Cultural Ecosystem Services Research Progress and Future Prospects: A Review

: Cultural ecosystem services (CESs) are direct inﬂuences on human well-being and sustainable development, and they have become increasingly important in the development and progress of society. This paper reviewed 1248 papers on CESs by using VOSviewer and CiteSpace software. On this basis, we also reviewed 179 selected empirical papers related to this ﬁeld, with regards to geographical locations, study objects, evaluation methods, and categories. The results show that: (1) the number of publications related to cultural services has increased year by year, and Ecosystem Services, Ecological Indicators, and Sustainability are the top three journals with the greatest amount of research published; (2) ecosystem services, beneﬁts, management, and social–ecological systems are perceptions closely related to CES research, which are also popular topics in the ﬁeld; (3) the results of a keyword detection show that the cultural landscapes, patterns, mental health services, social value, and other keywords were popular keywords used from 2005 to 2021; (4) CESs have mainly attracted the attention of many developed countries, and the cultural services in cities, oceans, and coastal areas have become the focus for researchers; (5) the recreation and ecotourism, and aesthetic values are the common categories of the CES empirical studies, while the knowledge system and the cultural diversity are the two categories with the least amount of research; (6) the evaluation methods of CESs are mostly carried out using a traditional questionnaire and interview, but mapping and modeling methods have been widely used in recent years.


Introduction
Many interesting concepts and topics have emerged through the process of interaction between humans and nature. Nature has made significant improvements to the well-being of humans through its diverse material and nonmaterial contributions. Humans can not only obtain food, raw materials, and clean water from the natural ecosystem, but they can also gain some non-material benefits, such as leisure and recreation, spiritual fulfillment, personal development, social relations, and aesthetic experience [1]. These nonmaterial interests are defined as the cultural ecosystem services (CESs) that link humans with nature. These interests are defined by the Millennium Ecosystem Assessment (MEA) as the nonmaterial benefits that people obtain from ecosystems through spiritual fulfillment, cognitive development, thinking, recreation, and aesthetic experience [2]. With the deepening of people's understanding of CESs, we can also consider these services as benefits that are provided by nature in order to improve quality of life [3].
In 1997, Costanza estimated the value of cultural services [4] and clarified that CESs have an important value, which promoted the research on the value and monetization of CESs. At the same time, it also made other authors realize the economic value and importance of CESs [5,6]. Since the MEA was held in 2005, social organizations and the tion [33], and so on, which makes an insufficient contribution to the overall understanding of the research hotspots and characteristics of cultural services. At present, the status of cultural service research in the overall ecosystem service research is increasing. We need to comprehensively evaluate some conclusions of the past research on cultural services through quantitative means in order to provide a reference basis and some clarification for future research. Bibliometric analysis can provide a pathway for the analysis of a large number of documents and identify the hotspots and evolutionary characteristics of previous studies [40,41]. However, bibliometric analysis has some limitations in literature content review. Systematic quantitative review can make up for the lack of understanding of the research content. Therefore, this study combines bibliometric analysis with systematic review methods. The former is used to analyze a large number of CES studies from 2005 to 2021, and the latter is used to review empirical studies. In this review, we try to review the following issues: (1) exploring the current research status and research hotspots of cultural services; (2) analyzing the geographic location, research objects, service categories, and research methods of an empirical study on cultural services; and (3) discussing the loopholes and deficiencies of the current research on cultural services and considering the future research prospects.

Papers Selection
The reviewed papers include two parts: the first comprises all CES-related peerreviewed journal articles from 2005 to 2021, and the second comprises empirical papers that are more closely related to this topic.
In the first part, for bibliometric analysis, papers published between 1 January 2005 and 31 December 2021 with the search terms "cultural ecosystem services" or "cultural services" in the titles, keywords, and abstracts were obtained from the Web of Science Core Collection (WoSCC), which is one of the primary sources for most bibliometric analyses [42]. A total of 1272 publications were obtained. All book chapters and conference abstracts were excluded, resulting in 1248 valid papers ( Figure 1).
limited to the discussion of cultural service indicators [14], research methods [39], classi fication [33], and so on, which makes an insufficient contribution to the overall under standing of the research hotspots and characteristics of cultural services. At present, th status of cultural service research in the overall ecosystem service research is increasing We need to comprehensively evaluate some conclusions of the past research on cultura services through quantitative means in order to provide a reference basis and some clari fication for future research. Bibliometric analysis can provide a pathway for the analysi of a large number of documents and identify the hotspots and evolutionary characteristic of previous studies [40,41]. However, bibliometric analysis has some limitations in litera ture content review. Systematic quantitative review can make up for the lack of under standing of the research content. Therefore, this study combines bibliometric analysis wit systematic review methods. The former is used to analyze a large number of CES studie from 2005 to 2021, and the latter is used to review empirical studies. In this review, we tr to review the following issues: (1) exploring the current research status and researc hotspots of cultural services; (2) analyzing the geographic location, research objects, ser vice categories, and research methods of an empirical study on cultural services; and (3 discussing the loopholes and deficiencies of the current research on cultural services and considering the future research prospects.

Papers Selection
The reviewed papers include two parts: the first comprises all CES-related peer-re viewed journal articles from 2005 to 2021, and the second comprises empirical papers tha are more closely related to this topic.
In the first part, for bibliometric analysis, papers published between January 1, 200 and December 31, 2021 with the search terms "cultural ecosystem services" or "cultura services" in the titles, keywords, and abstracts were obtained from the Web of Scienc Core Collection (WoSCC), which is one of the primary sources for most bibliometric anal yses [42]. A total of 1272 publications were obtained. All book chapters and conferenc abstracts were excluded, resulting in 1248 valid papers ( Figure 1).   In the second part, selected empirical papers were used for quantitative analyses to explore the state of CES empirical research and provide supportive evidence for future practice. In the previous review, two methods of quantitative review paper selection were included. One was to review all types of journal papers (e.g., case studies, reviews, and conceptual papers) (e.g., Milcu et al., 2013 [10]) and the other was to take only empirical research as the review object (e.g., Cheng et al., 2021 [43]). This review belongs to the second category. The process of the paper selection was as follows. First, 1248 papers obtained from WoSCC were imported into Endnote software for management. Second, the search function of Endnote was used to closely search related studies for the terms "cultural ecosystem services" in titles and keywords, and 225 closely related cultural services papers were obtained. Finally, non-empirical papers were eliminated by reading them one by one, and 179 empirical papers were selected for systematic review and analysis of CES categories and evaluation methods. Through an extensive literature review, the quantitative statistical results were explained, together with descriptive analysis (Figure 1).

VOSviewer
VOSviewer (https://www.vosviewer.com/, accessed on 24 January 2022) is a software tool used for constructing and visualizing bibliometric networks [44]. The software can construct networks of keywords, authors, organizations, countries, and cited references [45]. Keywords are the concentration and summary of the content and themes of the literature. By analyzing the high-frequency keywords in the whole field, the hotspots and frontiers of the research in this field can be explored. Keywords with emergent characteristics are words with a high-frequency change rate over a certain period of time, which can reflect the research trend [46]. Additionally, keyword co-occurrence analysis can detect keywords that appear in the same publication and identify frequently used keywords [45,47]. In this review, we construct a co-occurrence map in terms of keywords by using VOSviewer 1.6.18, which created by Nees Jan van Eck and Ludo Waltman in Leiden, The Netherlands.

CiteSpace
CiteSpace is a bibliometric analysis software developed by Dr. Chaomei Chen [48]. Cocitation analysis, such as analysis of cited reference, cited author, and cited journal, which is an important tool of CiteSpace, is widely used in quantitative literature review [49,50]. Citation burst analysis of keywords can reflect changes in interest in a professional field, thereby revealing research hotspots in different periods [51]. Therefore, this paper analyzes the research hotspots and frontiers of CES through the keyword citation burst tool of CiteSpace software.

Classification of the Identified Papers
To ensure the reliability of the review, a question set was formed with reference to Milcu's review [10]. For each reviewed paper, six questions that were formulated on the basis of our research objectives and based on a preliminary literature review and expert judgment were answered (see Table A1, Appendix A for more details). The questions sought to gather basic information about the reviewed literature: (1) whether the article was an empirical study, (2) the geographic location of the studies, (3) the research object of the studies, (4) CES categories addressed by the study, (5) the CES evaluation method, and (6) whether the paper used monetary or non-monetary methods.

Geographic Location of Studies
The geographic locations were recorded during the review of each paper. According to Cheng's guidance [43], we recorded the countries that conducted the study to indicate their geographic location. If a study was carried out in several countries, we would use regional descriptions to record its location, such as "Europe" or "Global."  [43]. Exploring different research objects is also conducive to our understanding of the sources of CESs and the process of obtaining them, which contributes to new knowledge systems for cultural services. For example, to study the cultural services provided by the marine ecosystem, some scholars chose the marine ecosystem as the research object [52], whereas others chose shellfish as the research object [53]. Although they all come from nature, the CES categories that they provide highly differ. The research subjects of this review were categorized into two groups. The first was a specific study area, which included urban, rural, forest, and park areas. The second category included animals and vegetation, such as plants and trees. Studies that contained multiple research objects were counted. For example, regional parks in urban areas are considered regional and urban parks.

Cultural Ecosystem Service Categories
This review examined CES categories based on the Millennium Ecosystem Assessment classification, including aesthetic values, cultural diversity, cultural heritage values, educational values, inspiration, recreation and ecotourism, sense of place, spiritual and religious values, social relations, and knowledge systems. The CES category of each empirical paper matching the MEA category was counted in this review [39].

Evaluation Methods
Many types of evaluation methods exist in the existing empirical research. However, research methods can generally be divided into two groups: monetary and nonmonetary [39,54]. On the basis of this classification, statistics were created, and all the methods used in each paper were classified. If multiple methods were used in the same article, they were recorded twice or more. For example, if the main method of a study is expert-based, but it is carried out in the form of interviews, expert-based and interview methods will be recorded separately.

Overview of General Papers
As Figure 2 shows, the number of papers on CESs increased rapidly from 2005 to 2021. The number of published papers has increased significantly, particularly in recent years. In 2021, 206 CES-related papers were published, which was 25.75 times that of 2005. Among all source journals, Ecosystem Services, Ecological Indicators, and Sustainability ranked the top three in terms of publication volume (Table 1).   Figure 3 shows the CES keyword co-occurrence map of 1248 papers. Each circle represents keywords in a given field [55]. The size of the circle reflects the number of papers of the corresponding CES keyword. The distance between two circles indicates the strength of the co-occurrence link between the keywords, and the closer the two circles are located to each other, the stronger the co-occurrence link between the corresponding keywords [56]. Colors represent clusters of keywords with strong co-occurrence links, and lines are used to indicate the link strengths between keywords. As shown in Figure 3, there were seven clusters in total with 14,449 links, and the total link strengths were 27,916. The central keyword of cluster 1 was cultural ecosystem services; the number of occurrences was 371, and the total link strength was 2680 (Table 2). It can be concluded that cluster 1 has strong links with cluster 2, whose central keyword was ecosystem services with an occurrence of 231 times, and the total link strength was 1644. The number of occurrences of other central keywords, such as management (co-occurrence: 217), biodiversity (co-occurrence: 176), perception (co-occurrence: 104), and benefits (co-occurrence: 77) were all above 50, with only social-ecological systems (27) being less than 50, while the total link strengths were all over 200 ( Table 2). The full set of co-occurrences of each keyword can be found in the review database.  Figure 3 shows the CES keyword co-occurrence map of 1248 papers. Each circle represents keywords in a given field [55]. The size of the circle reflects the number of papers of the corresponding CES keyword. The distance between two circles indicates the strength of the co-occurrence link between the keywords, and the closer the two circles are located to each other, the stronger the co-occurrence link between the corresponding keywords [56]. Colors represent clusters of keywords with strong co-occurrence links, and lines are used to indicate the link strengths between keywords. As shown in Figure 3, there were seven clusters in total with 14,449 links, and the total link strengths were 27,916. The central keyword of cluster 1 was cultural ecosystem services; the number of occurrences was 371, and the total link strength was 2680 (Table 2). It can be concluded that cluster 1 has strong links with cluster 2, whose central keyword was ecosystem services with an occurrence of 231 times, and the total link strength was 1644. The number of occurrences of other central keywords, such as management (co-occurrence: 217), biodiversity (co-occurrence: 176), perception (co-occurrence: 104), and benefits (co-occurrence: 77) were all above 50, with only social-ecological systems (27) being less than 50, while the total link strengths were all over 200 ( Table 2). The full set of co-occurrences of each keyword can be found in the review database.   Social-ecological systems 27 241

Burst Keywords Detection
The burst strength of burst detection indicates the intensity of focus on a given field [57]. The cultural landscape has been the main topic since 2013, with focus subsequently diversifying to include "pattern" (strength: 6.57) and "mental health service" (strength: 5.03); these keywords' strengths were above 5 and reflect the degree of their influence on CES research (Table 3). In terms of the keyword evolution process, "behavior"

Burst Keywords Detection
The burst strength of burst detection indicates the intensity of focus on a given field [57]. The cultural landscape has been the main topic since 2013, with focus subsequently diversifying to include "pattern" (strength: 6.57) and "mental health service" (strength: 5.03); these keywords' strengths were above 5 and reflect the degree of their influence on CES research (Table 3). In terms of the keyword evolution process, "behavior"  (Table 3).

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Geographic Location of the Studies
The reviewed studies have a clear geographic focus on Europe and North America. As shown in Figure 4, China ranked first with 18 studies, and the number of publications here accounted for 10.1% of the total. The USA ranked second with 17 studies, accounting for 8.9%, and Germany was third with 8.4%. In addition, the UK, Italy, Europe, Spain, Chile, and Australia had more than six each, and fewer than five studies were conducted in the remaining countries. In some countries, only one study was conducted, which is not shown in Figure 4.

Research Object of the Studies
Many studies have focused on the CESs of cities, oceans, and forests. Figure 5, most studies concentrated on urban areas (16.8%); the second most a focus on marine and coastal areas (12.8%); the third most common was a fo (11.7%). Of the 179 studies, 21 focused on regional and local areas. In additio mainly explored specific animals and plants, such as shellfish, sea trout, and

Research Object of the Studies
Many studies have focused on the CESs of cities, oceans, and forests. As shown in Figure 5, most studies concentrated on urban areas (16.8%); the second most common was a focus on marine and coastal areas (12.8%); the third most common was a focus on forests (11.7%). Of the 179 studies, 21 focused on regional and local areas. In addition, six studies mainly explored specific animals and plants, such as shellfish, sea trout, and trees.  Figure 5, most studies concentrated on urban areas (16.8%); the second most c a focus on marine and coastal areas (12.8%); the third most common was a foc (11.7%). Of the 179 studies, 21 focused on regional and local areas. In addition mainly explored specific animals and plants, such as shellfish, sea trout, and

CES Categories
The recreation and ecotourism services are the most used CES categori empirical papers assessing the recreation and ecotourism services, which ac 60.9% of the reviewed studies, followed by aesthetic values, cultural heritag itual and religious values (87,48.6%). Contrastingly, knowledge systems and versity were the least studied categories in the research field (Table 4).

CES Categories
The recreation and ecotourism services are the most used CES categories, with 109 empirical papers assessing the recreation and ecotourism services, which accounted for 60.9% of the reviewed studies, followed by aesthetic values, cultural heritage, and spiritual and religious values (87,48.6%). Contrastingly, knowledge systems and cultural diversity were the least studied categories in the research field (Table 4).

Evaluation Methods
Twenty-four CES evaluation methods were used. As Figure 6 shows, of all the 179 empirical papers, most studies used questionnaires to evaluate CESs, which ranked first among all methods (34.1%), followed by evaluation models (28.5%), interview methods (18.4%), participatory mapping methods (15.6%), and the social media method (14.0%). The number of studies using quantitative calculation, focus groups, observation, expert-based methods, and contingent valuation accounted for 2.8% to 6.7%. The use of other methods ranged from 0.6% to 2.2% in all studies. Moreover, the majority (92%) used non-monetary methods (see Supplementary Materials for more details).
The number of studies using quantitative calculation, focus groups, observa based methods, and contingent valuation accounted for 2.8% to 6.7%. The methods ranged from 0.6% to 2.2% in all studies. Moreover, the majority (92% monetary methods (see Supplementary Materials for more details).

Static Characteristics of CES Research Keywords
We used VOSviewer 1.6.18 to analyze the co-occurrence network of keyw research. It is found that "ecological services", "management", "perception" "biodiversity" and "social-ecological systems" are the most discussed topics ing research, and each keyword has a certain relationship (Figure 3). There ar ters in Figure 3 and cluster 1 has "cultural ecosystem services" as the central k

. Static Characteristics of CES Research Keywords
We used VOSviewer 1.6.18 to analyze the co-occurrence network of keywords in CES research. It is found that "ecological services", "management", "perception", "benefits", "biodiversity" and "social-ecological systems" are the most discussed topics in the existing research, and each keyword has a certain relationship (Figure 3). There are seven clusters in Figure 3 and cluster 1 has "cultural ecosystem services" as the central keyword. As the largest cluster, cultural ecological services are closely related to cluster 2 (central keyword: ecosystem services), cluster 6 (central keyword: social-ecological systems), and cluster 7 (central keyword: benefits). On the one hand, this shows that cultural services as a branch of ecosystem service research are an important part of ecosystem services. In empirical research, some studies on cultural services are often related to ecological services, such as provisioning services and supporting services. On the other hand, cultural services connect to nature and human beings and are the non-material benefits that people obtain from the natural ecosystem. Cluster 3 has "management" as the core keyword. In this cluster, national parks, cultural landscapes, and other keywords are highly related to management. In practice, many researchers have provided many policy suggestions for the management of national parks and cultural landscapes through research on cultural services [29,[58][59][60]. For example, Brown [58] used the participatory mapping method to investigate the value of Australian National Parks in land use and found that national parks have the potential to manage land use contradictions. Cluster 4 takes "biodiversity" as the core keyword, and stakeholders and cultural services are closely related to it. Biodiversity can provide richer cultural services for different stakeholders [61,62]. However, the protection of biodiversity requires the joint efforts of stakeholders [63]. The central keyword of cluster 5 is "perception", which is closely related to cultural services. In the research of cultural services, the perception of social groups such as community residents and tourists are both hot topics [31,61]. Mapping the perception of cultural services of relevant groups is an effective way to understand the intangible benefits, which is conducive to the visualization of cultural services and provides a basis for the management of local landscapes [64].

Dynamic Evolution of CES Research Keywords
The results of keyword detection show that cultural landscape, pattern, mental health services, social value, and other keywords were hot keywords from 2005 to 2021. To a certain extent, the evolution of keywords is consistent with the key areas of concern in various periods of cultural services. According to the detection results, the hot spots of cultural services can be roughly divided into three stages. The first stage (2005-2011) mainly focused on consumer behavior, cultural diversity, mental health services, and other topics. This stage belongs to the exploration stage of cultural services research. The MEA has promoted the development of CES research. However, as the academic research on it has just started, the topics discussed tend to be diversified, and the various hot keywords have not yet formed a close relationship. Many studies are still discussing the concept and connotation of cultural services, mostly qualitative studies [65,66]. The second stage (2012-2017) paid more attention to cultural landscapes, knowledge related to cultural services, agricultural cultural services, and so on. Since 2012, which coincided with the establishment of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES), more diverse academic fields have been represented by relevant studies [1]. The relationships, mechanisms, approaches, and knowledge systems between nature and human well-being have become the focus of academia. Some authors have conducted mapping research on cultural services generated by cultural landscapes, which promoted the quantitative measurement of cultural services [52,67]. Researchers also found that agricultural production can not only provide material benefits, such as food and wood but also shape agricultural landscapes in the process of humans engaging in agricultural activities [68,69]. These landscapes can also provide non-material benefits such as recreation and leisure for human beings [70][71][72]. This knowledge is conducive to forming a more comprehensive understanding of cultural services, so as to provide more information for the management of the agricultural landscape [73]. At this stage, the academic circles also began to pay attention to the discussion of research methods and the willingness to pay method alone became a hot topic. In the actual research process, the willingness to pay method has become an important means to measure the economic value of cultural services [74,75]. The third stage (2018-2021) focused on ecological restoration, outdoor recreation services, cultural service demand assessment, cultural service spatial pattern, social media data evaluation methods, and other topics. This stage belongs to the rapid development stage of cultural service research. The number of published documents began to increase explosively every year. The quantitative evaluation of cultural services has been further developed with the support of the development of information technology [76,77]. Therefore, scholars have begun to make more use of geographic information technology to evaluate the spatial pattern of cultural services and the demand for cultural services, which has provided a lot of valuable information for the management and decision-making of cultural services and the protection and development of landscapes.

CES in Developing Countries and Rural Areas Deserve Attention
In this study, 179 identified empirical papers were reviewed using quantitative methods. Of all the papers, many studies focus on the evaluation of CESs in developed countries and regions, especially in Europe and North America [78][79][80]. However, China has gradually become a focus of the research into CESs, and many empirical studies have emerged recently [81,82]. However, CESs have received limited attention in most developing countries and rural areas. The possible reasons are as follows: first, the relatively backward local economic level has not fully met the basic material needs of residents; second, the knowledge and value systems of most developing countries and indigenous communities have not yet formed effective communication with developed countries, and cultural services have a series of characteristics, such as being intangible, subjective, socially constructed, and dependent on human perception, which all require different tools and methods for their understanding [5,9,83]. This poses a challenge for research on cultural services in developing countries and rural areas. In recent years, many authors have begun to realize this problem and gradually pay attention to CESs in third-world countries. In fact, when CES research in developed countries is focused, it may be possible to draw some interesting or different conclusions from the CES research in some developing countries. In this sense, authors such as Santarem have conducted research on the supply of CESs in the Sahara-Sahel Ecological Reserve in Africa [84]. In terms of the study object, CESs in urban areas were one of the most important topics in research because some researchers thought that cities are complex adaptive systems embedded within even more complex adaptive ecosystems. In contrast, the attention focused on townships and rural areas does not match their importance in the provision of CES [85]. Therefore, the rural landscape deserves further attention in the research.

The Categories of CES Need Comprehensive Consideration
In all CES categories, many authors believe that aesthetic value is relatively easy for residents and tourists to perceive [86], and recreation and ecotourism can be assessed by landscape indicators or social media data [87]. Therefore, aesthetic value and recreational value are relatively easy to quantify, whether using traditional questionnaires or GIS mapping [88,89]. With the deepening of CES research, many authors advocate for supplementing the CES categories proposed by the MEA. As a result, several categories not explicitly included in the MEA framework are emerging in the research, such as social inclusion [90], identity [72], and existence value [91]. These categories are not only enriching the category of CES but also playing an important role in amending and supplementing the original CES evaluation framework. In addition, it is also conducive to improving people's understanding of the spiritual benefits they have obtained from the natural system and realizing that the generation of non-material benefits is a dynamic process. Although the ecosystem provides people with various CESs, it also greatly meets the spiritual needs of human beings. Moreover, beyond positive services, there is also some negative CESs derived from the human and natural system, namely disservices, which are more derived from the negative effects of human interaction with nature, such as crowdedness, unpleasantness, scariness, and noise [92].
Future research should evaluate comprehensive CES categories to provide more details for CES management. However, this not only requires researchers to establish more comprehensive CES evaluation indicators but also necessitates the development of multiple source data. In addition, the combination of various methods can help evaluate CESs. Generally, depicting the complexity of CES using a single indicator, method, or data point is difficult. Ultimately, the knowledge and methods of different disciplines must be integrated to solve this problem.

Mapping CESs Provides More Detail for Large-Scale Management and Planning
With the rapid development of global urbanization, the CES is of great significance to human health [93][94][95]. Ecosystem services exhibit strong spatial and temporal characteristics [96,97]. The application of the research method plays a crucial role in the CES evaluation process. This review shows that earlier research considers ecosystem services to be of significant economic value. Therefore, monetary methods have been adopted in many studies, such as the travel cost method [67] and the contingent valuation method [98]. However, in recent years, there has been a change in the understanding of CESs. Many authors believe that CESs are intangible and that it is difficult to assign them a monetary value [9]. Although the monetary method is still regarded as one of the main methods of CES valuation [54], our research results show that only a few studies in the current CES empirical research use this method to assess the value of cultural services ( Figure 6). Contrastingly, questionnaires, evaluation models, interviews, and participation mapping have become important methods for evaluating CESs. These methods, such as questionnaires and evaluation models, combined with CES mapping, have become important tools for CES visualization. Therefore, many studies on mapping CESs have emerged, which has played an important role in promoting the visualization of CESs [99,100]. Mapping CESs can overcome the limitations of economic value estimation and explain the spatial heterogeneity of CES demands more intuitively, thus serving planning and management more effectively [64]. In addition, mapping CESs from the perspective of stakeholders can take into account the ecological knowledge of places and people [101], which is beneficial for broadening people's cognition of CESs. Currently, CES mapping has become an important means for the academic community to analyze the preferences of tourists and locals for CESs. For example, Tobias Plieninger's study [13] found that tourists' and residents' perceptions of CESs are related to landscape features and land-cover forms. This has an important reference value for the planning and management of local cultural landscapes and land use. Moreover, the method of mapping CESs is applicable to large-scale research areas and can compare the evaluation results of cultural services in different regions, serving the planning and management of destinations such as tourist destinations and protected areas.

Conclusions
This paper conducts a bibliometric analysis and systematic review of the papers in the CES field. The following conclusions are drawn: (1) the number of publications in cultural service research has increased year by year, and Ecosystem Services, Ecological Indicators, and Sustainability are the top three journals with the most research published in the literature; (2) ecosystem services, benefits, management, social-ecological systems, and perception are closely related to CES research, which are also hot topics in a given field; (3) the results of keyword detection show that cultural landscape, pattern, mental health services, social value, and other keywords are hot keywords from 2005 to 2021; (4) the study of CESs has attracted the attention of developed countries, and the cultural services in urban, marine and coastal areas have become the focus of researchers; (5) recreation and ecotourism and aesthetic values are some common categories of CESs, while knowledge systems and cultural diversity are the two least-evaluated CESs; (6) the evaluation methods of CESs are still dominated by traditional social questionnaires and interviews, and mapping and modeling CESs have also become two important methods in research.
This review combines bibliometric and systematic review methods and obtains a series of valuable results. However, there are still some limitations in this review. First, the papers analyzed in this review are from the WoSCC, and the literature of other databases has not been analyzed. Secondly, this review aimed to explore the CES research hot topics; therefore, the bibliometric analysis carried out in this review is limited to the analysis of keywords, without bibliometric analysis of the authorship, country, references, and other items. Finally, this review uses a quantitative review method to analyze the previous literature only in terms of statistics and quantity. Thus, this review does not instruct the CES research framework or make a sufficient contribution to CES theory. However, it is helpful to recognize the conclusions and characteristics of CES research. Data Availability Statement: Cultural ecosystem services publications from 2005 to 2021 in this study were derived from ISI Web of Science (https://www.webofscience.com/wos/, accessed on 8 January 2022).