Geoheritage Conservation Enhanced by Spatial Data Mining of Paleontological Geosites: Case Study from Liaoning Province in China

Abstract

China boasts abundant geoheritage, including numerous paleontological geosites; however, many of these geosites are currently at high risk of degradation and face considerable challenges in protection and management. Using Liaoning Province as a case study, this study employs Geographic Information Systems (GIS) and spatial analysis to conduct the systematic data mining of provincial paleontological geosites. We quantitatively examine their spatiotemporal distribution patterns, identify key natural and socio-economic factors influencing their spatial occurrence, and pinpoint areas at high risk of degradation. Results reveal that the distribution of paleontological geosites across prefectural-city, regional, and geological time scales is highly uneven, leading to significant disparities in scientific research, resource allocation, and geotourism development. Significant spatial correlations are observed between the locations of these geosites and natural parameters as well as socio-economic indicators, providing a theoretical foundation for designing targeted conservation measures and precise management strategies. Based on these findings, the study proposes a multi-scale geoheritage conservation framework for Liaoning, which systematically addresses protection strategies across three distinct dimensions: at the prefectural-level city scale, through precise basic management, systematic investigation, and differentiated protection measures; at the regional scale, by enhancing collaborative mechanisms and establishing an integrated conservation network; and at the geological time scale, by deepening value recognition and promoting forward-looking conservation initiatives. This study not only offers tailored recommendations for conserving paleontological heritage in Liaoning, but also presents a transferable research model for other regions rich in paleontological resources worldwide, thereby bridging the gap between geoheritage conservation needs and practical solutions.

1. Introduction

Geosites are key localities that exhibit significant scientific value due to their geological features, providing insights into pivotal stages of Earth’s evolution [1,2,3]. A paleontological geosite constitutes a specialized subset of geosites, comprising both direct fossilized remains of organisms and indirect evidence—such as traces of biological activity—preserved within the geological record [4,5,6]. These geosites are recognized for their scientific, educational, or touristic importance. The designation “paleontological geosite” may refer to spatially restricted outcrops, isolated geological elements with exceptional characteristics, or extensive site clusters, depending on the application scenarios of the term “geosite” [2,3]. Irrespective of scale, paleontological geosites are characterized by their unique scientific significance and, in some instances, notable fragility, underscoring their critical role in advancing geological and paleontological research [3,7,8]. In some literature, depending on usage conventions and contextual requirements, paleontological geosites are also alternatively referred to as fossil sites [8], paleontological sites [9], or fossil localities [10].

However, paleontological geosites are highly susceptible to degradation and irreversible damage due to anthropogenic and natural threats, including erosion, mining, urban development, infrastructure expansion, landfill operations, and illicit fossil extraction [11]. This vulnerability stems from the significant societal demand for paleontological resources (fossils), which are often exploited for ornamental and commercial purposes, rendering them particularly prone to overharvesting and looting [12]. So, an excessive focus on their economic value may conflict with the principles of sustainable territorial development. Moreover, an exclusive focus on the conservation of paleontological geosites may lead to overly restrictive protection measures, inadvertently neglecting their critical scientific, educational, and geotourism value, thereby distorting the original intent of preservation efforts. Consequently, a scientific conservation strategy for paleontological geosites, along with the formulation of evidence-based fossil utilization planning, is essential to ensure the long-term preservation and sustainable use of these critical paleontological resources [13,14].

China boasts an abundant and diverse fossil record, making it one of the few nations globally with the most comprehensive assemblage of fossil types [15,16]. Key biotas such as the Chengjiang Biota [17], Guanling Biota [18], Yanliao Biota, and Jehol Biota [19], which have yielded discoveries of global scientific significance, underscore China’s pivotal role in paleontological research. Over the past three decades, Chinese paleontological studies have emerged as focal points of international scholarly attention. Breakthroughs in theoretical frameworks—including insights into avian origins and early evolutionary patterns [19,20], the origin of Eutherian mammals [21], and the evolutionary dynamics of Triassic marine reptiles [18]—have generated substantial global academic resonance. These paleontological advancements not only provide critical foundational data for understanding the tectonic evolution of ancient Chinese continental blocks, paleoenvironmental transformations, and chronostratigraphic correlation across geological eras, but also inform strategic planning for petroleum and solid mineral exploration [22,23]. Furthermore, they have established a robust foundation for scientific education, fossil conservation, and the sustainable utilization of paleontological resources [15,16].

The exceptionally continuous Precambrian-to-Quaternary stratigraphic succession in China preserves a globally significant assemblage of paleontological geosites, characterized by their remarkable richness and diversity that is unparalleled in most other regions of the world [15,16]. However, influenced by complex tectonic settings and differential preservation conditions, these geosites exhibit remarkable spatial heterogeneity and temporal disparity. This situation has resulted in research deficiencies in current Chinese paleontological studies: (i) Academic attention is predominantly focused on biological or taphonomic analyses of individual fossil specimens. (ii) There are only a few cases of systematic research on their spatiotemporal distribution patterns at the geosite-cluster scale [8,24]. Moreover, with rapid economic development and intensified anthropogenic activities, many crucial paleontological geosites in China are facing imminent threats of degradation, rendering their conservation an urgent priority. Consequently, spatial data mining of China’s paleontological geosites will establish a crucial foundation for targeted conservation strategies and precision management, while offering substantial scientific significance. This initiative represents both an imperative response to contemporary geoheritage protection challenges and a strategic opportunity to advance interdisciplinary development.

In summary, there is a growing recognition of the necessity to conduct spatial data mining on China’s paleontological geosites as a critical component of geoheritage conservation efforts. Accordingly, this study focuses on Liaoning Province (Figure 1)—a region exceptionally rich in paleontological resources—to analyze the spatial distribution patterns of paleontological geosites, the factors influencing their spatial locations, and their degradation risks using GIS techniques. Furthermore, we explore potential conservation strategies and sustainable utilization approaches for these valuable geosites. By establishing an interdisciplinary link between paleontology and geospatial information science, this study aims to contribute empirical data that advances our understanding of Earth’s evolutionary history of life, supports the preservation of geological heritage, and mitigates geosite degradation. Ultimately, it seeks to enhance both academic research and practical conservation efforts related to China’s paleontological geoheritage.

2. Materials and Methods

2.1. Study Area

Liaoning Province is situated in the southern part of northeastern China, covering a land area of 148,700 square kilometers (Figure 1). The Chifeng–Kaiyuan deep fault bisects Liaoning Province into northern and southern sections [25]. The southern section, encompassing the majority of the province, belongs to the North China Block, while the northern section represents the Paleozoic depression belt along the northern margin of the North China Block [22,25,26]. The North China Block exposures in Liaoning Province exhibit a well-developed stratigraphic sequence with the notable absence of Devonian strata (Figure 2). The Archean to Paleoproterozoic basement comprises an intensely metamorphosed and deformed rock series. The Mesoproterozoic to Late Paleozoic successions exhibit stable depositional environments with abundant fossil assemblages, characterized by typical North China-type lithostratigraphic, biostratigraphic, sequence stratigraphic, and chronostratigraphic features [27]. The Mesozoic continental volcanic–sedimentary sequences are particularly well-developed in western Liaoning, where the renowned Jehol Group and its exceptional Jehol Biota were first identified [22,25,28]. In contrast, the northern Paleozoic depression belt displays sporadic outcrops of Cambrian, Ordovician, Silurian, Carboniferous, and Permian strata. The Early Paleozoic sequences represent residual marine basin deposits, including Ordovician–Silurian volcanic–sedimentary formations, while the Late Paleozoic comprises marine and paralic volcanic–sedimentary successions [25,26,27].

The province administers 14 prefecture-level cities and has a resident population of 41.55 million (Figure 1). As of 2023, a total of 252 paleontological geosites have been systematically documented in Liaoning Province [24], with their distribution spanning all 14 prefecture-level cities. The comparative regional analysis demonstrates that the recorded quantity in Liaoning Province substantially exceeds the documented paleontological geosites in other provinces in China with systematically compiled inventories (e.g., Shandong Province: 133 [29]; Hebei Province: 110 [30]; Anhui Province: 74 [31]), indicating a pronounced regional predominance in paleontological resources distribution. Liaoning Province is globally renowned for its exceptionally rich paleontological resources, particularly from the Mesozoic Yanliao Biota and Jehol Biota [19,20,21,22,23,32,33]. The region’s well-preserved fossil specimens have yielded numerous groundbreaking discoveries, including feathered dinosaurs (e.g., Microraptor zhaoianus [34] and Anchiornis huxleyi [35]), early birds (e.g., Confuciusornis sanctus [36] and Longipteryx chaoyangensis [37]), early mammals (e.g., Eomaia scansoria [21] and Juramaia sinensis [38]), and diverse early angiosperms (e.g., Archaefructus sinensis [39] and Archaefructus liaoningensis [40]). These fossils provide unparalleled insights into the co-evolution of paleontology and paleoenvironment during a critical period of Earth’s history.

2.2. Data Source and Required Software

This study utilizes seven primary data categories (

Table 1. The multi-source geographic data source in this study.

No.	Data Name	Data Source	Data Cutoff Time	Citation
1	252 Paleontological geosites of Liaoning Province	List of paleontological geosites in Liaoning Province	2023	Reference [24]
2	Provincial, municipal and county administrative divisions of Liaoning Province	Administrative Division Dataset of China	2024	Reference [41]
3	DEM of Liaoning Province	SRTMDEM UTM 90 m resolution DEM dataset	2024	Reference [42]
4	Stratigraphic outcrop map of Liaoning Province	China’s 1:1,000,000-scale digital geological map spatial database	2017	Reference [43]
5	Road networks and water systems data of Liaoning Province	OpenStreetMap website	2024	Reference [44]
6	Population density and GDP per capita of Liaoning Province	China County Statistical Yearbook	2023	Reference [45]
7	Mineral Occurrences and Geological Hazard Sites in Liaoning Province	Mineral Occurrences and Geological Hazard Sites datasets of China	2021	Reference [46]

): (1) Comprehensive documentation of 252 paleontological geosites in Liaoning Province, including geological coordinates, geological period, and conservation status, compiled through field surveys and verified public archives (Table S1) [24]. (2) Geographic data for administrative divisions were obtained from the open-access website of China Temporal Sequence Administrative Map [41]. (3) The 90 m resolution Digital Elevation Model (DEM) was acquired from the Geospatial Data Cloud Platform [42]. (4) Stratigraphic outcrop information was extracted from China’s 1:1,000,000-scale digital geological map spatial database [43]. (5) Road networks and water systems were sourced from the OpenStreetMap website [44]. (6) The new population density and GDP per capita data were derived from the China County Statistical Yearbook [45]. (7) The Mineral Occurrences and Geological Hazard Sites dataset was provided by the Geographic Remote Sensing Ecological Network Platform (www.gisrs.cn, accessed on 11 December 2024) [46].

The aforementioned data were integrated into a GIS database of Liaoning Province, georeferenced in the WGS 84/UTM zone 49N coordinate system. Data extraction and analytical computations were performed using ArcGIS 10.8.1, QGIS 3.34.2, GeoDa 1.22, and Microsoft Excel 2021. Spatial layer integration, superposition analysis, and cartographic output were conducted primarily in QGIS 3.34.2.

2.3. Research Methods

2.3.1. Quantitative Analysis of the Distribution Patterns

• Imbalance Index

The imbalance index (S) quantifies the degree of distributional equilibrium of research subjects across different regions [8,47,48]. In this study, we employ Equation (1), derived from Lorenz curve analysis, to calculate the imbalance index of paleontological geosites in Liaoning Province.

$$S={\displaystyle \frac{{\displaystyle \sum _{i=1}^n{Y}_i-50\left(n+1\right)}}{100n-50\left(n+1\right)}},$$

(1)

where n represents the total number of paleontological geosites, and Y_i denotes the cumulative percentage of paleontological geosites across cities (taking only the numerical portion of percentages, omitting the % symbol), ranked in descending order of magnitude. The index S ranges from 0 to 1: S = 0 indicates a perfectly even distribution of paleontological geosites across all cities; S = 1 signifies complete spatial concentration, with all paleontological geosites confined to a single city; intermediate values (0 < S < 1) reflect varying degrees of distributional imbalance. Through the Lorenz curve [8,47,48], we can visually observe the equilibrium degree of the distribution of paleontological geosites in various prefecture-level cities of Liaoning Province.

2.

Clustering Analysis

Density-Based Spatial Clustering of Applications with Noise (DBSCAN) is an unsupervised machine learning algorithm that identifies clusters based on spatial density distribution [49,50]. This density-based clustering approach autonomously determines optimal clusters by adaptively adjusting two critical parameters: the minimum number of points (MinPts) and the neighborhood radius (ε) required to define a dense region. The parameter ε defines the radius of the neighborhood around a data point. When the number of points within ε exceeds the minimum threshold (MinPts), the region is considered dense.

Tuning the optimal parameters is a non-trivial task, as the combination of ε and MinPts significantly influences clustering results. The selection of initial values follows established spatial clustering conventions: (i) The MinPts must satisfy the inequality MinPts ≥ D + 1 (with D being the data dimensionality), ensuring sufficient points to define a dense region; (ii) The ε should not exceed the dataset’s average inter-point distance, pre-venting excessive cluster merging.

In this study, the DBSCAN algorithm was applied to identify spatial clusters and outliers of paleontological geoheritage sites in Liaoning Province. By leveraging the inherent spatial distribution characteristics of the data, DBSCAN effectively detects arbitrarily shaped clusters while simultaneously identifying spatial outliers, making it particularly suitable for analyzing complex paleontological geosites patterns.

3.

Average Nearest Neighbor

The average nearest neighbor ratio (R) is defined as Equations (2) and (3) [51,52].

$$R={\overline{r}}_i/r_E,$$

(2)

$$r_E=1/2\sqrt{D}=1/2\sqrt{m/A},$$

(3)

where ${\stackrel{\_}{r}}_i$ represents the mean nearest neighbor distance between paleontological geosites (treated as point elements); r_E is the expected (theoretical) nearest neighbor distance under complete spatial randomness; D represents the number of geosites per unit area; m represents the number of paleontological geosites; A represents the study area, which is generally the area of the smallest circumscribed rectangle that includes all the point elements. The R quantifies the spatial distribution pattern: R = 1 indicates a random distribution; R > 1 suggests a dispersed (uniform) distribution; R < 1 indicates a clustered (aggregated) distribution. Thus, the distribution patterns can be classified into three fundamental types: clustered, dispersed, or random.

4.

Kernel Density Estimation

Kernel density estimation is a nonparametric spatial statistical method used to analyze the density distribution of point elements within a defined neighborhood. The process involves assigning a kernel function to each data point, where the function’s influence diminishes with distance from the point. A critical parameter, the bandwidth (or search radius), determines the spatial scale of smoothing. For a given location (e.g., defined by latitude and longitude), the density value is computed by summing the contributions of all neighboring points within the bandwidth, then normalizing by the kernel’s effective area. In this study, measuring the superimposed density value of paleontological geosites surrounding each output raster yields the degree of discrete concentration of paleontological geosites, calculated as Equation (4) [53,54].

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

(4)

where h represents the bandwidth (search radius); n indicates the number of paleontological geosites whose Euclidean distance from geosite x is less than or equal to h; k[ ] is the kernel function; (x − x_i) denotes the Euclidean distance from the valuation geosite x to the geosite x_i.

The bandwidth is conceptually analogous to the bin width of a common histogram, determining the spatial extent over which each paleontological geosite exerts its influence. Larger bandwidth values produce smoother density surfaces, while smaller bandwidths capture finer-scale local density variations with greater detail [53]. In the ArcGIS 10.8.1 implementation, the initial bandwidth estimate is derived as 1/30th of the diagonal length of the dataset’s minimum bounding rectangle (automatically calculated from spatial ex-tent). The final bandwidth selection was optimized through: (i) an acceptable level of density surface smoothness; (ii) an excellent visualization clarity; (iii) geological interpretability validation.

2.3.2. Quantitative Analysis of Spatial Relationships

• Superposition Analysis

Superposition analysis is a geospatial technique that integrates two or more datasets from the same region to generate composite layers or maps with enhanced interpretability [8,51]. In this study, this method is employed to overlay the spatial distribution of paleontological geosites in Liaoning Province with multiple geographical and socioeconomic variables, including topography (elevation, slope aspect, and stratigraphic exposure) and socioeconomic factors (road proximity, population density, and Gross Domestic Product (GDP) per capita). This multi-factor superposition enables a systematic assessment of the spatial distribution patterns of paleontological geosites and their potential correlations with these variables. We further quantify these relationships through statistical analysis and graphical representations.

2.

Bivariate Spatial Autocorrelation Analysis

Bivariate spatial autocorrelation analysis enables the examination of spatial dependence between two variables while revealing their aggregated distribution patterns [55,56]. For the 100 county-level cities in Liaoning Province, we employ bivariate spatial autocorrelation analysis to investigate the spatial autocorrelation between paleontological geosites and mineral occurrences, as well as geological hazard sites. This approach enables the quantification of degradation risk among these paleontological geosites.

The bivariate global Moran’s I (Equation (5)) and bivariate local Moran’s I (Equation (6)) indices are employed to quantify these spatial associations at global and local scales, respectively.

$$I={\displaystyle \frac{n{\displaystyle \sum _{p=1}^k{\displaystyle \sum _{q=1}^k{w}_{pq}\left(x_p-\overline{x}\right)\left(y_q-\overline{y}\right)}}}{{\displaystyle \sum _{p=1}^k{\displaystyle \sum _{q=1}^k{w}_{pq}}{\displaystyle \sum _{p=1}^n{\left(x_p-\overline{x}\right)}^2}}}},$$

(5)

where k represents the number of spatial units (the county-level cities); p and q denote the location index-representing spatial units; x_p and x_q are the first variable and the second variable; $\stackrel{\_}{x}$ and $\stackrel{\_}{y}$ are the means of x and y; w_pq denotes the spatial weight value.

The bivariate Moran’s I statistic ranges from −1 to 1, with values approaching zero indicating weak or no spatial autocorrelation between the variables. Positive values (I > 0) demonstrate significant positive spatial autocorrelation (clustering of similar values), while negative values (I < 0) reflect negative spatial autocorrelation (dispersion of dissimilar values).

$$I_p=Z_p{\displaystyle \sum _{q=1}^k{w}_{pq}}{Z}_q,$$

(6)

where I_p represents the local spatial relationship between the two variables in spatial unit p; Z_p and Z_q are the standardized variances of the observed spatial units p and q.

Based on the calculated Ip values, five distinct spatial clustering patterns are identified: high-high (HH), high-low (HL), low-high (LH), low-low (LL), and non-significant clusters. The resulting Local Indicators of Spatial Association (LISA) distribution effectively characterizes both spatial clustering and local heterogeneity between the two variables.

3. Results

3.1. Spatial Distribution of Paleontological Geosites

3.1.1. Distribution Patterns at Prefecture-Level City Scale

As of 2023, a total of 252 paleontological geosites have been documented across Liaoning Province, distributed unevenly among its 14 prefecture-level cities (Figure 1;

Table 2. Summary of the imbalance index of paleontological geosites of the prefecture-level cities in Liaoning Province.

S/N	Prefecture-level City	Counts	Yi/%	Yi Cumulative Percentage/%	S
1	Chaoyang	108	42.86	42.86	0.6520
2	Jinzhou	30	11.90	54.76
3	Huludao	28	11.11	65.87
4	Dalian	23	9.13	75.00
5	Benxi	21	8.33	83.33
6	Fuxin	8	3.17	86.51
7	Tieling	8	3.17	89.68
8	Liaoyang	6	2.38	92.06
9	Fushun	5	1.98	94.05
10	Shenyang	4	1.59	95.63
11	Panjin	3	1.19	96.83
12	Yingkou	3	1.19	98.02
13	Dandong	3	1.19	99.21
14	Anshan	2	0.79	100.00

). The spatial distribution exhibits a significant difference, with Chaoyang containing the highest concentration, 108 paleontological geosites, accounting for 42.86% of the provincial total. The second to fifth most abundant cities are Jinzhou (30 geosites, 11.90%), Huludao (28 geosites, 11.11%), Dalian (23 geosites, 9.13%), and Benxi (21 geosites, 8.33%). The remaining nine prefecture-level cities exhibit notably fewer paleontological geosites, with none exceeding eight geosites (≤3.17% of the total).

To quantify the distributional imbalance of paleontological geosites across Liaoning Province, we calculated the imbalance index (S) using Equation (1) in Microsoft Excel 2021. The resulting value (S = 0.6520, Table 2) falls within the range of 0 to 1, indicating a highly uneven distribution of paleontological geosites at the prefecture-level city scale. To further visualize this disparity, we constructed a Lorenz curve by ranking the cities in descending order of paleontological geosite abundance and plotting their cumulative percentages (Figure 3). The pronounced upward convexity of the Lorenz curve—particularly its strong deviation from the line of uniform distribution—graphically confirms the extreme spatial imbalance of paleontological geosites in Liaoning Province.

3.1.2. Distribution Patterns at Regional Scale

Through DBSCAN clustering analysis, we identified well-defined spatial clusters with minimal outliers using optimal parameters in QGIS 3.34.2. Given that paleontological geosites are represented as 2D spatial data, we initially set ε = 10 km and MinPts = 3 based on the fundamental principles mentioned in Section 2.3.1. Then, we iteratively adjusted these values until achieving optimal cluster configurations characterized by: (i) well-defined cluster boundaries; (ii) minimal noise points; (iii) geologically meaningful spatial clusters. The final optimized parameters were determined to be ε = 49 km and MinPts = 5, which produced the most geologically interpretable clustering results while maintaining statistical robustness.

Spatial visualization revealed five distinct paleontological geosite clusters in Liaoning Province, with only eight outliers sparsely distributed across the province. The spatial distribution of these paleontological geosites exhibits a strong correlation with geological activity-induced geographic zonation. Accordingly, we delineated five paleontological geosite aggregation areas based on key geographic or geomorphological units (Figure 4): the Western Liaoning, Eastern Liaoning, Liaodong Peninsula, Northern Liaoning, and Northwestern Liaoning Paleontological Geosite Aggregation Areas (PGAAs).

The Western Liaoning PGAA represents the largest aggregation, comprising 169 paleontological geosites (67.06% of the provincial total). The Eastern Liaoning PGAA follows this with 36 paleontological geosites (14.29%) and the Liaodong Peninsula PGAA with 23 paleontological geosites (9.13%). The Northern Liaoning and Northwestern Liaoning PGAAs are comparatively smaller, each containing only eight paleontological geosites (3.17%, respectively). Notably, the eight non-clustered paleontological geosites, predominantly located near provincial boundaries, exhibit significantly more dispersed spatial distribution patterns compared to other geosites.

3.1.3. Distribution Patterns at Geological Time Scale

Based on the operational practices of Liaoning’s paleontological geosites administration, the paleontological geosites across the province are most appropriately classified into four major chronostratigraphic categories according to the stratigraphic ages: Precambrian, Paleozoic, Mesozoic, and Cenozoic. This systematic classification is fundamentally justified by the distinct paleontological assemblage characteristics preserved in strata of different geological periods [16,24].

The average nearest neighbor analysis yields an index for evaluating the specific clustering degree of point elements. This analysis enables a comparative study of which geological period exhibits the highest clustering intensity among paleontological geosites. In our calculations, we first assumed a random spatial distribution of paleontological geosites across all periods within Liaoning Province. Subsequently, using Equations (2) and (3), the nearest neighbor ratio and other parameters were computed through the Spatial Statistics Tools in ArcGIS 10.8.1 software (

Table 3. Summary of the average nearest neighbor of paleontological geosites in Liaoning Province.

Geological Age	Counts	Observed Mean Distance/m	Expected Mean Distance/m	Nearest Neighbor Ratio	Z-Score	p-Value
Total	252	7249.4461	13,397.1796	0.541117	−13.935829	0.000000
Cenozoic	47	22,038.2251	31,021.6376	0.710415	−3.798013	0.000146
Mesozoic	164	8534.7699	16,607.0206	0.513925	−11.908463	0.000000
Paleozoic	30	11,296.0590	38,828.7153	0.290920	−7.429962	0.000000
Precambrian	11	27,121.1182	64,123.5129	0.422951	−3.661337	0.000251

). The study area was defined as the minimum bounding rectangle encompassing all paleontological geosites in Liaoning Province.

For paleontological geosites from the Cenozoic, Mesozoic, Paleozoic, and Precambrian periods, the observed mean distances were all lower than their expected mean distances, yielding nearest neighbor ratios of 0.710415, 0.513925, 0.290920, and 0.422951, respectively. The analysis demonstrates that paleontological geosites across all four studied geological periods in Liaoning Province exhibit statistically significant clustering at the provincial scale, as evidenced by the nearest neighbor ratio values being consistently below 1. Furthermore, all four periods reveal Z-scores < −2.58 with p-values < 0.01, confirming the statistical significance of these results. Of the 252 total paleontological geosites in the province, 164 (65.08%) are Mesozoic in age. The nearest neighbor ratio for these Mesozoic paleontological geosites closely matches that of the total dataset (0.541117), indicating its dominant role in governing spatial distribution patterns.

While paleontological geosites in Liaoning Province exhibit clustered distribution patterns across all four geological periods, variations in their nearest neighbor ratio values reveal significant differences in clustering intensity. Under the condition of spatially aggregated distributions, lower nearest neighbor ratio values indicate stronger clustering. Our analysis demonstrates distinct temporal variations in spatial aggregation: the Paleozoic period displays the most pronounced clustering (nearest neighbor ratio = 0.290920), followed by the Precambrian (0.422951), Mesozoic (0.513925), and Cenozoic (0.541117) periods, indicating an initial intensification followed by progressive weakening of clustering intensity in the geological history.

Kernel density estimation was applied to quantify the spatial clustering of paleontological geosites in Liaoning Province using Equation (4). After iterative optimization of bandwidth parameters, a bandwidth (search radius) of 20 km was ultimately selected for kernel density estimation in ArcGIS 10.8.1, achieving an optimal balance between spatial accuracy and visual representation in the resultant density surfaces (Figure 5). The results reaffirm the pronounced spatial heterogeneity of paleontological geosites across the province. At the provincial scale, paleontological geosites exhibit a distinct primary concentration zone with an elliptical configuration, centered on Chaoyang and its vicinity, while other regions demonstrate negligible clustering.

Temporally stratified analysis reveals marked variations in distribution patterns of paleontological geosites across geological epochs. Precambrian paleontological geosites, though scarce, exhibit a single dominant cluster in southern Dalian. Paleozoic paleontological geosites are concentrated primarily near Benxi, with secondary concentrations in southern Dalian and southern Chaoyang. Mesozoic paleontological geosites display high spatial concentration, with two nearly contiguous high-density zones focused exclusively around Chaoyang. Cenozoic paleontological geosites form a left-leaning U-shaped agglomeration zone in north-central Liaoning, with sub-agglomeration zones in Chaoyang and Southern Dalian.

3.2. Potential Factors of Paleontological Geosites Locations

3.2.1. Elevation

Physical geography, particularly elevation factors, plays a significant role in the spatial distribution of paleontological geosites [8,57]. To analyze this relationship, DEM data for Liaoning Province were acquired and processed using ArcGIS 10.8.1. Elevation values were extracted and spatially correlated with the locations of known paleontological geosites (Figure 6). The elevations of paleontological geosites were determined as the median altitude within a 500 m radius centered on their coordinates.

Liaoning Province exhibits distinct topographic variations, characterized by undulating hills and mountains in its western and eastern regions, while the central, northern, and southeastern areas consist of low-lying plains. Elevation distribution analysis (

Table 4. Spatial relationship statistics of paleontological geosites and elevations in Liaoning Province.

Geological Age or Region	Elevation (m)
	[0, 100)	[100, 200)	[200, 300)	[300, 400)	[400, 800)	≥800
Total	20.63% (52)	20.63% (52)	18.25% (46)	14.68% (37)	25.40% (64)	0.40% (1)
Cenozoic	42.55% (20)	27.66% (13)	17.02% (8)	6.38% (3)	6.38% (3)	0.00% (0)
Mesozoic	8.54% (14)	21.34% (35)	18.90% (31)	18.90% (31)	32.32% (53)	0.00% (0)
Paleozoic	33.33% (10)	10.00% (3)	20.00% (6)	10.00% (3)	23.33% (7)	3.33% (1)
Precambrian	72.73% (8)	9.09% (1)	9.09% (1)	0.00% (0)	9.09% (1)	0.00% (0)
Entire province area	35.01%	18.28%	14.01%	11.07%	20.54%	1.10%

Square brackets denote closed intervals (inclusive of endpoints), whereas parentheses represent open intervals (exclusive of endpoints). Parenthetical values following percentages indicate the absolute counts of paleontological geosites.

) reveals that approximately 53.29% of the province lies in low elevations (<200 m), predominantly in the central, northern, and southeastern plains. Moderate elevations (200–300 m and 300–400 m) account for 14.01% and 11.07% of the total area, respectively, primarily occurring in transitional zones between plains and mountainous regions. Higher elevations (400–800 m) cover 20.54% of the province, mainly concentrated in the western and eastern regions. Notably, only 1.10% of the area exceeds 800 m, with these elevated zones restricted to localized mountain peaks in the western and eastern sectors.

The superposition analysis reveals distinct patterns among the 252 paleontological geosites in Liaoning Province’s elevations. The majority (n = 187, 74.21%) are concentrated at elevations below 400 m, with 104 geosites (41.27%) located in low-altitude areas (<200 m) and 83 geosites (32.94%) distributed across mid-elevation areas (200–400 m). Higher elevation areas (400–800 m) contain 64 geosites (25.40%), while only one geosite (0.40%) occurs above 800 m. Notably, the elevational distribution of these geosites varies significantly across geological periods. Cenozoic paleontological geosites (n = 33) are predominantly situated below 200 m (70.21%). In contrast, Mesozoic paleontological geosites (n = 164) exhibit a distinct preference for moderate to high elevations in western Liaoning, with 62 geosites (37.80%) located at elevations of 300–400 m and 53 (32.32%) at elevations of 400–800 m. Paleozoic paleontological geosites are primarily concentrated in the eastern part of the province at low to moderate elevations, with 13 geosites (43.33%) below 200 m and nine geosites (30.00%) at 200–400 m. Furthermore, 81.82% of Precambrian paleontological geosites occur in coastal lowland areas (<200 m).

Statistical results reveal a significant correlation between paleontological geosites’ distributions and elevations in Liaoning Province. Quantitatively, these geosites exhibit a moderate preference for mid- to high-elevation areas. Notably, the 200–400 m elevation range, covering 25.08% of the province’s area, contains 32.93% of all paleontological geosites. Similarly, the 400–800 m zone (20.54% of provincial area) hosts 25.40% of paleontological geosites. A temporally stratified analysis demonstrates distinct elevational preferences among geological periods. Mesozoic paleontological geosites predominantly occur in mid-to-high elevation areas (100–800 m) in Western Liaoning; Paleozoic paleontological geosites are concentrated in mid-to-high elevation zones (>100 m) of Eastern Liaoning and low elevation areas (<100 m) in Liaodong Peninsula; Cenozoic and Precambrian geosites show greater aggregation in low-elevation plains, particularly below 200 m. These distribution patterns suggest that while elevation represents one potential influencing factor, the clustered distribution of geosites across different periods may be more strongly associated with other geological structures.

3.2.2. Slope Aspect

Slope aspect is a critical factor in understanding both natural environmental patterns and the spatial distribution of paleontological geosites [58]. To evaluate its influence on fossil preservation and exposure, we conducted a spatial correlation analysis between slope aspect (derived from DEM data processed in ArcGIS 10.8.1) and the coordinates of known paleontological geosites in Liaoning Province (Figure 7). For each geosite, the representative slope aspect was defined as the median value within a 500 m radial buffer centered on its coordinates. This methodology mitigates localized topographic noise while robustly capturing the dominant slope orientation most relevant to fossilization processes.

The topography of Liaoning Province exhibits a distinctive horseshoe-shaped configuration, characterized by a general north-to-south inclination with descending gradients from both eastern and western flanks toward the central lowlands. For analytical purposes, slope aspects were categorized into four 90° intervals (45–135°, 135–225°, 225–315°, and 315–45°) (

Table 5. Spatial relationship statistics of paleontological geosites and slope aspects in Liaoning Province.

Geological Age or Region	Slope Aspects
	−1 (Flat Terrain)	[45°, 135°) (Eastern)	[135°, 225°) (Southern)	[225°, 315°) (Western)	[315°, 360°) and [0°, 45°) (Northern)
Total	0.00% (0)	35.32% (89)	41.67% (105)	21.43% (54)	1.59% (4)
Cenozoic	0.00% (0)	19.15% (9)	44.68% (21)	34.04% (16)	2.13% (1)
Mesozoic	0.00% (0)	40.85% (67)	39.63% (65)	18.90% (31)	0.61% (1)
Paleozoic	0.00% (0)	26.67% (8)	56.67% (17)	13.33% (4)	3.33% (1)
Precambrian	0.00% (0)	45.45% (5)	18.18% (2)	27.27% (3)	9.09% (1)
Entire province area	2.10%	26.71%	25.67%	25.83%	21.80%

Notes follow the same convention as Table 4.

). Analysis of provincial slope aspect distribution reveals that flat areas (no slope aspect) constitute 2.10% of the total area, while all other aspects each exceed 20% coverage. Specifically, east-facing slopes predominate (26.71%), followed by south-facing (25.67%), west-facing (25.83%), and north-facing slopes (21.80%). Regionally, the mountainous and hilly areas of western and eastern Liaoning, including the Liaodong Peninsula, predominantly exhibit eastern, southern, and western aspects. In contrast, the central plains display a more balanced aspect distribution.

The analysis of 252 paleontological geosites in Liaoning Province reveals a strong slope aspects preference, with only a few geosites exceptions. The 248 out of total paleontological geosites exhibit distinct aspect distributions: south-facing (105 geosites, 41.67%), west-facing (89 geosites, 35.32%), and east-facing (54 geosites, 21.43%). Only four geosites (1.59%) represent exceptions, exhibiting north-facing slope aspects. When examined by geological period, all periods show a predominant preference for sun-facing aspects (south-, west-, or east-facing), with north-facing geosites being exceptionally rare (only one per period). Specifically, Cenozoic and Paleozoic paleontological geosites demonstrate a south-facing preference (44.68% and 56.67% of their respective totals, comprising 21 and 17 geosites); Mesozoic paleontological geosites are predominantly east- and south-facing (40.85% and 39.63% of the total for the period, representing 67 and 65 geosites, respectively); Precambrian paleontological geosites show an east-facing predominance (45.45% of the total for the period).

Statistical analysis reveals a significant preferential distribution of paleontological geosites along sunward-facing slopes in Liaoning Province. This spatial distribution pattern likely results from the combined effects of natural erosion processes and anthropogenic activities. Sunward-facing slopes experience higher solar radiation and greater diurnal temperature fluctuations, promoting intense physical weathering (e.g., freeze–thaw cycles, thermal expansion–contraction) [59,60,61]. These processes accelerate rock fracturing and enhance the exposure of fossil-bearing strata. Additionally, sunward-facing slopes generally exhibit denser drainage networks (Figure 6) and greater accessibility in Liaoning Province, leading to higher exploration intensity and potential sampling bias due to preferential fossil discovery. Human disturbances, such as agricultural expansion, mining, and infrastructure development, are also more frequent on sunward-facing slopes, further increasing the likelihood of fossil exposure and detection [8,11].

3.2.3. Stratigraphic Exposure

The distribution of stratigraphic exposure serves as a fundamental control on the spatial occurrence of paleontological geosites [8,24,62,63]. However, the precise nature of their spatial correlation remains contingent upon regional geological contexts. To elucidate this relationship in Liaoning Province, we systematically extracted Liaoning stratigraphic outcrop data from the 1:1,000,000 Geological Map of China, categorizing them by geological age (Cenozoic, Mesozoic, Paleozoic, and Precambrian). These data were then spatially integrated with the documented locations of contemporaneous paleontological geosites, generating a series of overlay maps that delineate the spatial association between outcrop distribution and paleontological geosites distribution across distinct geological epochs (Figure 8).

The superposition analysis reveals a significant spatial coupling between the distribution of stratigraphic outcrops and paleontological geosites across Liaoning Province. The Precambrian strata exhibit extensive exposure in Liaoning, though their spatial distribution is highly heterogeneous (Figure 8A). Major outcrops are concentrated in Eastern Liaoning and the Liaodong Peninsula, with secondary occurrences in Western Liaoning and negligible exposure elsewhere. Notably, all currently documented Precambrian paleontological geosites are exclusively located in Eastern Liaoning and the Liaodong Peninsula—regions with the highest outcrop density—while no contemporaneous fossil sites have been discovered in other areas.

Paleozoic strata in Liaoning are spatially restricted, with fragmented exposures primarily distributed along the eastern and western parts of Liaoning, as well as the coastal zones of the Liaodong Peninsula (Figure 8B). These limited outcrop areas coincide with the principal occurrences of Paleozoic paleontological geosites, such as Dalian, Chaoyang, Benxi and Anshan. In contrast, other regions exhibit only sporadic Paleozoic strata exposures, and no significant geosites have been reported to date.

The Mesozoic continental volcanic–sedimentary sequences are exceptionally well-developed in Western Liaoning (Figure 2 and Figure 8C). Volcanic rocks are predominantly distributed across Eastern Liaoning and the Liaodong Peninsula, with lesser occurrences in Western Liaoning and sporadic exposures in northern Liaoning. In contrast, Mesozoic sedimentary rocks exhibit an inverse distribution pattern, forming extensive deposits in Western Liaoning, moderate accumulations in Northern Liaoning, and only isolated outcrops in eastern Liaoning and the Liaodong Peninsula. This alternating volcanic-basin architecture exerts control on the spatial distribution of Mesozoic paleontological geosites. The majority of Mesozoic paleontological geosites are concentrated in western Liaoning, where sedimentary basins are prevalent, while isolated occurrences have been confirmed in northern Liaoning and the Liaodong Peninsula.

Cenozoic strata in Liaoning are primarily constrained by fluvial and marine processes, with major accumulations occurring in extensive fluvial plains of central and northern Liaoning, intermontane fluvial basins in western and eastern Liaoning, and coastal zones along the Liaodong Peninsula (Figure 8D). Notably, Cenozoic paleontological geosites exhibit a distinct linear distribution paralleling these riverine and coastal depositional systems, demonstrating clear stratigraphic control on fossil preservation.

The spatial distribution of stratigraphic outcrops exerts a fundamental control on the preservation and distribution of paleontological geosites, serving as a determinant of geosite spatial patterns. Our analysis demonstrates a strong lithostratigraphic dependency of paleontological geosites, revealing that they predominantly occur in areas with concentrated coeval stratigraphic outcrops. Notably, we observe significant enrichment of paleontological geosites along the marginal zones of stratigraphic units, particularly in proximity to formation interfaces. These zones, representing periods of dramatic paleoenvironmental change in Earth’s history, constitute critical stratigraphic intervals for exceptional fossil preservation. The concentration of paleontological geosites at these boundaries suggests that major geological transitions created optimal taphonomic conditions for fossilization.

3.2.4. Road Proximity

Road infrastructure significantly influences the distribution and conservation of paleontological geosites [8], particularly as newly exposed roadcuts may directly reveal critical fossil-bearing strata. To assess the spatial relationship between paleontological geosites and adjacent road networks in Liaoning Province, road data were extracted from the OpenStreetMap website and overlaid with paleontological geosite locations using ArcGIS 10.8.1 (Figure 9). Proximity analysis was then conducted to calculate the minimum Euclidean Distance between each paleontological geosite center and the nearest road segment (

Table 6. Spatial relationship statistics of paleontological geosites and roadways in Liaoning Province.

Geological Age	The Distance to the Nearest Roadways (m)
	[0, 500)	[500, 1000)	[1000, 1500)	[1500, 2000)	[2000, 3000)	≥3000
Total	52.38% (132)	13.89% (35)	11.11% (28)	11.11% (28)	7.14% (18)	4.37% (11)
Cenozoic	65.96% (31)	14.89% (7)	10.64% (5)	4.26% (2)	4.26% (2)	0.00% (0)
Mesozoic	43.29% (71)	14.02% (23)	12.80% (21)	14.63% (24)	14.63% (15)	6.10% (10)
Paleozoic	66.67% (20)	13.33% (4)	6.67% (2)	6.67% (2)	6.67% (1)	3.33% (1)
Precambrian	90.91% (10)	9.09% (1)	0.00% (0)	0.00% (0)	0.00% (0)	0.00% (0)

Notes follow the same convention as Table 4.

).

Liaoning Province maintains an extensive and well-developed road transportation infrastructure system. As of 2024, the province’s total road network spans approximately 131,484 km, resulting in a mean road density of 89.34 km per 100 square kilometers. However, the spatial distribution of road infrastructure exhibits significant heterogeneity across the province. The road network demonstrates distinct regional disparities. The eastern mountainous region demonstrated relatively low road density due to challenging topography and consequent high construction costs. The flat plain and coastal zones exhibited substantially higher road density, benefiting from favorable terrain conditions and economic development priorities. Major financial centers, particularly Shenyang and Dalian, showed the highest road network densities, reflecting their economic significance and corresponding infrastructure investments.

The analysis reveals a pronounced spatial correlation between paleontological geosites and the location of roadway infrastructure. There is a statistically significant clustered distribution of paleontological geosites along major roadways, exhibiting a distinct distance–decay relationship. The distribution of 252 paleontological geosites reveals strong proximity patterns to roads, with 52.83% (n = 132) located within 500 m of roadways, while only 4.37% (n = 11) occur beyond 3000 m from roadways. This distribution exhibits slight temporal variations across geological periods: Precambrian geosites show the highest road affinity, with 90.91% situated within 500 m of roads; Paleozoic and Cenozoic geosites display intermediate proximity, with 66.67% and 65.96%, respectively, within 500 m buffers; Mesozoic geosites demonstrate relatively greater dispersion, with 43.29% in the immediate (<500 m) roadside zone. Mesozoic paleontological geosites show significantly weaker spatial association with road networks, with less than half located within 500 m roadside buffers, compared to that of Precambrian–Paleozoic–Cenozoic geosites.

The pronounced spatial clustering of paleontological geosites along roadways, exhibiting a statistically significant distance–decay relationship, is a result of synergistic geological and anthropogenic factors. Road construction activities generate extensive fresh exposures through engineered cuttings and excavations, significantly enhancing the detectability of fossiliferous strata that would otherwise remain concealed. This artificial exposure effect is particularly pronounced in valley corridors (e.g., Eastern Liaoning and Western Liaoning), where both natural outcrops of fossil-bearing formations and transportation routes preferentially concentrate due to geomorphological constraints. Proximity to roadways facilitates more rapid scientific discovery and subsequent conservation intervention, effectively creating a prioritized preservation for accessible paleontological geosites.

3.2.5. Population Density

The discovery of paleontological geosites is often directly linked to human activities [8,64,65], suggesting an intrinsic relationship between population density and the distribution of such geosites. To systematically examine this relationship in Liaoning Province, we first compiled the 2023 population density data at the county level, with missing values interpolated using historical records to ensure data completeness. Subsequently, we employed QGIS 3.34.2 to classify the population density data into six distinct categories using the natural breaks (Jenks) method. Finally, we conducted a superposition analysis by integrating the classified population density map with the geographic coordinates of known paleontological geosites, generating a composite distribution map to assess their spatial association (Figure 10).

Liaoning Province comprises 100 county-level cities distributed across 14 prefecture-level cities. Statistical analysis of population density reveals two pronounced spatial disparities in population distribution across the province. First, high population density areas exhibit marked concentration in the urban cores of all prefecture-level cities, while surrounding regions maintain substantially lower densities. Second, there is extreme heterogeneity in population distribution at the county level. The demographic landscape demonstrates: five county-level cities (all located in Shenyang and Dalian metropolitan cores) with exceptionally high densities (11,383–13,204 persons/km²), 29 county-level cities showing intermediate densities of 832–11,383 persons/km², the majority (66 county-level cities) displaying low population densities (63–832 persons/km²).

The superposition analysis reveals a statistically significant concentration of paleontological geosites in low-population-density regions of the province (

Table 7. Spatial relationship statistics of paleontological geosites and population density in Liaoning.

Geological Age	Population Density (Persons/km2)
	[63, 832)	[832, 2177)	[2177, 3980)	[3980, 6810)	[6810, 11,383)	[11,383, 13,204]
Total	96.83% (244)	2.37% (6)	0.40% (1)	0.00% (0)	0.40% (1)	0.00% (0)
Cenozoic	93.62% (44)	4.26% (2)	2.12% (1)	0.00% (0)	0.00% (0)	0.00% (0)
Mesozoic	99.39% (163)	0.61% (1)	0.00% (0)	0.00% (0)	0.00% (0)	0.00% (0)
Paleozoic	93.33% (28)	6.67% (2)	0.00% (0)	0.00% (0)	0.00% (0)	0.00% (0)
Precambrian	81.82% (9)	9.09% (1)	0.00% (0)	0.00% (0)	9.09% (1)	0.00% (0)

Notes follow the same convention as Table 4.

, Figure 10). Of the 252 documented paleontological geosites, 244 (96.83%) are located within the 66 county-level cities exhibiting the lowest population densities. In contrast, only eight paleontological geosites (3.17%) are found in the 29 medium-density cities (832–11,383 persons/km²), while remarkably, none were present in the five highest-density cities. The observed spatial correlation can be attributed to several geomorphological and anthropogenic factors. In Liaoning Province, regions with high population density are typically located near urban centers of prefecture-level cities, characterized by relatively low average elevations. In contrast, the paleontological geosites predominantly occur in mid-to-high elevation zones.

Temporally stratified analysis reveals distinct spatial relationships between the distribution of paleontological geosites and population density across different geological periods. Among the 11 Precambrian paleontological geosites, nine (81.82%) are located in county-level cities with the lowest population density. Of the 164 Mesozoic geosites, 163 (99.39%) occur in county-level cities with the lowest population density. For the 47 Cenozoic geosites, 44 (93.62%) are situated in county-level cities with the lowest population density. Similarly, 28 out of 30 paleontological geosites (93.33%) fall within county-level cities exhibiting the lowest population density. These results indicate that Mesozoic geosites are almost exclusively concentrated in the least densely populated regions of the province, a critical characteristic that should be valued in the formulation of conservation strategies.

As demonstrated in previous analyses, elevated terrains, such as mountains and hills, exhibit superior bedrock exposure conditions, thereby enhancing the detectability of paleontological remains. However, such topographically constrained areas are less conducive to large-scale urban development [8]. Furthermore, intensive human activities in densely populated regions may adversely affect the preservation and integrity of fossil resources. Conversely, low-population-density regions experience reduced surface modification, thereby facilitating better conservation of paleontological heritage.

3.2.6. GDP per Capita

Gross Domestic Product (GDP) per capita serves as a robust metric for assessing regional macroeconomic performance and is widely recognized as a key indicator of economic prosperity and living standards [66]. To investigate the spatial relationship between paleontological geosites and economic development in Liaoning Province, this study compiled county-level GDP per capita data for 2023 (with missing values imputed using interpolation methods). Using QGIS 3.34.2, the data were classified into six distinct categories through the natural breaks (Jenks) method, represented by a graduated color scheme, and subsequently overlaid with the distribution of paleontological geosites for spatial correlation analysis (Figure 11).

The superposition analysis reveals a significant inverse correlation between the distribution of paleontological geosites and regional economic development levels (measured by GDP per capita) in Liaoning Province (

Table 8. Spatial relationship statistics of paleontological geosites and GDP per capita in Liaoning Province.

Geological Age	GDP Per Capita (CNY)
	[6802, 23,590)	[23,590, 41,905)	[41,905, 76,698)	[76,698, 131,226)	[131,226, 176,382)	[176,382, 282,592]
Total	24.60% (62)	55.95% (141)	12.30% (31)	4.76% (12)	0.40% (1)	1.98% (5)
Cenozoic	14.89% (7)	44.68% (21)	27.66% (13)	10.64% (5)	2.13% (1)	0.00% (0)
Mesozoic	32.32% (53)	65.24% (107)	1.83% (3)	0.61% (1)	0.00% (0)	0.00% (0)
Paleozoic	6.67% (2)	36.67% (11)	40.00% (12)	6.67% (2)	0.00% (0)	10.00% (3)
Precambrian	0.00% (0)	18.18% (2)	27.27% (3)	36.36% (4)	0.00% (0)	18.18% (2)

Notes follow the same convention as Table 4.

). Notably, 203 paleontological geosites (80.55% of the provincial total) are concentrated in county-level cities with per capita GDP below 41,905 Chinese Yuan (CNY), which account for 52% of the total. Thirty-three paleontological geosites (17.06% of total) are situated in 43 intermediate economic county-level cities (41,905–131,226 CNY), while only six paleontological geosites (2.38%) occur in the 10 most economically developed county-level cities (131,266–282,592 CNY).

Temporally stratified analysis reveals distinct spatial relationships between the distribution of paleontological geosites and GDP per capita across different geological periods (Table 8). Among the Precambrian, Paleozoic, and Cenozoic paleontological geosites, 81.82%, 90.00%, and 97.87%, respectively, are located in county-level cities with a GDP per capita below 131,226 CNY. Notably, all Mesozoic paleontological geosites are situated in counties with GDP per capita under 131,226 CNY, among which 97.56% occur in areas where GDP per capita is less than 41,905 CNY. The more concentrated distribution of Mesozoic geosites in economically underdeveloped regions warrants greater attention in future geoheritage conservation efforts.

This robust negative correlation reveals a systematic spatial association between high paleontological significance (quantified by the density of paleontological geosites) and lower economic development (measured by per capita GDP), which may be attributed to multiple interrelated factors. First, fossil-rich areas predominantly occur in mountainous or hilly terrains (mid-to-high elevation) with well-exposed geological strata—landscapes that inherently constrain economic development due to poor accessibility and limited agricultural potential, thereby sustaining lower GDP levels. Second, regions with higher per capita GDP typically experience intensified industrialization, urbanization, and land-use modifications (e.g., infrastructure expansion, mining), which can directly degrade fossil-bearing strata or obscure critical outcrops. Third, per capita GDP often exhibits collinearity with population density (e.g., urban centers with high GDP and high human activity), and anthropogenic disturbances (e.g., agriculture, construction) may compromise the preservation of paleontological deposits.

3.3. Degradation Risk of Paleontological Geosites

The spatial relationship with potential environmental degradation factors has long been recognized as a critical determinant of geosite degradation risk [9,24,67,68]. To quantitatively assess the degradation risk of paleontological geosites in Liaoning Province, this study examines all county-level cities within the region. We extracted quantitative data on paleontological geosites, mineral occurrences, and geological hazard sites for each county-level city. Subsequently, a bivariate spatial autocorrelation analysis was conducted using GeoDa 1.22 to analyze the spatial distribution patterns between paleontological geosites and mineral occurrences or geological hazard sites (Figure 12).

The bivariate global Moran’s I analysis reveals a statistically significant yet weak positive spatial autocorrelation (I = 0.210, p = 0.006, z = 3.952) between the distributions of paleontological geosites and mineral occurrences at the county-level cities. This implies a tendency for county-level cities with higher densities of paleontological geosites to also possess greater concentrations of mineral resources, and conversely. However, despite statistical significance (p < 0.05, z > 1.96), the low Moran’s I value—approaching zero—suggests that the observed spatial association is weak and geographically constrained, with significant correlation detectable only in specific regions of Liaoning Province.

The local spatial analysis reveals distinct clustering patterns across 20 county-level cities: high-high clustering (n = 8), low-high clustering (n = 7), and low-low clustering (n = 5) (Figure 12A). No high-low clusters are observed. Notably, high-high clustering areas represent key regions where paleontological geosites spatially overlap with intensive mineral mining activities. These clusters are predominantly distributed within the Western Liaoning and Eastern Liaoning PGAAs, which also serve as the primary concentrations of Mesozoic and Paleozoic paleontological geosites. Specifically, these significant clusters are identified in eight county-level cities: Beipiao, Yixian, Longcheng, Chaoyang, Lingyuan, Kalaqin, Benxi, and Huanren. This spatial congruence underscores a potential conflict between geoheritage conservation and mineral resource development. Given this overlap, the high-high cluster regions are identified as having the highest degradation risk for paleontological geosites, necessitating prioritized conservation strategies.

The bivariate spatial autocorrelation analysis yielded a global Moran’s I index of 0.080 (p = 0.087, z = 1.417) for the relationship between paleontological geosites and geological hazard sites at the county level. This statistically non-significant result (p > 0.05, |z| < 1.96) indicates the absence of meaningful spatial correlation between paleontological geosites and geological hazard sites across Liaoning Province at a provincial scale. Nevertheless, the observed global Moran’s I index of 0.080, while not equal to zero, may still reflect localized spatial associations between these variables in certain subregions.

The local spatial analysis reveals important subregional clustering patterns among the 19 analyzed county-level cities. Three distinct spatial associations were identified: high-high clustering (n = 3), low-high clustering (n = 6), and low-low clustering (n = 10) (Figure 12B). High-high clusters are predominantly located in Benxi and Huanren within the Eastern Liaoning PGAA, as well as in Pulandian on the Liaodong Peninsula PGAA. These three county-level cities represent areas requiring particular attention, as they also coincide with regions characterized by high concentrations of Paleozoic and Precambrian paleontological geosites. These regions combine high paleontological resources with elevated exposure to geological hazards, resulting in significantly increased degradation risks for geoheritage geosites. This localized spatial coupling suggests that, while broad-scale correlations may be absent, targeted risk assessment remains crucial in specific areas of high vulnerability.

4. Discussion

4.1. Geoheritage Conservation and Management Dilemmas

Liaoning Province, recognized as one of China’s most fossil-rich regions, hosts 252 significant paleontological geosites. A multi-scale analysis of the spatial distribution patterns of paleontological geosites in Liaoning Province reveals pronounced heterogeneity across prefectural-city, regional, and geological time scales. These geosites are distributed across all 14 prefecture-level cities—a rare phenomenon among provincial-level administrative units in China that underscores the exceptional abundance and widespread nature of its paleontological resources.

However, the distribution is markedly uneven. Chaoyang City contains the highest concentration, with 108 paleontological geosites accounting for 42.86% of the provincial total, followed by Jinzhou, Huludao, Dalian, Benxi, and the other cities. Regional-scale spatial analysis has identified five distinct PGAAs: the Western Liaoning, Eastern Liaoning, Liaodong Peninsula, Northern Liaoning, and Northwestern Liaoning regions. Among these, the Western Liaoning PGAA, centered around Chaoyang, represents the most densely concentrated and significant cluster within the province. This area encompasses 169 paleontological geosites, accounting for 67.06% of the total in Liaoning.

The spatial distribution of paleontological geosites in Liaoning Province exhibits distinct temporal patterns across geologic periods, with significant variations in clustering intensity and geographic configuration. Quantitative analysis reveals a hierarchical clustering intensity: Paleozoic paleontological geosites demonstrates the strongest spatial aggregation, followed by Precambrian, Mesozoic, and Cenozoic geosites, respectively. Geographically, these paleontological geosites assemblages display period-specific distribution patterns: Precambrian paleontological geosites are predominantly concentrated along the coastal region of Liaodong Peninsula; Paleozoic paleontological geosites show marked clustering in the Eastern Liaoning; Mesozoic paleontological assemblages are primarily distributed around the Western Liaoning; Cenozoic paleontological geosites are located in the whole province.

The pronounced spatial heterogeneity has created significant disparities in conservation effectiveness and management practices. In Liaoning Province, this imbalance manifests through three critical dimensions:

(i) Research and Knowledge Gaps: While some PGAAs (e.g., Western Liaoning) receive substantial research attention, dispersed paleontological geosites in other regions remain understudied [22,24,33], resulting in data deficiencies that hinder a comprehensive understanding of key evolutionary events. This sampling bias may skew interpretations of regional biodiversity patterns through geologic time.

(ii) Resource Allocation Disparities: The fossil-rich Western Liaoning (Chaoyang-Jinzhou) benefits from preferential policy support, dedicated funding, and advanced conservation technologies [24,69,70]. Conversely, areas with dispersed paleontological geosites often experience management deficiencies, increasing their vulnerability to illegal fossil extraction and black-market trade—a particular concern for unique specimens [71,72].

(iii) Geotourism Development Imbalance: The concentration of paleontological geosites in certain regions has resulted in asymmetrical development, whereby geosites from particular geologic periods (e.g., the Mesozoic) attract research-driven tourism and associated economic benefits [71,73], while those from other periods frequently lack the foundation required to utilize their paleontological heritage. This economic disparity may inadvertently promote destructive land-use practices in areas where geosites receive inadequate attention or investment [11,74].

4.2. Multifactorial Controls on Paleontological Geosite Distribution

From a natural environmental perspective, the geospatial analysis reveals significant correlations between paleontological geosites and key topographic factors, including elevation, slope aspect, and stratigraphic exposure. Paleontological geosites exhibit a pronounced preference for mid- to high-elevation zones (200–800 m), a pattern that remains consistent across multiple geological periods. A strong aspect-related bias is also evident, with over 98% of documented geosites occurring on south-, east-, and west-facing slopes, whereas north-facing slopes are largely devoid of significant paleontological geosites. Nevertheless, the distribution of paleontological geosites is predominantly controlled by stratigraphic exposure conditions.

From a socioeconomic perspective, the distribution of paleontological geosites in Liaoning Province exhibits significant spatial correlations with proximity to roads, population density, and GDP per capita. Statistical analyses reveal that these paleontological geosites display strong spatial clustering along road networks, following a statistically significant distance–decay relationship. This pattern likely reflects preferential discovery, documentation, and subsequent inclusion in conservation inventories for sites proximal to transportation routes. Moreover, paleontological geosites are disproportionately concentrated in regions characterized by lower population density and lower economic development (GDP per capita).

The spatiotemporal heterogeneity of paleontological geosites distribution in Liaoning Province reveals a complex interplay between natural environmental constraints and socioeconomic influences. This distribution may be attributed to two interrelated mechanisms: first, the predominance of mountainous and hilly terrain—less amenable to urbanization—promotes favorable bedrock exposure, thereby enhancing the detectability of fossil remains; second, diminished anthropogenic disturbance in sparsely populated and economically underdeveloped regions supports the superior preservation of geoheritage features.

Regardless of whether the aforementioned natural and socioeconomic factors exhibit positive or negative spatial correlations with paleontological geosites, they collectively form a complex system governing their distribution and state. From a holistic perspective, natural environmental factors constitute the foundational conditions for the presence of paleontological geosites [8,16,24]. Variables such as elevation, slope aspect, and stratigraphic exposure fundamentally determine the location, existence, and surface exposure potential of geosites. On the other hand, socioeconomic conditions—including proximity to roads, population density, and GDP per capita—serve as necessary prerequisites for the discovery of these geosites, influencing the timing and likelihood of their detection.

It is important to note that these potential influencing factors are not static and may exhibit regional variations. Socioeconomic factors, in particular, are highly variable across different contexts. For example, prior spatial data mining in Shandong Province, China, revealed no significant correlation between paleontological geosite distribution and GDP per capita, while a stronger association was observed with disposable personal income [8]. These findings underscore the importance of adopting a dynamic and developmental perspective in geoheritage conservation and management, emphasizing the need to account for temporal changes and contextual differences in driving factors.

4.3. Multi-Scale Geoheritage Conservation Framework

Renowned for their uniqueness and rarity globally, the paleontological resources of Liaoning Province represent invaluable geoheritage [16,19,20]. To ensure their scientific, efficient, and sustainable conservation and management, it is imperative to establish a multi-scale conservation framework. The spatial distribution of paleontological geosites in Liaoning Province and their coupling relationships with multi-dimensional natural and socioeconomic factors provide a critical foundation for formulating targeted conservation and precise management strategies [8,75,76,77]. This proposal systematically outlines conservation strategies from three distinct dimensions: the prefectural-level city, the regional, and the geological time scale.

(i) Prefecture-level city scale: Strengthening management foundations and implementing precision-based differentiated governance.

As the primary implementing bodies for paleontological geoheritage conservation, municipal governments and relevant agencies play a decisive role in protection outcomes through the refinement of their operational practices. It is recommended that each city conduct periodic, comprehensive, and systematic surveys of its entire jurisdiction, with particular attention to paleontological geosites located in economically underdeveloped regions with low population density, as well as exposed sedimentary rock formations on sun-facing slopes at medium to high elevations. Furthermore, prefecture-level cities should implement precision classification and tiered management strategies tailored to their specific geoheritage endowments. Key cities with rich paleontological resources, such as Chaoyang, Jinzhou, Huludao, Dalian, and Benxi, should continue to advance the development of paleontological geoparks. Meanwhile, other prefectural-level cities should progressively establish specialized paleontological protection systems to ensure coordinated and effective conservation across the province.

(ii) Regional scale: Enhancing collaborative linkages and establishing an integrated conservation network.

The distribution of paleontological geosites transcends administrative boundaries, necessitating coordinated efforts across different PGAAs within the province that extend beyond prefectural divisions. Building on spatial correlation analyses between known paleontological geosites and multiple natural and socioeconomic factors, high-potential zones for paleontological resources (e.g., Western Liaoning PGAA) as well as areas at elevated risk of degradation (e.g., Western Liaoning PGAA, Eastern Liaoning PGAA, and Liaodong Peninsula PGAA) have been identified. Particularly for the Western Liaoning PGAA, which exhibits exceptionally rich paleontological resources, it is essential to establish a joint conference mechanism for paleontological heritage conservation. This mechanism would facilitate the development of unified protection standards, consistent monitoring indicators, and shared surveillance data for all paleontological geoheritage sites throughout the western Liaoning region.

(iii) Geological time scale: Deepening value understanding and guiding forward-looking conservation.

In practical conservation efforts, it is imperative to adopt a macro-perspective grounded in geological history for the management of paleontological geosites, so as to enhance recognition of the distinctions among sites from different geological periods and deepen the understanding of their scientific significance. It is recommended to further strengthen collaboration with research institutions to clarify the precise positions and unique contributions of these geosites within the tree of life. Furthermore, the design of integrated paleontological research and tourism routes, connecting key paleontological geosites across various geological periods, can effectively transform conservation outcomes into educational and economic benefits. Such initiatives will create a virtuous cycle in which generated resources can reciprocally support sustained preservation efforts.

5. Conclusions

Paleontological geosites in Liaoning Province are distributed across all 14 prefecture-level cities yet exhibit pronounced spatial heterogeneity. The majority of geosites are concentrated in regions such as Chaoyang, Jinzhou, Huludao, Dalian, and Benxi, forming five distinct spatial clusters across the province. Furthermore, paleontological geosites spanning from the Precambrian to the Cenozoic era demonstrate period-specific distribution patterns and varying degrees of spatial aggregation. This imbalance creates critical disparities in research focus, resource allocation, and geotourism development, necessitating region-specific management strategies to address geoheritage vulnerability and sustainable utilization.

The spatial distribution of paleontological geosites in Liaoning Province and their coupling relationships with multi-dimensional natural factors (elevation, slope-aspect, and stratigraphic exposure) and socioeconomic factors (road proximity, population density, and GDP per capita) provide a critical foundation for formulating targeted conservation and precise management strategies. The findings provide a theoretical basis for constructing a multi-scale conservation framework, which systematically addresses protection strategies across three distinct dimensions: the prefectural-level city scale (micro-operational perspective), the regional scale (meso-coordinative perspective), and the geological time scale (macro-value perspective).

At the prefectural-level city scale, the implementation of precision management, systematic surveying, and differentiated protection constitutes essential strategies for effective geoheritage conservation. Resource-rich cities (e.g., Chaoyang, Jinzhou, Huludao, Dalian, and Benxi) should prioritize developing paleontological geoparks, while others ought to establish specialized conservation systems, with focused attention on high-potential areas including low population density and economically underdeveloped regions and medium-to-high elevation sun-facing sedimentary exposures.

At the regional scale, it is crucial to enhance collaborative linkages and establish an integrated conservation network. The spatial correlation analyses identify high-potential (e.g., Western Liaoning PGAA) and high-risk zones (e.g., Western Liaoning, Eastern Liaoning and Liaodong Peninsula PGAAs), necessitating mechanisms such as a joint conference system for PGAAs (particularly for the Western Liaoning) to unify standards, monitoring, and data sharing across regions.

At the geological time scale, conservation strategies should integrate a macro-perspective of deep time to enhance scientific interpretation and long-term forward-looking conservation. Strengthened collaboration with research institutions is needed to clarify the evolutionary significance of geosites, while integrated research and tourism routes can translate conservation into sustainable educational and economic benefits, forming a virtuous cycle that supports ongoing preservation.