survey, the selection of these forests was based on five criteria, which are listed in

Table 1. Selection criteria for the “Kyoto Prefectural Shrine/Temple Forest Survey” [25].

	Selection Criteria
1	Primeval or near-pristine natural forest
2	Forest communities that represent the local landscape and shows typical community characteristics
3	Forests where no logging has occurred over the long term
4	Natural forests that are scientifically valuable because of their unique distribution under natural conditions or because they show typical transitional forms.
5	Forested areas of 1 ha or more and with forest age of 75 years or older

. The scope of this study encompasses the shrine/temple forests specified in the survey report, with a focus on the historical changes in forest cover from the past to the present. It is important to note that these boundaries do not necessarily represent the current ownership boundaries of the shrines and temples. Additionally, certain areas might not be part of the forest that was owned by shrines and temples during the Meiji era (1868–1912) because of the nationalization of temple forests under the “Land Tax Reform” in 1875. In this study, we conducted an analysis of the forest cover changes in the shrine/temple forests, taking into consideration the surrounding forest environment. To ensure consistency in the sample size of the dataset, buffer zones of the same area were established around each shrine/temple forest.

Among the 15 shrine/temple forests, No. 1, No. 2, and No. 10 are located at elevations from 500 to 900 m, while the other shrine/temple forests are all located at elevations below 500 m (Figure 1).

2.2. Research Method

In this study, we examined changes in the forest cover of Kyoto using data from three different periods: (i) the 1890s, which has been commonly studied for historical forest landscape analysis in Kyoto city; (ii) the 1980s, a period known for changes in forest composition, including a decline in Pinus densiflora forests; and (iii) after 2010, as the present day. The 1980s field-survey-based vegetation data were used as an intermediate indicator from the 1890s to the present in order to minimize errors due to the interpretation of aerial photographs [47]. As indicated earlier, the forest landscape in Kyoto city remained largely unchanged from the 1890s until the 1980s [35]. Additionally, due to the lack of available historical data for other periods [46], these three periods were selected for the analysis of this study.

Aerial photographs have been widely used for long-term forest cover comparisons due to the relatively small difference in data acquisition accuracy between them and satellite images [47,48,49]. However, due to the historical context of this study, topographical maps were also utilized to understand the vegetation during the 1890s [22,46,50]. Therefore, this study used a combination of topographic maps and aerial photographs (

Table 2. Data sources used for forest cover type mapping.

Period	Survey Site Numbers	Reference	Year	Publisher
1890s	1, 2	the topographical map surveyed in 1893 with a scale of 1/20,000	1893	Land survey department of Japan’s Army
	3–15	the topographical map surveyed in 1889 with a scale of 1/20,000	1889	Land survey department of Japan’s Army
1980s	3–15	the 1985 National Land Image (color aerial photograph) with a scale of 1/20,000	1985	the Geospatial Information Authority of Japan (GSI)
After 2010	1, 2	the 2010 National Land Image (color aerial photograph) with a scale of 1/10,000	2010	the Geospatial Information Authority of Japan (GSI)
	3–15	the 2020 National Land Image (color aerial photograph) with a scale of 1/10,000	2020	the Geospatial Information Authority of Japan (GSI)

, Figure 2) as primary data sources and referred to a vegetation map obtained from GIS data. The GIS vegetation maps were from the “3rd Natural Environment Survey on Vegetation” (1/50,000 scale, conducted in 1983) and the “6th and 7th Natural Environment Survey on Vegetation” (1/25,000 scale, conducted in 2004), provided by the Ministry of the Environment to aid in the interpretation of vegetation patterns. As historical resources from the 1890s, we used a topographical map with a scale of 1/20,000 surveyed in 1889 and a topographical map with a scale of 1/20,000 surveyed in 1893 (Annotation A3 in Appendix A) to understand the vegetation during the Meiji period (1890s) [22,46,50]. For the 1980s and after 2010, we utilized three major sources of aerial photographs: (i) the 1985 National Land Image (color aerial photograph taken by the Geospatial Information Authority of Japan (GSI)) with a scale of 1/20,000; (ii) the 2010 National Land Image (color aerial photograph taken by the GSI) with a scale of 1/10,000; and (iii) the 2020 National Land Image (color aerial photograph taken by the GSI) with a scale of 1/10,000 (Annotation A4 in Appendix A). In addition, the study utilized supporting data from the 2008 Digital National Land Information, published by the GSI (Orthoimage) and with a ground pixel size of 40 cm, as a reference for ArcGIS (ver. 3.1.0) georeferencing. This allowed the study to perform geometric correction for these aerial photographs before conducting the identification of the forest cover type. Moreover, geographical location information was added to these topographic maps by employing the Digital National Land Basic Map (Basic Geospatial information) for reference and for the utilization of georeferencing. The positions of shrine/temple buildings primarily served as the starting points for this process. Subsequently, three vegetation maps were created using the planar rectangular coordinate system JGD2000 VI as the basis.

The topographical map with a scale of 1/20,000 surveyed in 1889 was created for military purposes, with the Japanese army aiming to perform assessments to locate construction timber and fuelwood resources [22]. Therefore, the classification of forests on the map was primarily performed from the perspective of resource utilization. This classification was performed based on tree species with a high timber value, including Cryptomeria japonica, Chamaecyparis obtusa, and Pinus densiflora, as well as tree species with a high fuelwood value, including Quercus serrata or Quercus acuta, and coppice forests. For Cryptomeria japonica, Chamaecyparis obtusa, and Pinus densiflora, subcategories of mature or low-height trees were assigned. In the study area of this research, only Pinus densiflora had a distinction between mature and low-height trees, which resulted in the use of the classifications shown in Figure 3. When considering the influence of historical human activities, it is considered inadequate to evaluate the cultural value solely based on classifications derived from phytosociology [2]. To prevent this bias in the study, we used the resource-use-based forest classifications obtained from topographic maps created in the Meiji period as a basis for reading aerial photographs in the 1980s and after 2010. Figure 3 illustrates the correlation between the corresponding legend of these topographical maps and identifies the vegetation types from these aerial photographs. The symbols used in the topographical maps and their corresponding vegetation are in reference to Ogura (1992) [22] and Ogura (2012) [46]; these were verified in detail based on old photographs and geographical records.

Considering the potential natural vegetation in Kyoto city, it is known that all Cryptomeria japonica forests shown on the topographic map were established through afforestation [51]. Therefore, these Cryptomeria japonica forests on the topographic map corresponded to the planted coniferous forests observed in the aerial photographs. However, distinguishing between mature and low Pinus densiflora forests on the topographic map was not feasible in the aerial photographs. For this reason, both types were combined and classified as Pinus densiflora forests. It should be noted that the coppice forests found on the topographic map could not be identified as only coppice forests in the aerial photographs from the same period. Therefore, these coppice forests were aligned with natural conifer and broad-leaved mixed forests.

Apart from planted conifer forests of Cryptomeria japonica or Chamaecyparis obtusa, and the secondary forests of Pinus densiflora, the existing vegetation in Kyoto City includes broad-leaved trees such as Castanopsis spp. and Quercus spp. at elevations below 500 m [52]. On the other hand, at elevations between 500 m and 1000 m, the existing vegetation includes broad-leaved trees such as Quercus spp. and Acer spp., as well as natural coniferous trees such as Abies firma and Chamaecyparis obtuse [52].

Regarding cultural values, forest environments have been discussed within the context of cultural landscape heritage conservation [53,54] or evaluation of sacred natural sites, such as the Delos Initiative [55,56]. In 2021, the International Union for Conservation of Nature and Natural Resources (IUCN) developed guidance on the conservation and management of sacred natural sites from the perspective of both cultural values and ecological values [56]. In these contexts, cultural value refers to the significance that individuals or groups attribute to the forest, encompassing historical, aesthetic, economic, social, scientific, and various other types of value [54,56]. Importantly, the elements and significance that comprise cultural value can change over time, depending on the regional and temporal context [56,57]. This stipulates that cultural values encompass diverse and complex significance accumulated over time. Therefore, the assessment of cultural value requires reassessing the values inherited from previous generations in a contemporary context using a historical approach [57].

Through the following two procedures, the historical forest dynamics of shrine/temple forests from the 1890s to the present (after 2010) were elucidated, and an assessment of the cultural value of shrine/temple forests was conducted. Firstly, using the tabulated area function of ArcGIS, a cross-tabulation of forest cover changes was conducted to clarify the forest dynamics. Other research similar to this study also used topographic maps created in the 1890s and aerial photographs to validate the forest dynamics, and considered changes of 1 ha or more to be significant changes [58]. Therefore, in this study, changes of 1 ha or more were regarded as significant. Secondly, as part of the assessment of cultural value, hierarchical cluster analysis using the Ward Method was conducted to understand the trends in forest cover types for each shrine/temple forest in the 1890s. The forest cover types and their functional categorization using the data collected were classified into three categories based on the forest classifications obtained from the topographic maps and previous studies on the landscapes of shrine/temple forests during the 1890s. A designated definition of cultural value for each forest cover type allocated three functional categories according to the IUCN’s guideline [56] and the original meaning of shrine/temple forests’ existence [9,10] mentioned earlier in the study. The history of the cultural values of shrine/temple forests was explored by tracing the changes in the forest cover type of the three functional categories. The study examined which forest cover type matched the cultural value of each functional category, and how such cultural values transformed due to changes in the forest cover type.

(i)

Landscape forest

We defined the forests dominated by “mature Pinus densiflora forests” or “coniferous forests such as Cryptomeria japonica” in the 1890s as landscape forests. The “mature Pinus densiflora forest” and “coniferous forests such as Cryptomeria japonica” were identified as landscape elements owing to their support of the solemnity of shrines and temples in the 1890s [22,23]. Furthermore, these tree species have timber production functions and were utilized as construction materials for shrine/temple buildings and as general timber. This finding suggests that, along with their landscaping function, these forests also encompassed timber production functions. Hence, the term landscape forest in this study can be regarded as also having timber production functions. The cultural value of landscape forests can be defined as their ability to support the solemnity of shrines and temples and simultaneously provide construction materials for the management of shrines and temples. This includes the knowledge of landscape forestry that enables timber production while maintaining the religious solemnity of shrine/temple forests.

(ii)

Fuelwood forest

Forests dominated by “low Pinus densiflora forests”, coppice forests and broad-leaved forests were classified as fuelwood forests. “Coppice forests” are known as fuelwood forests based on the investigation of the “Kyoto Prefecture Geography” and photographs [22]. Moreover, “low Pinus densiflora forests” are considered to have grown on land that was extensively logged during the 1870s, based on an investigation of their age [23,51]. These types of forests are known to have been formed due to the high demand for forest resources, primarily firewood and charcoal [51,59]. Additionally, “broad-leaved forests”, correspond to the legend “Quercus serrata or Quercus acuta forest” on the topographical map (1/20,000, surveyed in 1889). It is known for its use as firewood and charcoal. The cultural value of fuelwood forests can be defined as their ability to provide a sustainable supply of firewood and charcoal materials for supporting the livelihoods of people associated with shrines and temples. This includes the knowledge and practice of routine forest maintenance by those community members.

(iii)

Mixed function forest (landscape and fuelwood)

Forests that possess characteristics of both landscape forests and fuelwood forests were classified as mixed-function forests (landscape and fuelwood). These forests were composed of “mature Pinus densiflora forests” and “coniferous forests such as Cryptomeria japonica”, which represent landscape forests. In addition, they were composed of “low Pinus densiflora forests”, “coppice forests” and “broad-leaved forests”, which represent fuelwood forests. The cultural value of mixed-function forests can be defined as their ability to support the solemnity of shrines and temples, as well as accommodate the production of timber, fuelwood, and charcoal materials. Therefore, these forests support the operation of shrines and temples while simultaneously supporting the livelihood of people associated with shrines and temples. This includes knowledge regarding the selection of suitable locations for each function, and the skills and techniques required to maintain all functions at the same time.

3. Results

3.1. The Historical Forest Dynamics of Shrine/Temple Forests from the 1890s to the Present (after 2010)

3.1.1. The Changes in Each Forest Cover Type from the 1890s to 1980s

Regarding overall change, there was a shift to conifer afforestation, broadleaf tree dominance, and mixed-forest formation within the forest cover types (Figure 4 and Figure 5). Pinus densiflora forests, by the 1980s, were reduced to only 30–40% of their presence in the 1890s. Approximately 30% of the remaining forest cover types were transformed into coniferous planted forests, and around 40% transitioned into broad-leaved or mixed forests. Comparing the trends in change with the buffer areas, it was found that the afforestation of coniferous trees progressed in the early stages, particularly in the areas that were formerly used for fuelwood production (LP, C, BL), as well as grasslands or wastelands (GL). In terms of conifer afforestation, only about 30% of the forests that were Cryptomeria japonica forests in the 1890s remained as mature planted conifer forests, with approximately 40% transitioning into mixed forests. On the other hand, about 30% of the forests that were Cryptomeria japonica forests in the 1890s were converted to young coniferous planted forests or newly coniferous planted forests, with logging and replanting carried out until the 1980s. In former coppice forests, approximately 50% had been transitioned into coniferous planted forests, with half of them being young coniferous planted forests or newly coniferous planted forests. In this context, silvicultural practices involving logging and replanting were carried out until around the 1980s.

3.1.2. The Changes in Each Forest Cover Type from 1980s to the Present (after 2010)

There has been an observed growth in coniferous planted forests, the conversion of Pinus densiflora forests into broad-leaved forests and the conversion of grasslands into coniferous planted forests or broad-leaved forests (Figure 4 and Figure 5). However, significant changes in forest cover types were rarely observed.

3.2. The Forest Cover Types and Their Functional Categories

3.2.1. Forest Cover in the 1890s

Regarding the proportional area for each forest cover within the target shrine/temple forests, we conducted a hierarchical cluster analysis using the Ward Method, resulting in the following five cluster classifications (Figure 6):

• Cluster-1 (1890): Forest dominated by coppice forests, occupying more than 50% of the forested area.
• Cluster-2 (1890s): Forest dominated by coniferous forests such as Cryptomeria japonica, occupying approximately 80% of the forested area.
• Cluster-3 (1890s): Forest dominated by broad-leaved forests, occupying approximately 40% of the forested area.
• Cluster-4 (1890s): Forest dominated by low Pinus densiflora forests with a height of less than 2.7 m, occupying approximately 80% of the forested area.
• Cluster-5 (1890s): Forest dominated by mature Pinus densiflora forests with a height of 5.4 m or more, occupying more than 40% of the forested area.

Among these, Cluster-1 (1890s) and Cluster-4 (1890s) were characterized by an area of approximately 80% or more being occupied by low vegetation, such as low Pinus densiflora forests and coppice forests (No. 14 represents the combined area of low Pinus densiflora forests and coppice forests). A total of seven shrine/temple forests were classified into these two clusters. On the other hand, Cluster-2 (1890s) and Cluster-5 (1890s) were characterized by mature Pinus densiflora forests and coniferous forests such as Cryptomeria japonica, respectively. Each cluster contains three shrine/temple forests, with a total of six shrine/temple forests classified into these two clusters. Cluster-3 (1890s) was characterized by broad-leaved forests. Two temple forests were classified into this cluster. Furthermore, among the shrine/temple forests classified into Cluster-1 (1890s) and Cluster-4 (1890s), the relatively large shrine/temple forests (20 ha or more) showed a significant proportion of mature Pinus densiflora forests or coniferous forests such as Cryptomeria japonica, as seen in No. 10 and No. 13. Similarly, even in Cluster-3 (1890s), the large temple forests, such as No. 3 and No. 12, exhibited a similar trend.

In the 1890s, shrine/temple forests were classified into different structures, each with its main forest cover types: Coniferous forests such as Cryptomeria japonica (Cluster-2 (1890s)), mature Pinus densiflora forests (Cluster-5 (1890s)), low-vegetation forests, including low Pinus densiflora forests and coppice forests (Cluster-1 (1890s), Cluster-4 (1890s)), and broad-leaved forests (Cluster-3 (1890s)). In the shrine/temple forests dominated by mature Pinus densiflora and coniferous forests such as Cryptomeria japonica (Cluster-2 (1890s), Cluster-5 (1890s)), as well as the larger shrine and temple forests (20 ha or more; Cluster-3 (1890s), No. 10, No. 13), there were noticeable differences in forest structure compared to the surrounding areas (buffer), mainly based on the proportion of mature Pinus densiflora and coniferous forests such as Cryptomeria japonica. On the other hand, the shrine/temple forests classified into Cluster-1 (1890s) and Cluster-4 (1890s), primarily dominated by low vegetation such as low Pinus densiflora forests and coppices forests, did not exhibit clear differences in forest cover when compared to the surrounding areas (buffer).

Based on the above, these clusters were classified into the following functional categories:

(1)

Landscape forests; Cluster-2 (1890s), Cluster-5 (1890s).

(2)

Fuelwood forests; Cluster-1 (1890s), Cluster-4 (1890s).

(3)

Mixed-function forests (landscape and fuelwood); Cluster-3 (1890s).

3.2.2. Forest Cover in the 1980s

Similar to the previous section, a hierarchical cluster analysis was conducted, and the results classified the data into the following three clusters (Figure 6):

• Cluster-1 (1980s): Forest dominated by Pinus densiflora forests or broad-leaved forests, occupying more than 70% of the forested area.
• Cluster-2 (1980s): Forest dominated by mature to old-growth planted coniferous forests, occupying more than 70% of the forested area.
• Cluster-3 (1980s): Forest dominated by young and newly planted coniferous forests, occupying approximately 30% or more but less than 70% of the forested area.

From the composition of the forest cover type, Cluster-1 (1980s) is considered to retain characteristics from both Cluster-3 (1890s), which is dominated by broad-leaved forests, and Cluster-5 (1890s), which is dominated by mature Pinus densiflora forests. Similarly, Cluster-3 (1980s) is thought to retain characteristics from Cluster-2 (1890s), which is dominated by coniferous forests such as Cryptomeria japonica. However, shrine/temple forests that shared similar forest cover and functions in the 1890s exhibited distinct characteristics in their forest cover by the 1980s. In the forest cover of the 1980s, the clusters were primarily formed based on the proportion of planted coniferous forests. The landscape forests, which were dominated by mature Pinus densiflora forests, and the fuelwood forests had disappeared before the 1980s. Among the landscape forests, No. 15 deviated from the characteristics of Cluster-1 (1980s), showing an increase in the proportion of planted coniferous forests and mixed forests. Furthermore, apart from No. 13, the areas classified into fuelwood forests were dominated by low vegetation, such as low Pinus densiflora forests, and areas occupied by coppice forests in the 1890s had more than 30% to 70% of the forested area occupied by planted coniferous forests in the 1980s.

In the 1890s, there were no noticeable differences in forest cover between the shrine/temple forests and the surrounding areas (buffer). However, in the 1980s, many of these shrine/temple forests showed differences with the surrounding forest cover. This is mainly attributed to the proportion of mature to old-growth planted coniferous forests (No. 5, No. 6, No. 8, No. 9, No. 10 and No. 13). The shrine/temple forests classified into Cluster-3 (1980s) exhibited the presence of young planted coniferous forests, so replanting was conducted within the shrine/temple forests. However, the proportion of these young planted coniferous forests did not necessarily follow the forest cover observed in the 1890s, or the shrine/temple forest area trends. Instead, it was likely that the proportions were determined based on the individual policies of each shrine and temple.

3.2.3. Forest Cover after 2010

In the same manner as the previous section, a hierarchical cluster analysis was conducted, and the results classified the data into the following three clusters (Figure 6):

• Cluster-1 (After 2010): Forest dominated by Pinus densiflora forests or broad-leaved forests, occupying more than 70% of the forested area.
• Cluster-2 (After 2010): Forest dominated by mature to old-growth planted coniferous forests, occupying more than 50% of the forested area.
• Cluster-3 (After 2010): Forest dominated by mature to old-growth planted coniferous forests or young planted coniferous forests, occupying approximately 30% or more but less than 50% of the forested area.

After 2010, the forest cover clusters were classified in a similar way to those in the 1980s. Cluster-1 (After 2010) consisted of the same temples as Cluster-1 (1980s), and the change in forest cover indicated a transformation from Pinus densiflora forests to broad-leaved forests. Furthermore, Cluster-2 (After 2010) included not only the shrine/temple forests classified into Cluster-2 (1980s), but also some of the shrine/temple forests from Cluster-3 (1980s). All of these, except for No. 2, were classified into fuelwood forests in the 1890s, which were dominated by low vegetation such as low Pinus densiflora forests and coppice forests. Cluster-3 (After 2010) had the same composition as Cluster-3 (1980s) (No. 1, No. 3, No. 4, No. 8, No. 9 and No. 15), indicating a change in forest cover characterized by the aging of planted coniferous forests. Young planted coniferous forests were observed only in Cluster-2 (After 2010) or some parts of Cluster-3 (After 2010) (No. 2, No. 8, No. 9, No. 10 and No. 15). This indicates that planted forest management involving logging and replanting is rarely being conducted in these areas.

In the 1980s, many shrine/temple forests showed significant differences with the surrounding forest cover, mainly due to the proportion of mature to old-growth planted coniferous forests. However, in the forest cover after 2010, the aging of planted coniferous forests is evident in the shrine/temple forests and the surrounding areas (buffer). Additionally, young planted coniferous forests are limited to very few locations. As a result, except for these few exceptions (No. 5 and No. 6), there are no significant differences between the shrine/temple forest cover and the surrounding areas (buffer)’s forest cover.

4. Discussion

4.1. The Historical Forest Dynamics of Shrine/Temple Forests in Kyoto City

An overview of the forest dynamics in shrine/temple forests during the study periods (from the 1890s to the 1980s, and from the 1980s to after 2010) reveals significant changes from the 1890s to the 1980s. Changes include the afforestation of conifer trees; Cryptomeria japonica or Chamaecyparis obtusa in the early stages, especially in former fuelwood forests; and the transformation of Cryptomeria japonica forests and Pinus densiflora forests into broad-leaved forests or mixed forests. On the other hand, from the 1980s to after 2010, there were some instances of silvicultural practices involving the logging and replanting of planted coniferous forests. However, these practices were observed in only about 30% of the entire planted coniferous forests in shrine/temple forests. The predominant changes during this period were natural transitions, such as the growth of planted coniferous forests and the transformation of Pinus densiflora forests into broad-leaved forests when affected by pine wilt disease.

In this section, an analysis of the background factors that led to afforestation in former fuelwood forests from the 1890s to the 1980s is presented as follows:

One of the selection criteria for the Kyoto Prefectural Shrine/Temple Forests Survey is a forest age of 75 years and older. It can be assumed that many of the mature to old-growth planted coniferous forests observed in the 1980s were planted between 1868 and 1926 (Japanese Meiji and Taisho periods). In the shrine/temple forests near the urban areas of Kyoto city, rampant logging occurred in the 1870s, whereby shrine/temples tried to sell timbers to generate income before the government’s nationalization of their lands [23]. Regarding these areas affected by rampant logging, large-scale afforestation, mainly consisting of Cryptomeria japonica and Chamaecyparis obtusa, was carried out by the national government from 1884 to 1892 [59] (Annotation A5 in Appendix A). According to Nakajima (1996) [23], during this period, there was a shift in the management policy of national forests in the Kyoto prefecture. Specifically, in 1886, the management of national forests in the Kyoto prefecture was transferred from the Kyoto Prefectural Government to the Ministry of Agriculture and Commerce. After this transfer, the management policy of national forests shifted from “preserving and restoring scenic beauty” to “conserving the national territory and increasing forest products”. As a result, the landscape conservation practices that had been carried out in the Kyoto prefecture for many years were largely disregarded, and a standardized management approach was implemented by the national government. Between the 1890s and 1920s, there were significant developments regarding the ownership and management of shrine/temple forests. Various laws were enacted, such as the National Forest Act of 1899. These laws resulted in the return of the ownership and management of certain forested areas to shrines and temples [60,61]. Additionally, the Forest Act was revised in 1907 to establish the forest planning system, which allowed shrines and temples to engage in forestry management on the returned forested lands. This policy marked a recognition of the rights of shrines and temples to manage and utilize their own forest resources for forestry activities. During the same period, in 1909, the Kyoto Prefectural government issued a directive on planting in shrine grounds and the management of shrine forests (issued by the Ministry of Home Affairs, document number 629) [62]. This directive aimed to encourage forestry management through afforestation in shrine forests, implying that many of the planted coniferous forests established around shrines and temples between the 1890s and 1920s were likely created by either the government or shrines and temples, with the intention of conserving their owned lands and producing timber.

4.2. The History of the Cultural Values of Shrine/Temple Forests

Considering the forest dynamics and the results of the cluster analysis, an examination of the history of the cultural values of shrine/temple forests is conducted as follows.

4.2.1. The Cultural Value of Landscape Forests

As mentioned above, the cultural value of landscape forests is defined as their ability to support the solemnity of shrines and temples while simultaneously providing construction materials for the management of shrines and temples.

Summarizing the changes in landscape forests, the majority of landscape forests existent in the 1890s had disappeared by the 1980s due to the expansion of broad-leaved forests caused by pine wilt disease. Since 2010, semi-natural forests, such as broad-leaved forests or conifer and broad-leaved mixed forests, have become predominant, assuming a landscaping function (except for No. 2). Culturally valuable landscape forests previously supported by mature Pinus densiflora forests and coniferous forests have shifted to become semi-natural forests and mainly broad-leaf forests. This implies that the functions encompassed within the cultural value of landscape forests have shifted without prioritizing the timber production function that traditionally coexisted. If the current policies continue to only support natural or semi-natural forest restoration and landscape conservation, the timber production function of landscape forests will be lost. This will result in a further loss of traditional knowledge regarding landscape forestry that enables timber production while maintaining the religious solemnity of shrine/temple forests.

4.2.2. The Cultural Value of Fuelwood Forests

As mentioned earlier in the study, the cultural value of fuelwood forests is defined as their ability to provide a sustainable supply of firewood and charcoal materials for supporting the livelihoods of people associated with shrines and temples.

Approximately half of shrine/temple forests were classified as fuelwood forests in the 1890s. However, many of the fuelwood forests had undergone significant transformations by the 1980s, transitioning into planted coniferous forests or semi-natural forests such as broad-leaf forests or conifer and broad-leaf mixed forests. This indicated that, by the 1980s, the cultural value of fuelwood forests had disappeared compared to the value of fuelwood forests in the 1890s. There was a conversion to semi-natural forests that resulted in a functional shift from the fuelwood production function into landscaping function. This shift contributed to a land conservation function and timber production function within the planted coniferous forests. The shift in the function of the forests in the 1980s implies the multi-layered cultural value of landscape forests, land conservation forests, or timber production forests coexisting simultaneously within the 1890s culturally valued fuelwood forest. However, after 2010, intentional forest management for the timber production function was not efficiently maintained. The continued absence of management indicates that the multi-layered cultural values of these forests since the 1980s will not be renewed. Consequently, this will lead to the homogenization of the cultural characteristics of shrine/temple forests in the long term. Such homogenization of shrine/temple forests will affect the opportunities associated with cultural traditional practices like fuelwood collection and will regenerate the unique forest landscape shaped by these traditional practices related to forest resource use.

4.2.3. The Cultural Value of Mixed-Function Forests (Landscape and Fuelwood)

The cultural value of mixed-function forests is defined as the their ability to support the solemnity of shrines and temples, as well as accommodate the production of timber, fuelwood, and charcoal materials. In the collected data, larger shrine/temple forests (20 ha or more) are classified into this category. The trends observed in forest cover changes up until 2010 in this category were similar to the two categories mentioned above. The landscaping function was mainly supported by semi-natural forests. The fuelwood production function had disappeared, transforming into the landscaping function, land conservation function, and timber production function.

The cultural value of mixed-function forests varied between shrine/temple forests. Some preserved the characteristics of landscape forests, especially with the existence of Pinus densiflora. On the other hand, other shrine/temple forests had a higher proportion of planted coniferous forests, acting as timber production forests, and some maintained both cultural values. The cultural value of mixed-function forests can be described as the value associated with the simultaneous preservation of different functions in different areas within shrine/temple forests. Through zoning and other forest management practices, the mixed function of these forest adds intrinsic value to the forest as more than a natural resource. Maintaining such a cultural practice requires a wealth of knowledge and an understanding of the forest management practices best suited to the mixed-function forest environment. Therefore, if management practices disregard the cultural value of mixed-function forests, traditional knowledge regarding the selection of suitable locations for functional categories of forests, as well as the skills and techniques required to simultaneously maintain such a special forest dynamic, may also be lost with time.

5. Conclusions

In this study, we examined forest cover change in shrine/temple forests and their cultural value in Kyoto city. The comparison between forest cover in the 1890s and in the 1980s revealed two major trends: (i) landscape forests have shifted towards semi-natural forests, and (ii) fuelwood forests have transformed into planted coniferous forests. In the first trend, the analysis of cultural value suggested that the current policies, which only support natural or semi-natural forest restoration and landscape conservation, may lead to the trivialization of the cultural value of landscape forests. As mentioned earlier, a continued lack of forest management and the transition to climax forests will lead to a decrease in biodiversity [36,38,39]. For the conservation of the cultural values of landscape forests, in addition to maintaining semi-natural forests through regular management, reforestation with Pinus densiflora or planted conifer forests can contribute to enhancing the cultural value of landscape forests. The second trend indicated that multi-layered cultural value has been formed through the history of forest management, although functional change has occurred. In some of the shrine/temple forests examined in this study, it was found that planted forest management strategies involving logging and replanting were conducted until the 1980s. This implies that within the context of human activities, the management of forest stands shaped by ongoing practices has been consistently carried out. This maintenance, even when the forest has different functions, such is the case with fuelwood forests or timber production forests, signifies the renewal and preservation of layered cultural values even after the Meiji period (1868–1912) in the mountainous and foothill areas of Kyoto city.

Based on the above, the cultural value of planted coniferous forests between the 1890s and 1920s can be observed as being intentionally established for land conservation and timber production. This is particularly in areas that were fuelwood forests in the 1890s. These planted coniferous forests were found to have shaped the distinctive forest cover of shrine/temple forests in later years, indicating that planted coniferous forests within shrine/temple forests have retained their multi-layered cultural value dating back a century. However, it was found that, after 2010, planted forest management strategies within shrine/temple forests have been rarely implemented. This has resulted in the aging of planted coniferous forests that were originally intended for land conservation and timber production. The ongoing lack of management implies that these forests’ multi-layered cultural value will not be renewed, potentially leading to the homogenization of cultural characteristics in the long term.

For the shrine/temple forests in Kyoto city, the preservation and regeneration of Pinus densiflora forests have been emphasized. Similarly, for the planted coniferous forests without forest management practices, there has been a push towards transitioning these areas to semi-natural forests, with the intended purpose of conserving historical landscapes and enhancing biodiversity [24]. As revealed in this study, the planted coniferous forests formed within shrine/temple forests are not adequately managed. Consequently, these planted forests are being converted into semi-natural forests that require less management, without considering their multi-layered cultural values dating back a century. In this process, after intense thinning, a mixed forest of coniferous and broad-leaved trees will be achieved through natural regeneration [24]. There is a concern that the cultural features formed in interaction with humans have become fragmented, and that tracking those cultural features will be difficult.

This study revealed that the planted coniferous forests in the mountainous and foothill areas of Kyoto city have multi-layered cultural value. The assessment and conservation of the cultural value of planted forests run counter to the global trend of restoring “natural” forests. However, the disregard for cultural characteristics shaped by historical human activities implies that these forests’ multi-layered cultural value will not be renewed. While past traditional management strategies continue to influence forest characteristics [4], there is a concern that in the long term, it may lead to the homogenization of the cultural characteristics of each forest. There is also the potential loss of traditional knowledge and techniques that have supported the cultural value of shrine/temple forests. A further detailed examination of individual cases will be required to address the homogenization of cultural characteristics and the loss of traditional knowledge and techniques. However, it must be considered that, in this study, the differences in forest cover between the shrine/temple forests and their buffer area were no longer clear after 2010. There is a risk that the distinctiveness of shrine/temple forests and the characteristics and cultural values formed individually by each shrine/temple may disappear. In the conservation of forests formed in close association with human activities, such as shrine/temple forests, there is a need to understand forest dynamics and evaluate cultural values through a historical approach. Due to the varied size and historical background of each shrine/temple forest, the administration may find it challenging to provide comprehensive guidelines or policies through a historical approach. Based on historical evidence, a bottom-up approach would be more feasible. In this context, individual shrines and temples taking the initiative in zoning their forest for the inheritance of cultural value would be effective, whilst conserving their historical landscapes and biodiversity.