The Tangjiahe Area of the Giant Panda National Park (TGPNP) is located in Qingchuan County, Sichuan Province in China. This conservation area was rated as an A-level nature reserve by the World Wildlife Fund for Nature (WWF), covering an area of 40,000 hectares and harboring a variety of rare and endangered species, such as giant pandas, snub-nosed monkeys, and takins. The total population of the surrounding communities is nearly 9,500, among which around 1,100 residents live within the PA boundary (Yang, 2022; Zhang et al., 2023). We select TGPNP as a case study for the following reasons:

Firstly, the mountainous landform of TGPNP shapes a scattered, isolated rural settlement pattern, referring that there is apparent heterogeneity in terms of environmental conditions. In addition, the scattered distribution of tourist attraction resources inside and outside the reserve may have effects on fairness perceptions of local residents (Mbise et al., 2021).

Secondly, the establishment of the TGPNP has imposed diverse impacts on surrounding communities. For example, there are relatively severe human–wildlife conflicts within and around PAs, leading to loss of livestock and crops. Additionally, the TGPNP has a strong tourist attraction to the public, thus bringing about tourists and employment opportunities to surrounding communities (Carvache-Franco et al., 2021).

Thirdly, the Administration of Tangjiahe Area (ATA) has carried out community-based conservation practices with local communities since 1978, such as joint fire prevention and infrastructure building supports, organizing co-management committees, and China Bee Breeding Cooperative. These diverse co-management strategies may have diverse impact on local fairness perceptions (Chen et al., 2022; Zhang et al., 2023).

We selected four villages (Luoyigou, Yinping, Weiba, and Dongqiao) in the gateway town (Qingxi) of TGPNP as our research site. Luoyigou is only village located within the boundary of TGPNP, whereas the other three villages are situated outside (Figure 1). The basic information of those villages is displayed in Table 1. Among them, Luoyigou village is located within the boundary of PA, with the most intensive human–wildlife conflicts. Local residents there mainly rely on agriculture and farmhouse tourism for a living. In contrast, community-based tourism in Yinping village has gained large popularity in the past 2 decades, providing substantial revenue for family inns and restaurants. Being adjacent to TGPNP, local residents in Dongqiao village are mostly engaged in agriculture, breeding, and migrant work, while those of Weiba village are mainly employed by a local stone processing company (Chen et al., 2022; Yang, 2022).

The survey sample comprised 361 residents in Qingxi Town. The proportions of men and women were 45 and 55%, respectively. Many surveyed residents were in elder age groups, with 70% aged 50 and over. About 66% respondents completed only primary or secondary school, and 22% never attended school. A vast majority of respondents (89%) have lived here for more than 20 years. Residents from Yinping and Luoyigou villages were less dependent on agriculture, with 31 and 48% of respondents made a living by agriculture, while the percentages for Weiba (66%) and Dongqiao (57%) were obviously larger. Residents with an average annual household income of less than 30,000 yuan accounted for the highest proportion in Dongqiao (80%), followed by Luoyigou (67%), Weiba (61%), and Yinping (54%) subsequently (see Supplementary Table C).

As depicted in Figure 3A, the perceived recognitional equity was obviously lower inside the border of TGPNP compared to the outside. The recognitional equity score of Luoyigou was the lowest (Mean = 3.23) compared to the remaining three villages out of the PA boundary (Dongqiao = 3.97, Weiba = 3.81, and Yinping = 3.86). With respect to the procedural equity, there demonstrated a gradual increase from northeast to southwest (Figure 3B). Weiba village situated in the southwestern tip of our research region scored the highest (PE = 2.51) among all. In terms of distributional equity, the core clusters were concentrated in the northwestern part of the study area (Figure 3C). Among four selected villages, the highest distributional equity score was found in Luoyigou village (Mean = 3.03), whereas the lowest was displayed in Dongqiao village (Mean = 2.43).

In addition, Figure 3D clearly presented the spatial distribution of perceived combined equity, which shared a similar pattern with procedural equity and gradually increased from the northeast to southwest. Dwellers in Weiba village tended to have more positive feelings for combined fairness, the average score of which was 2.77. By contrast, residents in Dongqiao village were frequently more disappointed with their lot (Mean = 2.41; see Supplementary Table D).

According to OLS, both the statistical values of the Lagrange multiplier (lag) and the Lagrange multiplier (error) were significant at the level of p < 0.01, presented in Table 3. The statistical values of robust LM (lag) and robust LM (error) were also significant, verifying the necessity to construct the spatial econometric models. The SEM with a higher fitting coefficient than the SLM was selected for subsequent analysis (Anselin et al., 2004). The LogL of the SEM model was larger than that of the OLS model, while the AIC and the slope change (SC) values were much smaller than those of the OLS model, indicating that the fitting performance of the SEM model had significantly improved compared with the OLS model. Thus, we used results from spatial regressions for subsequent analyses.

As shown in Table 4, local respondents living closer to the PA boundary and PA authority tended to develop more negative judgments on combined social equity, while those who residing near tourism spots were more likely to perceive adversely. For recognitional equity, the proximity with PA boundary and PA authority was negatively linked local residents’ fairness perceptions. As to perceptions toward procedural equity, surrounding communities with easy access to PA authority were more likely to perceive negatively, while those who with close connectivity to tourism spots behaved in an opposite way. In addition, positive judgments toward distributional equity were more likely to be found among local residents who lived far away from PA boundary and PA authority, and those who lived close to tourism spots (see Supplementary Table E).

Our research results indicate that the spatial distribution of perceived justice varies across three equity dimensions. While the perceived recognitional equity is largely divided by the PA boundary, the overall procedural justice obviously increases from the northeast to the southwest, and the optimism in local distributional perceptions tend to be clustered in the northwest direction. Despite numerous causes can be responsible for this heterogeneity, one innovative perspective of our study is to probe into its spatial predicators.

First of all, the accessibility to PA boundary was the most influential factor on recognitional perceptions among all spatial variables. This refers to the fact the remoteness from the PA boundary increases locals’ positive judgments for recognitional justice. For one thing, the land use and livelihood within the TGPNP boundary are strictly restricted and controlled according to the Regulations on the Management of Nature Reserves. This leads to overwhelming low scores of legal and traditional rights (RE3) and land ownership (RE4) among respondents living inside the PA, adversely impacting recognitional equity. For another, this is also partially due to the widely existed human–wildlife conflicts, threatening the livelihood of local residents (RE2) in close proximity to the TGPNP. In our interviews, villagers of Luoyigou complained that “wild boars came to have crops every day,” so they were heavily “affected by such wildlife accidents” for “losing the harvest throughout the year.” Those findings corroborate the results of Digun-Aweto et al. (2020) who presented that the proximity to national parks was most of the time accompanied by pessimism toward conservation, for those nearby communities were much more affected by crop losses caused by wildlife.

One unexpected finding of our study was that the accessibility to PA authority has a negative effect on residents’ perceptions across all dimensions of fairness. Proximity to the ATA can decrease local communities’ perceived attitudes to the fairness of conservation. This finding, though surprisingly, reflects the fact that the governance of ATA is questioned by locals. One potential cause is that PA authorities can impose restrictions on local access to natural resources, and even the act of designating PAs is seen by some as a form of land grabbing. Furthermore, there is also possibly that due to a lack of information-sharing channels and insufficient participatory approaches provided by ATA diminishes their reputation. It is observed in TGPNP that some dwellers living far away mentioned that they did not know what the ATA is, while those who living close complained the lack of participatory opportunities. As one interviewee grumbled, “If the ATA asked me to give suggestions or get involved with conservation, I’d love to; but the thing is they never asked me.” In that case, accessibility to the ATA, though adds up to the probability to get access to information, can enhance the likelihood to feel “marginalized” or “excluded from the decision-making process” at the same time. This finding is consistent with study of He and Wei (2022) in Qianjiangyuan National Park, where local farmers spending more time on the Internet tended to develop more negative attitudes toward conservation, illustrating the exposure to information could bring down feelings for justice.

Our research findings also indicated that the tourism accessibility imposes a positive-going impact on procedural and distributional fairness among local residents. The proximity with tourist attractions is frequently associated with higher scores of procedural and distributional judgments. This finding verifies a prevailing situation that local residents living close to tourist attractions are usually provided with more participatory chances and diversified livelihood choices, such as employment in hotels and hostels (Nelson and Makko, 2005; Kideghesho et al., 2007; Lobora, 2016; Kaaya and Chapman, 2017; Mashauri, 2017). In the case study of TGPNP, local residents in tourism-based villages were more likely to develop more positive perceptions toward distributional justice (Yinping and Luoyigou, mean = 2.93) than non-tourism villages (Weiba and Dongqiao, mean = 2.48). Mashauri (2017) argued that such inequitable benefit-sharing mechanisms can impede locals from developing supportive attitudes toward conservation in adjacent communities of Serengeti National Park.

However, our study found no impact of traffic accessibility on perceived social equity in the context of TGPNA. We authors suppose this is potentially due to the “Village to Village” project, a road building and improvement project arranged for all villages surrounding the TGPNA, ensuring relatively equitable access to transport infrastructure among local dwellers. He and Wei (2022) noticed that farmers with higher levels of satisfaction toward the basic infrastructure tended to have more positive conservation attitudes in Qianjiangyuan National Park. Such correlations deserve further exploration and verified in the context of TGPNP.

Our study, though attentively focusing on the spatial distribution of perceived fairness, never intends to deny the driving force of biodiversity attributes and policy drivers behind. From the perspective of spatial justice, human spatiality is socially constructed. This means that what is regarded as fair and just will vary from place to place, and may be affected and modified by the characteristics of a specific place decision (Plant, 1998). Therefore, the potential for seeking spatial justice by modifying human spatiality according to the characteristics of a particular place should exist in all situations. Soja (2010) argued that spatial justice and injustice can be seen as both outcome and process, as geographies or distributional patterns that are in themselves just/unjust as the processes that produce these outcomes. If PA authorities fail to implement fair and legal allocation policies and protection compensation mechanisms, it will not only weaken the communities’ trust in political institutions, but also undermine the level of spatial justice. In short, as mentioned above: local characteristics, interacted with structural factors such as governance policy systems, can exert significant impact on the rights and participation of residents, thus consequently forming the spatiality of injustice.

Given that “oppression and inequity are rife at all geographic scales” (Crampton and Elden, 2007), and the importance of seeking more equitably managed PAs we put forward some proposals for consideration by the PA authority. First, it is common to spot human–wildlife conflicts surrounding the TGPNA, yet the current compensation program for wildlife conflicts is merely undertaken in Luoyigou village, causing widespread feelings for injustice among the remaining villages. Therefore, we suggest extending this program to other adjacent villages, providing economic compensation for losses of local livelihoods damaged by the unexpected intruding of wildlife. Second, the restrictions imposed on land use within the TGPNP boundary have led to recognitional injustice, based on which we argue for initiating franchise mechanisms provided for inside dwellers to strive for more equitably distributed income. Third, the proximity with ATA can obviously reduce fairness perceptions, which is apparently due to opaque policies and low participation in the policy formulation process. Therefore, we call upon to optimize the community-based participation procedures and involve locals in the planning, management, and policy formulation process. Fourth, as the inclination of tourism policies lead to disappointed feelings among residents who live far away, we advocate for more spatially scattered tourism spots to balance income among communities. In particular, it is worthwhile to increase the added value of agricultural and forestry products (e.g., honey, persimmon, and fungus) in surrounding forest-dependent communities (Zhang et al., 2023).