The Spatial and Temporal Characteristics and Influencing Factors of Intangible Cultural Heritage in Fujian Province

Abstract

Background: The UNESCO has defined the concept of intangible cultural heritage (ICH). Additionally, China joined the related convention in 2004, incorporating the protection of intangible heritage into its national strategy. By conducting a detailed analysis of the spatial distribution of national and provincial ICHs in Fujian Province, this study aims to explore the characteristics of their spatio-temporal evolution and the related influencing mechanisms, thereby offering theoretical references for the protection of intangible heritage. Utilizing five batches of national-level and seven batches of provincial-level ICH projects in Fujian as samples, this study employs ArcGIS for data organization and applies geographical concentration indices, average nearest neighbor indices, kernel density, and GeoDetector to explore the spatiotemporal distribution characteristics and influencing mechanisms of intangible heritage. There are at least three key findings in our research: First, ICH resources in Fujian exhibit a coastal concentration and inland dispersion pattern, with notable regional and hierarchical disparities. Second, univariate analysis reveals that socio-economic variables (e.g., GDP, population density) exert stronger explanatory power on overall ICH distribution, whereas cultural factors demonstrate more localized effects in less-developed or peripheral areas. Third, bivariate interaction results indicate that the combined influence between some specific socio economic/cultural variables and the other 20 variables generates enhanced synergistic effects, especially in shaping the distribution of selected ICH in national and provincial levels with distinct performances, highlighting the importance of multi-factor coordination in heritage recognition and protection.

1. Introduction

Intangible cultural heritage (ICH) encompasses the living traditions, expressions, and practices that communities recognize as part of their cultural legacy. Since the 2003 UNESCO Convention, ICH has been widely acknowledged as vital for cultural diversity and sustainable development [1,2,3,4]. In many contexts, ICH serves as a symbol of collective identity and social memory, shaping how communities remember, narrate, and legitimize their pasts [5,6,7,8]. In China in particular, ICH enjoys strategic importance: it not only reinforces national identity and “cultural self-confidence,” but it also underpins development policies such as rural revitalization and cultural soft power promotion [9,10,11]. Reflecting this, China has built a multilayered ICH protection system—from the UNESCO list (first Chinese inscription in 2001) through multiple national-level batch listings (starting in 2006) to provincial and municipal registers—anchored by a 2011 national ICH Law [12,13]. This state-led framework is often described as a “production-based” safeguarding paradigm that heavily involves government agencies in nomination, certification and funding [14,15]. Consequently, ICH in China is deeply intertwined with cultural policy and geopolitical agendas (the “heritage regime” perspective).

Against this backdrop, scholars in heritage geography and cultural policy emphasize that the spatial distribution of ICH reflects complex ecological, social, and institutional processes [16,17]. From a cultural-ecological standpoint, ICH patterns emerge from long-term interactions between human societies and their environments: the physical landscape provides the material context for traditions, while social, economic and historical forces shape which practices flourish and where [18,19]. In parallel, cultural policy scholars highlight that recognition criteria, funding mechanisms, and heritage promotion at national and local levels have a decisive impact on ICH/TCH (tangible cultural heritage is abbreviated as TCH in this paper) viability and location [20,21,22]. Crucially, these natural, social and institutional drivers exhibit strong spatial heterogeneity and scale-dependence: factors that dominate at one scale or region may be less important elsewhere. Thus, understanding ICH geography demands attention to multiple dimensions and factor interactions—a point underscored in recent research on heritage distributions [4].

Empirical studies of ICH distribution have proliferated recently, often leveraging geographic information systems (GIS) and spatial statistics. For example, Zhang, Cui [8] mapped China’s 3610 national ICH items and found a pronounced “east-dense, west-sparse” clustering. Similarly, Hu, Yang [23] used geo-detection to reveal that Eastern China holds the most ICH while a clear “polycentric agglomeration” pattern emerges nationwide. At regional scales, scholars have also identified distinctive patterns. Lin, Zhang [24] show that ICH along the Ming Great Wall, there forms a “three cores and one belt” structure, driven by intersecting natural, economic and social factors. Wu, Yang [25] describe a “small aggregation, large dispersion” spatial structure for Huizhou’s heritage, reflecting both geography and cultural networks. Across China’s urbanizing coast–inland gradient, studies consistently find that socio-economic drivers (population, GDP, urbanization) are often more influential for ICH density than purely physical factors [26,27]. For instance, in Fujian Province, Han, Tao [26] demonstrate that at the city level, ICH distribution is most strongly correlated with population size, urbanization and GDP, whereas only at finer (county) scales do physical factors (elevation, vegetation) gain relative importance. Likewise, Ren, Li [27] show for the Guangdong–Hong Kong–Macao region that economic variables dominate ICH clustering, with tourism hubs like Macao and Guangzhou as high-density cores. These findings echo global case studies where ICH tends to agglomerate in centers of wealth and mobility.

Despite these advances, significant gaps remain. Most existing work treats ICH as a homogeneous category or focuses on one administrative level (e.g., national or local), and few directly compare ICH recognized at different scales. In China’s context, past analyses have often examined either national lists or provincial lists in isolation [28,29,30,31]. Moreover, conventional methods (e.g., kernel density, trend surfaces, regression) typically identify single dominant factors but do not fully capture how multiple drivers interact to shape ICH distribution patterns. In heritage studies more broadly, the politics of designation and issues of identity demand attention: for example, listed ICH can reinforce ethnic or regional claims, and thus ICH mapping is inherently spatially political. Yet, quantitative work on ICH in China has rarely integrated this perspective. In sum, we lack a comprehensive, multilevel analysis that (a) juxtaposes national and provincial ICH distributions within one region, (b) incorporates China’s multi-batch nomination history and policy context, and (c) applies spatial frameworks sensitive to factor heterogeneity and interactions.

To address these gaps, this study takes Fujian Province as a case to investigate the spatial differentiation of national- and provincial-level ICH. We employ a multi-stage geospatial strategy: First, we quantify spatial patterns using global indices (geographic concentration) and point-pattern metrics (average nearest neighbor, kernel density) to capture cluster vs. dispersion tendencies. We then apply the GeoDetector model—including the detection of one-factor influence and two-factor interaction—on a 6 km × 6 km grid of Fujian to identify the explanatory power and synergy of 21 human and environmental factors (Figures 5 and 6 and Table 7). By explicitly comparing ICH at two designation levels and across five national versus seven provincial batches, we ask three questions: First, how are national vs. provincial ICH items spatially distributed and clustered in Fujian? We measure and contrast their concentration, dispersion and orientation patterns, elucidating coastal vs. inland or core vs. peripheral differences. Second, what underlying factors drive these spatial patterns? We test a wide set of natural (topography, climate), infrastructural (transport, location), socio-economic (population, GDP, urbanization), and cultural-industry variables, extending previous models. Third, how do the roles of these factors differ between national- and provincial-level ICH? We explore how geographic, socio economic, and cultural contexts result in divergent spatial logics for top-tier versus local heritage.

This paper makes four interrelated contributions to the spatiality study of ICH recognition and protection. First, at the theoretical level, it develops a multi-level, mechanism-oriented perspective by simultaneously examining national- and provincial-level designation systems within a single region, moving beyond static descriptions of “where heritage is” to explain why different levels of ICH concentrate or disperse and how socio-economic structures, cultural platforms (e.g., scenic sites, TCH units) and policy recognition regimes jointly shape spatial inequalities and coverage. Second, methodologically, it integrates geographic concentration indices, average nearest neighbor analysis and kernel density estimation with a 6 km × 6 km grid-based GeoDetector framework, enabling a systematic comparison of both single-factor effects and interaction effects among 21 natural, transportation, socio-economic and cultural-industry variables and thus revealing how specific factor combinations (such as accessibility × population or elevation × economic development) reshape ICH configurations. Third, empirically, it offers the first comprehensive two-tiered analysis of Fujian’s ICH, showing that while both national- and provincial-level items exhibit strong coastal clustering, provincial-level ICH is more numerous and spatially diffuse, extending protection more effectively towards inland and relatively underdeveloped areas, with socio-economic variables emerging as primary drivers and cultural factors (TCH, A-level scenic spots) exerting particularly strong category-specific influences. Finally, in practical terms, the identified interaction mechanisms provide an evidence base for differentiated, spatially sensitive safeguarding strategies—such as strengthening coastal–inland ICH corridors, enhancing support for peripheral communities with rich but under-recognized heritage, and tailoring interventions by ICH category—offering a transferable framework for balancing symbolic concentration with more equitable multi-scalar protection.

In summary, this study advances the spatial analysis of ICH by bridging global theoretical debates and China-specific dynamics, employing rigorous geospatial methods, and highlighting how socio-economic and cultural landscapes shape the recognition and protection of heritage at different levels. These insights not only enrich academic discourse but also support planners and cultural managers in designing spatially differentiated ICH safeguarding and development policies for Fujian and beyond.

2. Research Methodology

2.1. Data Sources

The data utilized in this study were sourced from the official website of the Fujian Provincial Department of Culture and Tourism. The dataset comprises five batches of nationally recognized and seven batches of provincially recognized ICH items, including extended lists of ICH (Beginning with the second batch of the National List of Intangible Cultural Heritage, an extended list was introduced to include identical ICH items that are practiced or inherited in different regions or by different communities and groups. Although the extended items share the same project code as those on the original national list, they exhibit variations in characteristics, inheritance status, and protection institutions. © Official Website of China Intangible Cultural Heritage (https://www.ihchina.cn/project accessed on 11 November 2023). In total, 140 national-level and 767 provincial-level ICH items were selected for analysis. To investigate the spatial distribution and density patterns of ICH across Fujian Province, kernel density estimation (KDE) was conducted using ArcGIS (version 10.7) Both the bandwidth and kernel function were set to default values within the software. Following KDE analysis, the results were reclassified into seven categories using the Natural Breaks (Jenks) classification method, which adaptively identifies cluster thresholds based on the data’s inherent distribution. This approach enables an effective capture of spatial aggregation patterns of ICH items. Additionally, a Geographic Detector model was employed to construct a multidimensional framework of influencing factors across four domains: transportation, natural geography, socio-economic conditions, and cultural industries. A total of 21 explanatory variables were included in the analysis (see

Table 1. Data source.

Data Type	Code	Variables	Data Sources
Natural Factors	X1	Elevation	NASA Earth Observation Data https://neo.gsfc.nasa.gov/ accessed on 7 November 2022
	X2	Slope
	X3	Slope Direction
Transportation Location Factors	X4	Road Density	Open Street Map (OSM) https://www.openstreetmap.org/ accessed on 7 November 2022
	X5	River Network Density
	X6	Length of Highways in Operation	Fujian Statistical Yearbook 2021 http://tjj.fujian.gov.cn/tongjinianjian/dz2021/index.htm accessed on 7 November 2022
	X7	Total Passengers
Socio Economic Factors	X8	Gross Domestic Product
	X9	Primary Industry
	X10	Secondary Industry
	X11	Tertiary Industry
	X12	Per Capita GDP
	X13	Annual Per Capita Disposable Income
	X14	Population Density
	X15	Lever of Township
	X16	Budgetary Expenditures of Local Government
Cultural Industry Factors	X17	Per Capita Expenditure of Rural Households on Education, Culture and Recreation
	X18	Per Capita Expenditure of Urban Households on Education, Culture and Recreation
	X19	Total Per Capita Expenditure on Education, Culture, and Recreation
	X20	A-grade Tourist Attractions	Fujian Provincial Department of Culture and Tourism https://wlt.fujian.gov.cn/ accessed on 7 November 2022
	X21	National Tangible Cultural Heritage (TCH)	Central People’s Government https://www.gov.cn/ accessed on 7 November 2022
		Provincial Tangible Cultural Heritage (TCH)	The People’s Government of Fujian Province https://fujian.gov.cn/ accessed on 7 November 2022

and Figure 4).

In this study, the management and protection services for each ICH item are closely tied to its proximate cultural protection institution. For instance, the protection of “Lantern Riddles (Shishi City)” is specifically implemented through the Shishi City Cultural Center. Although ICH items often span across multiple communities or regions, their localized administration through nearby heritage institutions allows for spatial analysis to be grounded on the geocoded locations of these affiliated cultural units. This approach enables a more accurate and operational mapping of ICH distribution based on the spatial coordinates of their managing institutions.

In addition, considering the influencing factors, existing research has primarily focused on three domains—natural geography, socio-economic conditions, and historical–cultural contexts—typically including variables such as topography, precipitation, water systems, per capita GDP, transportation accessibility, and population density [32,33,34]. However, these studies often covered a relatively narrow scope and seldom incorporated a systematic integration of socio-economic and cultural-industrial factors. Building upon this foundation, our study expands the potential driving factors to 21 variables across four dimensions: natural environment, transportation, socio-economic conditions, and cultural industries. This broader framework allows for a more comprehensive examination of the multifaceted mechanisms influencing the spatial distribution of ICH (Please see Table 1).

2.2. Methodological Model

To ensure a comprehensive understanding of the spatial characteristics of intangible cultural heritage (ICH) in Fujian Province, this study integrates three complementary spatial concentration measures rather than relying on a single index. The Geographical Concentration Index provides a macro-level evaluation of the overall distributional imbalance and concentration intensity across administrative units, capturing regional disparities in heritage recognition. The Average Nearest Neighbor Index refines this perspective by quantifying the proximity and spatial organization of individual heritage sites, identifying whether their distribution is clustered or dispersed. The Kernel Density Analysis, in turn, transforms discrete point data into continuous spatial surfaces, enabling fine-grained visualization of hotspot areas and serving as the spatial basis for subsequent geographic detector analysis. Collectively, these three approaches form a hierarchical analytical framework that bridges macro-level distribution assessment, meso-level structural identification, and micro-level spatial mechanism exploration, thereby enhancing both the robustness and interpretability of spatial analysis results.

2.2.1. Geographic Concentration Index

To assess the spatial unevenness of intangible cultural heritage (ICH) distribution across Fujian Province, this study employs the Geographic Concentration Index (G). This metric captures the extent to which ICH items are concentrated in a few cities or more evenly dispersed, offering a quantitative basis for understanding spatial disparities between coastal and inland areas, as well as between national- and provincial-level heritage. The specific calculation formula is as follows [35,36]:

$$G=100\sqrt{{\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}{\left(\frac{X_{\mathrm{i}}}{T}\right)}^2}}$$

(1)

In this equation: X_i represents the number of ICH items in the i-th prefecture-level city in Fujian Province; T is the total number of ICH items in the province; and nnn is the number of prefecture-level cities. The resulting index G ranges from 0 to 100. A higher value of G indicates a more concentrated spatial distribution, while a lower value suggests a more dispersed pattern.

$$G_0=100\sqrt{{\sum }_{i=1}^n{\left({\displaystyle \frac{X_i^{\mathit{avg}}}{T}}\right)}^2}, X_i^{avg}={\displaystyle \frac{T}{n}}$$

(2)

This study sets a baseline G₀ to represent the hypothetical concentration level under a perfectly uniform distribution across all prefecture-level cities. When G > G₀, the spatial distribution of ICH is considered relatively concentrated; when G < G₀, it is deemed relatively dispersed. This measure is especially useful in comparing spatial disparities between national and provincial heritage items, and in tracing how different types of ICH (e.g., traditional crafts, folk customs, traditional medicine) exhibit varying concentration patterns.

2.2.2. Average Nearest Neighbor

To further characterize the spatial distribution patterns of ICH point features, this study employs the Average Nearest Neighbor (ANN) index [37], which quantifies spatial proximity to distinguish whether heritage sites are clustered, randomly distributed, or uniformly dispersed across Fujian Province. The specific formula for calculation is as follows:

$$ANN={\displaystyle \frac{{\overline{D}}_O}{{\overline{D}}_E}} {\overline{D}}_O={\displaystyle \frac{\sum _{i=1}^n{d}_i}{n}} {\overline{D}}_E={\displaystyle \frac{0.5}{\sqrt{\raisebox{1ex}{$n$}\!\left/ \!\raisebox{-1ex}{$A$}\right.}}}$$

(3)

In Equation (3): ${\overline{D}}_O$ represents the observed average distance between each element and its nearest neighbor, ${\overline{D}}_E$ denotes the expected average distance between specified elements under a random pattern, ANN stands for the average nearest neighbor index, A denotes the study area’s area, and n is the total number of point features under study. When ANN = 1, the point features exhibit a random distribution; when ANN < 1, it indicates an aggregated distribution; when ANN > 1, the point features demonstrate a uniform distribution; and when ANN = 0, the point features are entirely concentrated.

2.2.3. Kernel Density Analysis

Kernel density analysis [38] is used to estimate the spatial concentration of ICH points by calculating the density of heritage and related influencing elements within a specified neighborhood. This method allows for the continuous representation of point (and line) feature intensity across space, aiding in the identification of core areas and spatial gradients. The specific formula for computation is as follows:

$${\widehat{f}}_h\left(x\right)={\displaystyle \frac{1}{n}}{\sum }_{i=1}^n{K}_h\left(x-x_i\right)={\displaystyle \frac{1}{nh}}\sum _{i=1}^{n}K\left({\displaystyle \frac{x-x_i}{h}}\right)$$

(4)

In Equation (4): ${\widehat{f}}_h\left(x\right)$ represents the estimated nuclear density of cultural heritage/influencing elements, n denotes the sample size, h stands for the bandwidth, k(x) denotes the kernel function, and (x − x_i) signifies the distance from the estimated point x to the event x_i. A higher value of ${\widehat{f}}_h\left(x\right)$ indicates a greater concentration in the distribution of ICH/influencing elements, while conversely, a lower value suggests a more dispersed distribution.

2.2.4. GeoDetector Model

The geospatial detector serves as a dependable method for exploring spatial differentiation and quantifying driving factors. It elucidates the intrinsic connections between driving factors and target variables within space [39].

To elucidate the spatial driving mechanisms of intangible cultural heritage (ICH), this study first employs Kernel Density Estimation (KDE) to transform discrete ICH locations into a continuous density surface, identifying clustering patterns. This density surface serves as the dependent variable. Subsequently, the study area was partitioned using a uniform 6 km × 6 km grid, and the centroid of each grid cell (N = 3396) was used to systematically sample values for both the ICH density and a suite of independent variables (e.g., elevation, population density, GDP). This gridding approach ensures a consistent spatial scale and mitigates localized outlier effects.

As the GeoDetector model requires categorical inputs, all continuous independent variables were discretized using the Natural Breaks (Jenks) classification method, partitioning each driver into seven strata. This adaptive method is optimal as it maximizes intra-class similarity while maximizing inter-class differences. This process enhances the model’s capacity to detect potential non-linear relationships between the driving factors and the spatial distribution of ICH. Its specific computational formula is as follows:

The factor detector is utilized for differentiation and factor detection: it can probe the spatial heterogeneity of the dependent variable Y, measured by q-values, expressed as:

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

(5)

In Equation (5): where h = 1,…, L represents the stratification of Y or X; N_h and N denote the number of units in stratum h and the entire region, respectively; ${\sigma }_h^2$ and ${\sigma }^2$ signify the variance within stratum h and the total variance of the entire region, while SSW denotes the within-stratum sum of squares, and SST denotes the total sum of squares for the entire region. The range of q lies between [0, 1], with a higher q indicating a stronger explanatory power of the independent variable X on attribute Y, and conversely, a weaker explanatory power.

The interaction detector identifies the explanatory power of multiple influencing factors on geographical phenomena when they interact [40]. By comparing the q-values of single and paired factors, the interactions are categorized as nonlinear weakening, single linear weakening, double factor enhancement, independence, or nonlinear enhancement, thereby revealing how factor interactions influence the explanatory power of the spatial distribution of intangible cultural heritage. The specific interaction types are shown in

Table 2. Types of interaction between two covariates.

Judgment Basis	Interaction Type
q(X1 ∩ X2) < min(q(X1), q(X2))	Nonlinear weakening
min(q(X1), q(X2)) < q(X1 ∩ X2) < max(q(X1), q(X2))	Single linear weakening
q(X1 ∩ X2) > max(q(X1), q(X2))	Double factor enhancement
q(X1 ∩ X2) = q(X1) + q(X2)	Independence
q(X1 ∩ X2) > q(X1) + q(X2)	Nonlinear enhancement

.

3. Distribution Patterns of ICH in Fujian Province

3.1. Overview of the Spatial Distribution of ICH in Fujian Province

Fujian Province has a total of 767 ICH items (T) and 9 prefecture-level administrative districts (n). The calculated Geographic Concentration Index (G) for intangible heritage in Fujian is 34.67. Assuming an even distribution of intangible heritage across each prefecture-level district in Fujian, each would have approximately 85.78 heritage items. The calculated baseline index G₀ = 33.33. Since G > G₀, this indicates that the distribution of ICH in Fujian exhibits significant spatial clustering characteristics. Overall, geographically, whether at the national or provincial level, the density of ICH spaces in the eastern coastal areas of Fujian Province is higher than in the inland areas. There is a broad clustering along the coast, forming a significant lineation. The main urban areas of three core cities—Fuzhou, Quanzhou, and Xiamen-Zhangzhou—constitute significant centers. Additionally, a core belt comprising the eastern coastal cities of Ningde, Fuzhou, Putian, Quanzhou, Xiamen, and Zhangzhou shows a density trend of being more concentrated in the south than the north, and more in the east than the west (Figure 1 and

Table 3. Quantity of ICH in Fujian Province by Province and City.

Municipal-Level Administrative Divisions	Quanzhou	Fuzhou	Ningde	Zhangzhou	Xiamen	Putian	Longyan	Nanping	Sanming	Total
National-level ICH	35	20	22	16	16	10	8	7	6	140
Percentage/%	25.00	14.29	15.71	11.43	11.43	7.14	5.71	5.00	4.29	1
Provincial-level ICH	129	123	90	89	62	63	60	76	75	767
Percentage/%	16.82	16.04	11.73	11.60	8.08	8.21	7.82	9.91	9.78	1

).

The province has 767 ICH items, including 140 of national significance. National-level heritage is predominantly concentrated along the coastal regions, with sparse and isolated distribution in the inland areas. Both national and provincial intangible heritages share core areas in the urban centers of Fuzhou, Xiamen, Zhangzhou, and Quanzhou, and are characterized by a common core belt along the eastern coast. However, compared to national-level ICH, the provincial-level heritage items are more numerous and exhibit a broader geographic distribution, with more pronounced distribution characteristics (Figure 2). By comparing the overall kernel density of provincial-level ICH in Fujian with topographic maps, one can observe a close correlation between the distribution of heritage sites and the topography and water systems. Heritage sites are more densely distributed in flat terrains and areas with slight undulations, especially along the Min River, Jianxi, Shaxi, and Futunxi areas. The trend in the kernel density paths of heritage sites closely mirrors that of the rivers and their buffer zones (Figure 2). Furthermore, the river basins of Ningde’s Jiaoxi and Huotongxi, Quanzhou’s Jinjiang, Zhangzhou’s Jiulong River, and Longyan’s Ting River are all significant areas for the distribution of ICH.

3.2. Structural Characteristics of ICH in Fujian Province

In accordance with Document No. 19 issued by the State Council in 2008, this study classifies intangible cultural heritage (ICH) in Fujian Province into ten major categories: folk literature (Y1), traditional drama (Y5), folk vocal art (Y10), traditional medicine (Y6), folklore (Y2), traditional music (Y7), traditional dance (Y4), traditional fine arts (Y2), traditional craftsmanship (Y1), and traditional sports, recreational activities, and acrobatics (Y3). The distribution of ICH items across these categories is markedly imbalanced. Among the national-level ICH items in Fujian, traditional craftsmanship and folklore constitute the largest shares, while folk literature and traditional medicine are the least represented. A similar pattern is observed at the provincial level, where traditional craftsmanship and folklore dominate, whereas folk literature and folk vocal art are underrepresented (see

Table 4. Composition of ICH in Fujian Province.

	Traditional Crafts (Y1)	Traditional Fine Arts (Y2)	Traditional Sports, Recreational Activities and Acrobatics (Y3)	Traditional Dance (Y4)	Traditional Drama (Y5)	Traditional Medicine (Y6)	Traditional Music (Y7)	Folk Literature (Y8)	Folklore (Y9)	Folk Vocal Art (Y10)	Total (Y)
National-level	34	17	7	8	23	4	9	3	29	6	140
Provincial-level	279	49	46	37	59	41	54	29	156	17	767

).

Overall, the typological structure of ICH in Fujian is heavily skewed toward traditional craftsmanship and folklore, with comparatively fewer items in folk literature, folk vocal art, and traditional medicine. Traditional craftsmanship often gives rise to a wide range of culturally embedded everyday artifacts, while folklore is deeply intertwined with the dynamics of daily life and enriched by multicultural influences, contributing to their numerical dominance in the ICH inventory. In contrast, categories such as folk literature, folk vocal art, and traditional medicine tend to involve more complex bodies of knowledge and technical practices, which raise the threshold for effective transmission.

Notably, certain items within folk literature and folk vocal art have lost their relevance to contemporary public life. For instance, heritage forms like ballads and poetic anthologies are no longer integral to everyday entertainment and have not successfully transitioned into the digital or experience-based economies. This detachment from both tangible and virtual markets hinders their visibility and recognition in the heritage selection process, leading to their relatively limited representation. Similarly, while a few traditional medicine items have been recognized at the national level—primarily those related to acupuncture and bone-setting techniques—other culturally significant practices with strong local endorsement, such as medicinal teas, therapeutic diets, and Daoist health exercises, remain underrepresented.

In terms of structural composition, national-level recognition tends to favor categories with higher aesthetic and symbolic value. For example, traditional drama (16.6%) and traditional fine arts (13.1%) constitute significantly higher proportions of national-level ICH compared to provincial-level (7.9% and 6.5%, respectively). Conversely, traditional craftsmanship accounts for a much higher proportion at the provincial level (36.3%) than at the national level (22.8%). This suggests that national-level recognition prioritizes heritage items with greater representational value and public visibility, while provincial-level recognition more comprehensively encompasses techniques and practices embedded in everyday production and life, reflecting the broader cultural foundation of local communities (see Table 4).

3.3. The Spatial Distribution Characteristics of ICH Across Ten Categories in Fujian Province

Fujian Province hosts a total of 140 national-level and 767 provincial-level intangible cultural heritage (ICH) items. Using ArcGIS, this study calculated the overall average nearest neighbor index for national-level ICH to be approximately 0.54, and for provincial-level ICH to be about 0.30, indicating a statistically significant clustering tendency across all heritage items. However, variations in clustering intensity are evident across different levels and categories of ICH. At the national level, categories such as traditional drama and folklore exhibit pronounced clustered distributions. In contrast, folk vocal art, traditional sports, traditional fine arts, and traditional craftsmanship display more random distribution patterns. Meanwhile, traditional dance, traditional medicine, traditional music, and folk literature tend to be more spatially dispersed. At the provincial level, by comparison, most categories exhibit significantly clustered patterns, with the exception of traditional dance, which also shows clustering but with less intensity, and folk vocal art, which remains randomly distributed. These results reflect the differences in spatial organization and distribution logic between national and provincial ICH systems (see

Table 5. Analysis of Average Nearest Neighbor Index.

		Y	Y1	Y2	Y3	Y4	Y5	Y6	Y7	Y8	Y9	Y10
National	Number/Unit	140	34	17	7	8	23	4	9	3	29	6
	Average Observation Distance	9033.21	25,590.16	25,952.24	41,381.03	80,323.55	24,014.39	63,197.51	64,469.09	149,986.50	26,287.00	48,904.97
	Expected Neighbor Distance	16,828.89	29,459.88	31,876.51	36,189.56	46,910.16	34,970.74	41,967.67	47,171.92	28,367.10	31,717.06	47,649.65
	Nearest Neighbor Ratio	0.54	0.87	0.81	1.14	1.71	0.69	1.51	1.37	5.29	0.83	1.03
	Confidence Factor (Z)	−10.71	−1.44	−1.59	0.73	3.85	−2.94	2.37	2.10	4.21	−1.76	0.13
	Confidence Factor (p)	0.00	0.15	0.11	0.47	0.00	0.00	0.02	0.04	0.00	0.08	0.89
	Distribution Type	Significantly Clustered	Random	Random	Random	Dispersed	Clustered	Dispersed	Dispersed	Dispersed	Clustered	Random
Provincial	Number/Unit	767	279	49	46	37	59	41	54	29	156	17
	Average Observation Distance	2412.69	5393.89	19,772.10	17,977.78	26,633.91	18,807.36	21,495.67	20,490.69	18,151.18	10,748.47	46,901.44
	Expected Neighbor Distance	8155.49	13,453.92	27,253.95	24,639.46	32,303.36	28,355.68	29,531.69	26,617.84	29,932.62	17,022.76	42,719.84
	Nearest Neighbor Ratio	0.30	0.40	0.73	0.73	0.82	0.66	0.73	0.77	0.61	0.63	1.10
	Confidence Factor (Z)	−37.31	−19.14	−3.86	−3.51	−2.04	−4.95	−3.33	−3.24	−4.05	−8.81	0.77
	Confidence Factor (p)	0.00	0.00	0.00	0.00	0.04	0.00	0.00	0.00	0.00	0.00	0.44
	Distribution Type	Significantly Clustered	Significantly Clustered	Significantly Clustered	Significantly Clustered	Clustered	Significantly Clustered	Significantly Clustered	Significantly Clustered	Significantly Clustered	Significantly Clustered	Random

Y—Overall ICH, Y1—Traditional Crafts, Y2—Traditional Fine Arts, Y3—Traditional Sports, Recreational Activities and Acrobatics, Y4—Traditional Dance, Y5—Traditional Drama, Y6—Traditional Medicine, Y7—Traditional Music, Y8—Folk Literature, Y9—Folklore, Y10—Folk Vocal Art.

).

Secondly, although the overall distribution of ICH is concentrated in the eastern coastal regions of Fujian, the distribution patterns and spatial extents do not fully overlap. Across both national and provincial levels, traditional craftsmanship and traditional drama exhibit relatively balanced spatial distributions, particularly at the provincial level, where such items are present in almost all municipalities. Traditional music and traditional fine arts tend to follow a linear belt-shaped pattern, most prominently concentrated in the Putian–Quanzhou corridor. In contrast, traditional medicine, along with traditional sports, recreational activities, and acrobatics, demonstrates a point-clustered distribution, primarily located in the cities of Fuzhou and Quanzhou. On the other hand, at the provincial level, categories such as folklore, traditional dance, folk literature, and folk vocal art display a wider geographic spread, whereas at the national level, these same categories are more likely to form compact, clustered aggregations. As illustrated in Figure 2, national-level ICH designation shows a clear bias toward coastal areas, with inland regions being relatively underrepresented. Some inland zones even appear as spatial blanks in terms of recognized ICH items, further reflecting the uneven spatial allocation and regional skewness in heritage recognition.

Notably, the same ICH category may exhibit contrasting spatial patterns across different administrative levels. For instance, traditional craftsmanship is randomly distributed at the national level but shifts to significant clustering at the provincial level; the same is observed for traditional fine arts. These differences indicate that provincial-level clustering reflects the formation of localized cultural ecologies, often resulting in belt-like or zonal concentrations. In contrast, the random or dispersed distribution at the national level suggests a more selective recognition process that favors a limited number of highly representative or landmark cultural nodes (see Table 5).

3.4. Temporal Evolution and Distribution Patterns of ICH Across Different Batches

The provincial-level batches of intangible cultural heritage (ICH) announced in 2005, 2007, 2009, 2011, 2017, 2019, and 2022 were analyzed using the Average Nearest Neighbor (ANN) method in ArcGIS. As shown in Table 5, all seven batches exhibit a clustered spatial distribution, though the degree of clustering varies slightly across batches. The first batch displays the lowest ANN ratio, while the fifth batch shows the highest, suggesting a gradual spatial diffusion of newly recognized items over time. In terms of quantity, the number of recognized ICH items follows the order of 2022 > 2005 > 2007 > 2011 > 2019 > 2017, indicating an expansion of recognition coverage. The ANN ratios range from smallest to largest in the years 2005, 2009, 2007, 2022, 2019, 2011, and 2017, with the 2017 batch reaching an ANN value of 0.858, further confirming a strong clustering tendency in Fujian’s ICH distribution (Figure 3 and

Table 6. Analysis of Average Nearest Neighbor Index for ICH at Different Years.

	Provincial ICH	First Batch	Second Batch	Third Batch	Fourth Batch	Fifth Batch	Sixth Batch	Seventh Batch
Year	Total	2005	2007	2009	2011	2017	2019	2022
Number/Unit	767	134	129	92	82	67	74	189
Average Observation Distance	2412.688	8032.284	9960.954	108,81.665	16,474.474	21,925.854	15,189.047	8593.154
Expected Neighbor Distance	8155.487	18,113.501	17,877.695	21,554.848	22,866.683	25,559.66	23,137.281	16,282.425
Nearest Neighbor Ratio	0.296	0.443	0.557	0.505	0.72	0.858	0.656	0.528
Confidence Factor (Z)	−37.308	−12.325	−9.622	−9.086	−4.843	−2.226	−5.653	−12.42
Confidence Factor (p)	0	0	0	0	0	0.026	0	0

).

It is important to clarify that this temporal evolution does not represent intrinsic changes in the ICH itself—whose cultural content and traditions are relatively stable—but rather reflects the institutional progression of ICH recognition and policy implementation in Fujian Province. Each “batch” corresponds to an official round of government approval, registration, and resource allocation, thereby indicating the province’s evolving administrative capacity and prioritization in heritage protection.

Spatially, national-level ICH remains concentrated along the eastern coastal regions, with its main centers of recognition shifting from southern Fujian to Putian and subsequently to Fuzhou, reflecting the dynamic adjustment of heritage management focus rather than the movement of the heritage itself. In contrast, inland areas have transitioned from sporadic recognition to increasingly clustered distributions, with Ningde showing notable growth in more recent batches. At the provincial level, ICH exhibits a broader spatial spread across the eastern coastal belt, progressively extending toward non-coastal and inland regions. This gradual expansion indicates that local governments are actively pursuing more balanced spatial recognition among cities, optimizing policy coverage and cultural resource allocation across the province (Figure 3).

4. Univariate Driving Forces Influencing the Spatial Distribution of Intangible Cultural Heritage

Building upon previous research, this study refines and categorizes four major dimensions—natural environment, transportation accessibility, socio-economic development, and cultural industry—into 21 specific driving factors (

Table 7. q-Statistics of Explanatory Power for Influencing Factors on the Spatial Distribution of Intangible Cultural Heritage (ICH).

	Dependent Variable	Geographical Detector Results of Intangible Cultural Heritage in Fujian Province
Independent Variable		Y		Y1		Y2		Y3		Y4		Y5		Y6		Y7		Y8		Y9		Y10
Natural Factors	X1	0.27	0.29	0.21	0.16	0.31	0.28	0.24	0.11	0.18	0.06	0.17	0.18	0.20	0.19	0.24	0.16	0.23	0.22	0.21	0.26	0.27	0.26
	X2	0.05	0.04	0.03	0.01	0.05	0.04	0.03	0.01	0.03	0.01	0.02	0.02	0.03	0.02	0.03	0.03	0.03	0.02	0.03	0.04	0.03	0.04
	X3	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
	X4	0.28	0.31	0.25	0.25	0.21	0.17	0.37	0.29	0.15	0.10	0.08	0.13	0.34	0.21	0.25	0.24	0.12	0.11	0.21	0.20	0.05	0.13
Transportation Location Factors	X5	0.42	0.44	0.31	0.26	0.34	0.29	0.44	0.30	0.23	0.08	0.18	0.19	0.39	0.34	0.31	0.28	0.21	0.22	0.32	0.37	0.17	0.31
	X6	0.21	0.18	0.20	0.15	0.23	0.15	0.16	0.08	0.11	0.08	0.11	0.08	0.14	0.11	0.16	0.14	0.12	0.14	0.16	0.14	0.09	0.17
	X7	0.12	0.25	0.13	0.34	0.21	0.36	0.28	0.33	0.17	0.14	0.13	0.15	0.25	0.46	0.29	0.21	0.14	0.34	0.19	0.30	0.26	0.43
Economic and Social Factors	X8	0.28	0.44	0.26	0.38	0.46	0.48	0.36	0.25	0.31	0.20	0.18	0.33	0.32	0.28	0.43	0.45	0.27	0.35	0.33	0.42	0.18	0.17
	X9	0.09	0.13	0.08	0.10	0.11	0.10	0.11	0.09	0.08	0.02	0.14	0.15	0.11	0.15	0.06	0.07	0.07	0.07	0.12	0.11	0.17	0.19
	X10	0.26	0.43	0.23	0.35	0.37	0.43	0.36	0.22	0.33	0.21	0.18	0.30	0.31	0.26	0.40	0.37	0.31	0.35	0.32	0.44	0.20	0.19
	X11	0.21	0.35	0.19	0.25	0.27	0.28	0.28	0.20	0.19	0.21	0.13	0.24	0.27	0.25	0.38	0.34	0.18	0.21	0.23	0.30	0.16	0.18
	X12	0.05	0.06	0.04	0.03	0.05	0.03	0.08	0.06	0.15	0.16	0.08	0.06	0.09	0.07	0.11	0.08	0.03	0.02	0.05	0.05	0.02	0.03
	X13	0.18	0.24	0.20	0.29	0.42	0.42	0.28	0.30	0.25	0.21	0.04	0.11	0.21	0.29	0.40	0.48	0.12	0.25	0.27	0.32	0.12	0.28
	X14	0.43	0.58	0.36	0.39	0.60	0.54	0.46	0.24	0.39	0.19	0.24	0.34	0.37	0.31	0.48	0.47	0.42	0.42	0.45	0.53	0.25	0.30
	X15	0.10	0.13	0.11	0.17	0.08	0.06	0.07	0.07	0.16	0.10	0.03	0.06	0.11	0.06	0.11	0.07	0.07	0.15	0.07	0.08	0.10	0.05
	X16	0.21	0.35	0.18	0.24	0.30	0.30	0.32	0.30	0.26	0.12	0.22	0.23	0.31	0.23	0.31	0.34	0.21	0.27	0.25	0.33	0.12	0.14
Cultural Factors	X17	0.13	0.25	0.14	0.25	0.33	0.47	0.24	0.32	0.18	0.20	0.10	0.14	0.23	0.40	0.37	0.45	0.15	0.32	0.17	0.22	0.17	0.23
	X18	0.17	0.34	0.18	0.44	0.26	0.40	0.32	0.33	0.16	0.12	0.18	0.24	0.26	0.42	0.27	0.30	0.20	0.46	0.27	0.38	0.15	0.24
	X19	0.17	0.33	0.17	0.43	0.25	0.39	0.31	0.32	0.13	0.13	0.18	0.24	0.25	0.43	0.27	0.30	0.18	0.41	0.28	0.36	0.25	0.35
	X20	0.49	0.43	0.44	0.28	0.47	0.36	0.47	0.41	0.29	0.09	0.10	0.16	0.44	0.24	0.42	0.33	0.24	0.19	0.39	0.32	0.08	0.15
	X21	0.43	0.48	0.35	0.30	0.40	0.38	0.38	0.20	0.22	0.13	0.23	0.27	0.37	0.22	0.44	0.34	0.29	0.20	0.33	0.31	0.19	0.21
Total		4.56	6.04	4.08	5.08	5.73	5.96	5.61	4.45	3.99	2.56	2.73	3.62	5.03	4.95	5.73	5.47	3.58	4.71	4.65	5.49	3.01	4.02
Provincial Traditional Intangible Cultural Heritage						numerical range			0.00														0.60
National Traditional Intangible Heritage						numerical range			0.00														0.58

Note: X1-Elevation, X2-Slope, X3-Slope Direction, X4-Road Density, X5-River Network Density, X6-Total Passengers, X7-Length of Highways in Operation, X8-Gross Domestic Product, X9-Primary Industry, X10-Secondary Industry, X11-Tertiary Industry, X12-Per Capita GDP, X13-Annual Per Capita Disposable Income, X14-Population Density, X15-Lever of Township, X16-Budgetary Expenditures of Local Government, X17-Per Capita Expenditure of Rural Households on Education, Culture and Recreation, X18-Per Capita Expenditure of Urban Households on Education, Culture and Recreation, X19-Total Per Capita Expenditure on Education, Culture, and Recreation X20-A-grade Tourist Attractions, X21-National tangible Cultural Heritage. Y-Overall ICH, Y1-Traditional Crafts, Y2-Traditional Fine Arts, Y3-Traditional Sports, Recreational Activities and Acrobatics, Y4-Traditional Dance, Y5-Traditional Drama, Y6-Traditional Medicine, Y7-Traditional Y8-Folk Literature, Y9-Folklore, Y 10-Folk Vocal Art.

), in order to systematically examine their influence on the recognition and spatial distribution of intangible cultural heritage (ICH). First, each of the 21 factors was processed using the Natural Breaks classification method with seven levels via ArcGIS (Figure 4), and the entire Fujian Province was divided into a 6 km × 6 km grid, resulting in 3396 grid cells.

In particular, the 6 km resolution strikes a robust balance between spatial accuracy and computational efficiency. On the one hand, it enables stable detection of ICH clusters along cultural nodes and corridors while minimizing fragmentation and random noise caused by overly fine grids. On the other hand, it aligns with the functional spatial scope of heritage transmission ecosystems centered around protection units—capturing frequent activities such as apprenticeship, practice, production, and market interactions—thus serving as a reasonable proxy for practitioners’ daily mobility range. Moreover, in the context of large-sample spatial analysis, the 6 km grid effectively controls computational load and mitigates the influence of local anomalies on overall pattern recognition. Although multi-scale sensitivity tests may be explored in future work, the current grid scale sufficiently supports this study’s core spatial conclusions.

Next, the kernel density results for all 21 driving factors were rasterized as independent variables (X), while the kernel density of ICH served as the dependent variable (Y). Subsequently, geographic detector analysis was applied to evaluate the explanatory power of each single factor on ICH distribution across the province, yielding statistically valid q-values (Table 7) and p-values (

Table A1. Significance Test Results (p-Values) of the Same Factors in Relation to Figure 6.

Data Type	Code	Y		Y1		Y2		Y3		Y4		Y5		Y6		Y7		Y8		Y9		Y10
Natural Factors	X1	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X2	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X3	0.172	0.633	0.493	0.352	0.561	0.793	0.177	0.183	0.423	0.573	0.632	0.211	0.522	0.585	0.188	0.814	0.775	0.692	0.301	0.745	0.716	0.367
Transportation Location Factors	X4	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X5	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X6	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X7	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
Socio Economic Factors	X8	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X9	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X10	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X11	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X12	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X13	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X14	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X15	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X16	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
Cultural Industry Factors	X17	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X18	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X19	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X20	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
	X21	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000

Provincial-level; National-level; p-values > 0.005.

).

4.1. Natural Factors

The spatial distribution of both national- and provincial-level intangible cultural heritage (ICH) in Fujian is significantly influenced by elevation and hydrographic conditions, while slope and aspect exert the least impact among all examined factors. Given the province’s complex topography and dense river networks, areas characterized by flat terrain and proximity to water bodies have historically been prioritized as human settlements. These geographical features constitute fundamental material conditions for the emergence of ICH resources. In particular, the southeastern coastal region of Fujian combines favorable geographic advantages—such as low elevation, nearness to the ocean, and abundance of rivers—which create an optimal natural environment for the concentrated coexistence of diverse ICH forms. For instance, the region’s robust maritime production and exchange activities, shaped by geographic conditions, have laid the foundation for the formation and flourishing of ICH traditions such as Mazu worship and cultural practices influenced by Southeast Asian exchanges.

4.2. Transportation and Location Factors

Transportation-related factors primarily include road density, total highway mileage, and total passenger volume of highway transportation. Among them, road density emerges as the most influential factor within the transportation category, significantly impacting the distribution of both national- and provincial-level ICH. The core corridor of dense road networks lies along the eastern coastal region, which closely aligns with the major areas of ICH concentration. Transportation factors exert particularly strong influence on Quyi (folk vocal art) and traditional medicine. As a mobile performance art, Quyi typically relies on robust transportation infrastructure to facilitate regional circulation, while the enhanced human mobility enabled by convenient transportation networks also facilitates the dissemination of traditional medicine. Notably, the total passenger volume of highway transportation has a greater impact on provincial-level traditional medicine, which is likely attributable to its wider geographic dispersion—allowing it to reach and benefit more remote and less accessible populations.

4.3. Socio-Economic Factors

From the perspective of univariate driving forces, population density stands out as the most critical socioeconomic factor influencing the distribution of various ICH categories, followed by gross domestic product (GDP) and the value added by the secondary and tertiary industries. On the one hand, the transmission of ICH often relies on oral teaching and apprenticeship within small communities, which implies that only those who are geographically proximate to inheritors and willing to carry on the legacy can become effective carriers of ICH continuity. Therefore, the demographic scale of a given area—specifically whether it is sufficient to generate potential successors—serves as a fundamental condition for sustaining and revitalizing ICH practices. In this regard, the densely populated eastern coastal cities of Fujian, such as Fuzhou, Xiamen, Zhangzhou, Quanzhou, and Putian, provide a favorable demographic base for the flourishing of various ICH items.

On the other hand, the secondary and tertiary industries exert more significant impacts on ICH recognition and protection compared to the primary sector. While many ICH forms historically emerged from agrarian practices, the dynamics of modern transmission and institutional recognition are increasingly shaped by urban manufacturing and service industries. This shift suggests that the evolving socio-economic context plays a pivotal role in reconfiguring the relevance and support for ICH in contemporary settings. Spatial distribution patterns further confirm that certified ICH resources are concentrated in coastal cities such as Quanzhou, Fuzhou, Zhangzhou, and Xiamen. These urban centers possess a robust economic foundation and exhibit higher demand for cultural and spiritual life, which, in turn, strengthens their capacity to invest in and sustain ICH development. Moreover, Fuzhou, as the political, economic, and cultural capital of Fujian Province, benefits from preferential policy support that facilitates ICH application and protection. Likewise, the economic leadership of Xiamen, Zhangzhou, and Quanzhou drives cultural consumption and enhances the institutional mechanisms for identifying and protecting ICH practices.

In terms of specific ICH categories, socioeconomic factors demonstrate not only strong overall influence but also notable disparities between the national and provincial levels of heritage recognition. Among the four categories of driving factors examined, socioeconomic variables generally exhibit the highest q-values in relation to the overall spatial distribution of ICH, indicating their dominant role. Notably, traditional music shows a consistently strong spatial correlation with nearly all socioeconomic indicators at both national and provincial levels, suggesting a robust and stable dependency on socioeconomic contexts for its development and preservation. By contrast, for traditional craftsmanship and traditional drama, the explanatory power of socioeconomic variables is more pronounced at the national level than at the provincial level, highlighting the tendency of nationally recognized projects in these categories to align with regions of greater economic development. Conversely, for traditional fine arts, traditional medicine, and traditional sports, recreational activities, and acrobatics, the influence of socioeconomic factors appears to be stronger on the distribution of provincially recognized items, suggesting that these types of ICH may be more reliant on localized economic conditions and support mechanisms at the subnational level.

The findings in Section 3.3 highlight the central role of socio-economic conditions in shaping the spatial distribution and viability of intangible cultural heritage (ICH). High population density, urban-based industries, and regional economic strength create favorable environments for ICH transmission by ensuring the presence of potential inheritors and an active cultural market. This indicates that policies aimed at ICH protection should not operate in isolation but instead align with broader demographic and industrial planning. The differentiated influence of socio-economic factors across ICH categories—e.g., the strong correlation of traditional music with urban indicators, or the distinct provincial-level support needed for traditional medicine—also suggests that category-specific strategies are essential. More broadly, the results imply that sustainable ICH protection requires embedding cultural practices within evolving urban and industrial contexts, where economic vitality can both enable recognition and foster long-term resilience. These insights offer wider relevance for other regions undergoing demographic shifts or economic transitions, underscoring the importance of adaptive, context-sensitive approaches in heritage governance.

4.4. Cultural Factors

Among the cultural factors, the variable representing designated sites of tangible cultural heritage demonstrates a relatively high q-value, indicating a significant influence on the spatial distribution of intangible cultural heritage (ICH). In many cases, these tangible heritage sites—or the structures themselves—serve as key venues for the generation, transmission, and dynamic display of ICH. Additionally, A-level tourist attractions exhibit even stronger explanatory power for the distribution of provincial-level ICH than both tangible heritage sites and topographic elevation factors. Notably, their influence is more pronounced at the provincial level than at the national level, suggesting that the development of the tourism sector in Fujian plays a critical role in the recognition and protection of provincial-level ICH. On the other hand, per capita urban and rural household expenditure on education, culture, and entertainment shows higher q-values in relation to national-level ICH distribution compared to provincial-level cases. This pattern implies that national-level ICH tends to comprise projects with higher visibility and greater supply-side costs—such as performances, exhibitions, competitions, or IP development—making their spatial distribution more sensitive to the chain of “willingness to pay—capacity to supply—platform accessibility.” In contrast, provincial-level ICH is more compatible with widely distributed, locally embedded forms such as everyday cultural practices and traditional craftsmanship. Moreover, compared with rural residents, the cultural and educational expenditures of urban residents generally exert a stronger influence on the distribution of ICH. While conventional socio-economic indicators such as urbanization rate and per capita GDP did not exhibit strong explanatory power in the single-factor analysis, closely related cultural indicators—particularly the proportion of tertiary industry and household cultural expenditure—emerged as significant drivers of ICH recognition and spatial differentiation.

In terms of specific heritage categories, the spatial distribution of traditional medicine-related ICH in Fujian is particularly influenced by cultural factors. These heritage items are primarily concentrated in the cities of Fuzhou, Xiamen, Quanzhou, and Zhangzhou, showing significant spatial clustering. These urban areas are characterized by dense populations and advanced economic development, where public demand and purchasing capacity for traditional medicine-related ICH are relatively strong. At the national level, traditional medicine heritage is more likely to be distributed in economically developed regions due to favorable financial conditions and supportive local policies. Moreover, such heritage items typically possess strong reputational value, further enhancing their potential for development and recognition. In contrast, provincial-level traditional medicine ICH exhibits a more dispersed spatial pattern and is less influenced by factors such as household expenditure on education, culture, and entertainment. This broader distribution enables provincial-level heritage to cover less developed regions, thereby addressing the needs of lower-income groups. The availability of raw medicinal materials in rural areas—where they are more easily accessible and affordable—also contributes to the survival and transmission of such heritage. Moreover, provincial projects often receive targeted local support, as evidenced in the recognition of traditional medical knowledge and practices among the She, Hakka, and Min ethnic groups. These characteristics highlight the important role of provincial-level recognition in enhancing spatial equity within the traditional medicine heritage protection system. Additionally, unlike the relatively limited influence of household cultural expenditure, A-level scenic areas and designated tangible heritage sites exert a significantly stronger influence on the distribution of provincial-level traditional medicine ICH. This indicates that the protection of material cultural heritage and the development of regional tourism resources play an increasingly important role in facilitating the recognition and protection of local traditional medicine heritage.

This study underscores that the protection of intangible cultural heritage (ICH) is most effective when embedded within broader cultural ecologies that include tangible heritage sites, tourism infrastructure, and patterns of cultural consumption. The significant influence of scenic areas and tangible heritage sites on provincial-level ICH distribution suggests that integrating ICH into place-based development strategies—such as cultural tourism, heritage zones, and community museums—can amplify both its visibility and vitality. At the same time, the stronger correlation between national-level ICH and household cultural expenditures highlights the importance of demand-side factors in sustaining more resource-intensive heritage forms. These findings point to the need for differentiated policy frameworks: while high-profile ICH projects may require targeted investment and platform-building in economically developed urban centers, locally embedded forms of ICH—such as traditional medicine, craftsmanship, and rural rituals—benefit from dispersed, equity-oriented protection mechanisms. The case of traditional medicine in Fujian illustrates how cultural identity, ethnic diversity, and resource accessibility shape ICH viability, calling for adaptive governance models that are sensitive to regional socio-cultural conditions. More broadly, these insights offer valuable guidance for global heritage systems seeking to reconcile cultural significance with spatial justice, by promoting inclusive, context-specific strategies for protecting ICH across both urban and rural landscapes.

5. Interactive Effects of Paired Factors on the Spatial Distribution of ICH

Through a systematic analysis of Bivariate interactions, this study identifies two key driving mechanisms underlying the spatial distribution of ICH: (1) universally adaptive common enhancement factors, and (2) structurally dominant elements reflecting hierarchical differentiation and spatial strategic orientation. First, five “foundational–developmental” variables—slope (X2), aspect (X3), gross domestic product of the primary industry (X9), per capita GDP (X12), length of highways in operation (X6) and urbanization level (X15)—exhibit consistently enhanced explanatory power when paired with most other environmental factors, predominantly demonstrating nonlinear enhancement effects. This synergy is not driven by any single natural or socioeconomic variable; rather, it emerges from the systematic amplification of explanatory effects through the interactions between these structural base factors and a variety of dimensions, including transportation accessibility, industrial composition, population distribution, cultural expenditure, and tourism resources. Together, they form a multifactor collaborative system characterized by both structural stability and wide adaptability. This finding not only provides a unified theoretical perspective for understanding the structural logic of overall ICH distribution patterns but also lays the groundwork for assessing the robustness and generalizability of spatial interaction mechanisms (Figure 5 and Figure 6).

Second, in the spatial distribution of provincial-level ICH, both A-level scenic spots (X20) and designated tangible cultural heritage protection units (X21) demonstrate consistently stronger interactive enhancement effects when paired with other variables, exhibiting higher degrees of variable dominance and spatial coordination. In contrast, these two factors do not show the same level of systemic amplification in the distribution mechanism of national-level ICH. This divergence suggests that provincial-level ICH relies more heavily on spatial anchor points such as cultural tourism zones and heritage protection platforms. Through the coordinated configuration with factors like transportation accessibility, population density, service infrastructure, and cultural consumption, a “platform–support” spatial structure emerges—designed to activate heritage resources across broader regions and enhance their exposure and accessibility. In comparison, national-level ICH tends to cluster in areas with superior resource endowments and stronger cultural symbolism, reflecting a structurally selective aggregation pattern shaped by higher thresholds of recognition and stronger dependence on policy prioritization. This layered identification of Bivariate interaction mechanisms not only elucidates the differentiated drivers between hierarchical levels of ICH but also provides theoretical support for more targeted policy design and differentiated intervention strategies.

Building upon this, further analysis reveals that road mileage (X5) and population density (X14) demonstrate strong two-factor interactive driving effects in the overall distribution of both national and provincial-level ICH, underscoring the fundamental importance of “locational accessibility” and “population agglomeration” as baseline spatial conditions. However, a more disaggregated category-wise analysis shows that such interactive enhancement effects are particularly pronounced in traditional fine arts (Y2) and folk customs (Y9). Notably, this effect does not stem from the dominant influence of X5 or X14 in isolation; rather, it reflects the synergistic amplification of explanatory power when these variables interact with others such as industrial structure, cultural expenditure, and tourism facilities. Traditional fine arts often follow a “production–exhibition–commerce” dissemination model, which requires high levels of market connectivity and visitor access. In contrast, folk customs are deeply embedded in the everyday lives of densely populated urban and rural communities, relying on collective participation and localized cultural ecosystems. As a result, both types of ICH exhibit heightened sensitivity to specific variable interactions within a “spatial support–social carrier” network, leading to more prominent spatial clustering under multi-factorial driving forces. This finding deepens our understanding of how different types of ICH respond to environmental variables and offers a theoretical foundation for developing type-specific and spatially differentiated intervention strategies.

The identification of interaction effects between two factors offers critical insights for designing integrated, context-sensitive heritage protection strategies. The revealed synergy among foundational–developmental variables (e.g., GDP, slope, urbanization) suggests that ICH protection should not rely on isolated cultural inputs alone but rather be embedded within a stable, multi-dimensional development framework. This implies that heritage policies can achieve greater impact when coordinated with urban planning, infrastructure development, and public service provision. At the same time, the dominance of scenic areas and tangible heritage sites in shaping provincial-level ICH distribution highlights the value of spatial anchoring through tourism and material heritage platforms. Policymakers should therefore foster “platform-based” models—such as cultural corridors or integrated tourism–heritage zones—that link intangible heritage with high-accessibility hubs. Moreover, the strong interaction effects between road networks/population density and other influencing factors call for strategic investments in transportation and service accessibility to expand the reach and relevance of ICH in both urban and rural settings. Type-specific strategies are also essential: for instance, traditional fine arts benefit from market connectivity, while folk customs thrive in dense, participatory communities. These findings advocate for a differentiated approach to heritage protection—tailored to both spatial configurations and cultural typologies—thereby supporting more inclusive, resilient, and sustainable cultural development pathways.

6. Discussion and Conclusions

The spatial distribution of ICH in Fujian Province exhibits distinct regional characteristics, reflecting not only the deep influence of geographical conditions on cultural transmission but also the spatial divergence in heritage protection between coastal and inland areas. The coastal zones—particularly the urban clusters of Fuzhou, Quanzhou, and Xiamen-Zhangzhou—serve as major concentration areas of ICH resources, shaped by advantageous transportation networks, robust economic development, and dense populations. This has resulted in a spatial pattern characterized by “stronger in the south than in the north, and denser in the east than in the west.” In contrast, the inland areas—constrained by terrain and relatively limited economic development—present more scattered and fragmented clusters of ICH, with smaller-scale and locally distinct distributions.

Moreover, the spatial morphology of different ICH categories—such as linear belts, clustered groups, or isolated points—further illustrates the geographically embedded advantages and constraints associated with specific cultural expressions. In terms of recognition, coastal regions clearly dominate in both national- and provincial-level listings, whereas many ICH resources in inland regions remain under-acknowledged. However, temporal analysis reveals an emerging trend toward spatial diffusion, with protection efforts increasingly extending into interior areas, indicating a transition from a coastal-centered recognition model to a more inclusive, regionally balanced protection landscape. This spatiotemporal evolution suggests that future ICH strategies should strive to reconcile cultural diversity with geographic disparity by constructing protection mechanisms that extend from core urban areas to peripheral rural regions. In doing so, a more comprehensive and equitable protection system can be established, ensuring the continuity of regional cultural heritage across varied spatial contexts.

The theoretical and methodological contribution of this study lies in the construction of a multidimensional and multilevel analytical framework that reveals the complex interplay between socio-economic and cultural factors in shaping the spatial distribution of ICH. By integrating the analysis of both socio-economic drivers and cultural dynamics, the framework challenges the conventional static and qualitative approaches that have long dominated ICH research and instead foregrounds the dynamic, interactive nature of ICH as a socio-cultural phenomenon embedded in space. The single-factor detection enables the identification of key variables with dominant explanatory power, offering insights into the primary forces underlying heritage distribution. In contrast, the Bivariate interaction analysis uncovers mechanisms of synergistic amplification across diverse dimensions, highlighting how multiple drivers coalesce to influence ICH patterns in a nonlinear and context-dependent manner. The integration of both methods facilitates a more comprehensive and systematic understanding of spatial heritage logics. Taken together, these approaches provide a robust analytical foundation for developing more targeted and coordinated ICH conservation strategies, supporting a shift from mere factor identification to mechanism-oriented policy design and adaptive spatial governance.

From the perspective of single-factor explanatory power, socio-economic variables such as population density (X14), gross regional product (X8), and emerging industrial structures (X10, X11) are not merely external conditions for the persistence of ICH; rather, they are identified as critical drivers that directly shape the inheritance and development of ICH. These findings reveal a profound interaction between heritage dynamics, urbanization processes, and industrial transformation. Particularly in economically developed areas, ICH benefits from a solid economic foundation, which enables its dissemination, revitalization, and integration into broader development agendas—thereby forming a virtuous cycle between heritage protection and socio-economic advancement. This not only promotes the sustainable transmission of ICH but also reinforces its value as a “living fossil” within the cultural industry.

Moreover, the analysis of cultural variables—especially material cultural heritage sites (X21) and national A-level scenic areas (X20)—further enriches the theoretical discourse on ICH by revealing how the spatial presence and institutional protection of tangible heritage can act as pivotal anchors for the dynamic inheritance and innovative renewal of intangible practices. These cultural drivers, mediated through heritage policy frameworks and tourism development, highlight the synergistic linkages between material and immaterial heritage dimensions.

Beyond the single-factor effects identified above, this study further employs bivariate interaction detection within the GeoDetector framework to deepen the explanatory power and analytical precision of the ICH spatial distribution mechanisms. Unlike single-factor analysis, which isolates the independent effects of individual variables, the bivariate method captures the synergistic and amplifying effects among interacting factors, offering a more nuanced understanding of composite drivers. The results reveal that foundational development-related variables—including slope (X2), aspect (X3), primary industry output (X9), per capita GDP (X12), and urbanization level (X15)—consistently demonstrate nonlinear enhancement effects when paired with other variables such as population, industry, culture, or transport, significantly strengthening their explanatory power. Additionally, Class-A scenic sites (X20) and material heritage protection units (X21) exhibit stronger interaction effects in the provincial ICH system, highlighting the pivotal role of cultural carriers and supportive environments in facilitating ICH diffusion and enhancing local cultural vitality. This analytical framework deepens empirical insights into multi-tiered ICH spatial dynamics, with its core contribution lying in the identification of interaction effects that shift spatial explanation from static, univariate causality to dynamic, relational cognition. At the policy level, this approach provides a robust foundation for targeted decision-making by promoting spatial governance strategies centered on multi-factor coordination, thereby enabling more precise and adaptive allocation of cultural resources across regions and designation levels.

Taken together, the comparative analysis of national- and provincial-level ICH in Fujian reveals that spatial patterns of “where heritage is” are inseparable from “who has the capacity to be recognized.” National-level ICH is highly concentrated in economically advanced coastal cities such as Fuzhou, Xiamen, Quanzhou, and Zhangzhou, where stronger fiscal capacity, cultural visibility, and policy support converge to produce a structurally selective pattern of recognition. Provincial-level ICH, by contrast, is more numerous and more widely diffused, extending into inland and relatively underdeveloped areas and more closely embedded in everyday cultural ecologies, especially in traditional crafts and performing arts. The single-factor and interaction detection results further show that socio-economic indicators and cultural–tourism infrastructures shape these two tiers in different ways: national-level items are more sensitive to high cultural expenditure and symbolic platforms, whereas provincial-level items depend more on the joint configuration of scenic sites, tangible heritage units, accessibility, and local social conditions. In this sense, the multi-level spatial structure of ICH does not merely mirror underlying cultural resources but also reflects the politics of designation—uneven nomination capacity, differentiated policy attention, and territorially biased visibility. By empirically unpacking how recognition hierarchies, development disparities, and spatial mechanisms intersect, this study contributes to a more critical understanding of heritage governance as a process that can both reproduce and mitigate spatial inequality. These insights provide a conceptual and evidential basis for designing differentiated ICH protection strategies that not only safeguard emblematic national items but also strengthen provincial-level systems as tools for spatial rebalancing, supporting peripheral communities and diverse cultural identities that are otherwise at risk of remaining under-recognized.

Through this integrated analysis, this study proposes a comprehensive theoretical model that transcends the traditional dichotomy between cultural protection and development. It offers a new interpretive lens for understanding the complexity and dynamism of ICH recognition and protection in contemporary society. The interdisciplinary, multi-scalar approach deepens our knowledge of the spatial logic of ICH and provides theoretical support for its sustainable use and transmission within the broader context of urban modernization and industrial upgrading. Ultimately, the analytical framework established herein contributes a valuable reference point for future research and policy formulation on ICH conservation and revitalization.

Despite the insights offered by this study, several limitations should be acknowledged. First, the analysis relies on officially recognized ICH datasets, which may not fully capture informal or community-based heritage practices, potentially introducing selection bias. Second, while the GeoDetector model provides a robust framework for identifying spatial heterogeneity and interaction effects, it does not account for temporal dynamics or local spatial dependencies in the way that time-sensitive or distance-decay models like GTWR or GWR might allow. Third, although we utilized a 6 km grid-based approach to enhance spatial resolution, certain micro-scale variations in cultural landscapes may still be underrepresented. Therefore, future research could benefit from incorporating parametric models such as GWR or Poisson regression to further validate and complement the non-parametric findings presented here. Additionally, the integration of fieldwork, participatory mapping, or qualitative interviews could provide deeper contextual understanding of local ICH transmission mechanisms. Expanding the scope to include inter-provincial comparisons or longitudinal datasets would also enhance the generalizability and policy relevance of the findings.

In addition, future research on intangible cultural heritage (ICH) should move beyond regional constraints and pursue cross-regional comparisons and dynamic analyses of spatial mechanisms within diverse socio-economic contexts. Against the backdrop of simultaneous global urbanization and cultural restructuring, ICH should not be viewed merely as a static cultural emblem, but as a dynamic force embedded in local social networks and actively participating in regional transformation processes. Subsequent studies are encouraged to focus on the multi-scalar evolution of ICH distribution and the coordination of governance mechanisms, exploring how to achieve fairness, adaptability, and sustainability in heritage protection across different stages of development, policy frameworks, and cultural traditions. Moreover, incorporating dynamic variables such as population mobility, industrial restructuring, technological mediation, and digital dissemination into analytical frameworks may help uncover the tensions and synergies between ICH and contemporary lifestyles. By constructing integrated theoretical models and methodological tools, future research can facilitate a paradigm shift from a heritage-centric ontology to a multi-dimensional “culture–society–space” system, offering refined and insightful knowledge for sustaining cultural diversity and fostering regional resilience on a global scale.