The research team, led by Lanyue Zhou of Ludwig-Maximilians-University Munich together with Hanqing Bao, Qian Wang of the Chinese Academy of Meteorological Sciences, and Junhao Li of the Northwest Institute of Eco-Environment and Resources, assembled panel data for 13 cities in western China covering the period from 2003 to 2023. Western China is an ideal natural laboratory for this question. The region spans arid basins, high plateaus, and mountain corridors, and its tourism economies range from heavily subsidized infrastructure hubs to remote destinations that depend almost entirely on scenic landscapes and cultural heritage. At the same time, the region’s development is strikingly uneven, meaning that two cities facing nearly identical climate hazards may have vastly different capacities to cope.

Methodologically, the study goes beyond the standard linear regression approach that dominates much of the climate-and-economy literature. Linear models implicitly assume that each additional unit of climate risk does the same amount of damage regardless of context. The authors instead applied panel threshold regression, a technique that searches the data for critical values of a switching variable, such as per capita GDP or a composite climate risk index, at which the relationship between climate exposure and tourism performance changes its sign or strength. They complemented this with interaction analysis to test whether development variables moderate the effects of climate exposure, and with contribution decomposition to quantify how much each factor, exposure, sensitivity, adaptive capacity, and readiness, contributes to observed tourism vulnerability and resilience.

The framework itself draws on the established vocabulary of climate vulnerability science. Exposure describes how much a city’s tourism sector is physically and economically in harm’s way from hazards such as drought, extreme heat, flooding, and sandstorms. Sensitivity captures how dependent the local economy is on climate-sensitive assets, including natural attractions and seasonal visitor flows. Adaptive capacity measures the resources, institutions, and infrastructure available to respond, while readiness reflects governance quality and the willingness to deploy those resources effectively. Vulnerability and resilience, in this framing, are not single numbers but emergent properties of how these four dimensions combine.

The headline result is a pair of thresholds with very different implications. When per capita GDP exceeds 21,357 CNY, the constraining effect of climate exposure on tourism is partly mitigated. Wealthier cities can invest in protective infrastructure, diversify their tourism products, and maintain governance systems that absorb shocks, and in some cases they manage to reconfigure exposure itself into a differentiated tourism resource, for example by marketing distinctive climate or landscape conditions that poorer competitors cannot replicate. In other words, money does not make climate risk disappear, but above this income threshold it changes what risk means for the tourism economy.

The second threshold tells the opposite story. When the composite climate risk index exceeds 1.49, the negative effect of exposure intensifies rather than fades. Under high-risk conditions, resilience and its supporting conditions, basic adaptive capacity and institutional readiness, become decisive for system stability. A city that crosses this risk line without strong adaptive foundations sees its tourism performance deteriorate disproportionately, because hazards no longer arrive as isolated events but as compounding pressures on water resources, transport networks, visitor safety, and seasonal reliability. The study’s contribution decomposition confirms that the weight of different factors shifts across these regimes: exposure dominates vulnerability in low-income, high-risk settings, while resilience indicators dominate outcomes once cities have crossed the development threshold.

Perhaps the most striking finding is that the 13 cities do not follow a single trajectory. The city-level analysis identifies three distinct transition pathways. Development-driven cities enhance resilience primarily through economic growth, using rising incomes to buy down their vulnerability. Tourism-driven cities improve their tourism reception capacity, expanding hotels, transport links, and service quality so that the sector itself becomes more robust even as hazards persist. Risk-response cities, typically those operating under high climate risk, become heavily dependent on basic adaptive capacity and readiness, meaning their stability hinges on governance and emergency response rather than on wealth or sector size. These pathways matter because they imply that a uniform national adaptation policy would misfire: a prescription that works for a development-driven city could leave a risk-response city dangerously exposed.

The study also delivers a methodological warning for anyone measuring tourism under climate stress. Tourism revenue, the authors found, is more sensitive than tourism intensity, a measure of visitor volume relative to local scale, to threshold shifts. Revenue-based indicators capture latent vulnerability and dynamic change that volume-based metrics smooth over, because spending patterns respond faster to perceived risk, infrastructure disruption, and shifting destination image than raw visitor counts do. For monitoring systems designed to give early warning of tourism decline, the choice of indicator is therefore not neutral: revenue-based measures are better at revealing when a system is quietly approaching a tipping point.

Why should readers far beyond western China care? The answer lies in the concept of threshold dependence itself. Much climate adaptation planning assumes proportionality, that incremental investments yield incremental protection. This study shows that in tourism systems, the returns to adaptation are discontinuous. A city just below the income threshold may gain little from the same adaptation spending that transforms outcomes for a city just above it, and a modest increase in composite climate risk can flip a manageable hazard regime into a destabilizing one. Similar threshold dynamics have been documented in coral reef tourism, ski economies, and coastal real estate, suggesting that the western China findings are a regional window onto a global pattern rather than a local curiosity.

The policy implications are correspondingly specific. The authors argue for threshold-sensitive and place-specific sustainability governance, meaning that adaptation planning should identify where each city sits relative to its critical thresholds and prioritize accordingly. For cities below the development threshold, the priority is building basic adaptive capacity and readiness before climate risk intensifies, since these are the factors that determine stability under high-risk conditions. For cities above it, the opportunity lies in converting exposure into differentiated tourism resources while continuing to strengthen resilience buffers. And for all of them, monitoring should track revenue-based indicators alongside visitor numbers to catch latent vulnerability early. As climate risk continues to rise across the world’s climate-sensitive regions, the study suggests that the most important question is not simply how risky the climate has become, but which side of the threshold a destination is standing on when the next shock arrives.

Subject of Research: Threshold effects of climate risk and economic development on tourism resilience in western Chinese cities

Article Title: Climate risk, uneven development, and threshold-dependent tourism resilience in selected cities of western China

Article References: Zhou, L., Bao, H., Wang, Q., & Li, J. (2026). Climate risk, uneven development, and threshold-dependent tourism resilience in selected cities of western China. Discover Sustainability, 7(1), Article 1669. https://doi.org/10.1007/s43621-026-04551-5

Image Credits: AI Generated

DOI: 10.1007/s43621-026-04551-5

Keywords: climate risk, tourism resilience, threshold effects, western China, adaptive capacity, climate vulnerability, panel threshold regression, sustainability governance, tourism economics, climate change adaptation, uneven development, city-level panel data

The study, authored by Chenchen Wu of Anhui Normal University in China, departs from conventional econometric practice in a way that matters for how governments interpret the numbers. Most analyses of the climate-tourism nexus rely on mean-based regression techniques, which estimate a single average effect and implicitly assume the relationship holds everywhere. Wu instead applies the Method of Moments Quantile Regression approach, a technique introduced by Machado and Silva in 2019 that estimates effects across the entire conditional distribution of the outcome variable. In practical terms, this means the analysis can distinguish between what drives tourism in destinations receiving relatively few arrivals and what drives tourism in destinations already flooded with visitors, rather than collapsing both into one misleading average.

The dataset draws on World Bank records for ten Asia-Pacific economies over nearly a quarter century, a period that encompasses the region’s extraordinary tourism boom, the global financial crisis, and the catastrophic collapse and recovery of international travel surrounding the COVID-19 pandemic. Four explanatory variables anchor the analysis. Carbon emissions capture the carbon-intensive character of economic activity. Ecological footprint, a measure developed by Wackernagel and Rees, quantifies the total biologically productive area an economy consumes, making it a broader indicator of environmental degradation than emissions alone. Trade openness reflects a country’s integration into global commerce, and gross domestic product stands in for overall economic scale.

The results on carbon emissions are perhaps the most counterintuitive. Across the lower, median, and upper-median quantiles of the arrivals distribution, carbon emissions show a positive and statistically significant association with tourism arrivals. This may seem paradoxical at first, but it is consistent with the idea that in developing and middle-income destinations, carbon-intensive industrial activity and tourism growth often rise together as part of the same expansionary economic process: more flights, more hotels, more energy consumption, and more visitors arriving in lockstep. Yet at the very highest quantile of arrivals, among the region’s most heavily visited destinations, the carbon effect loses statistical significance entirely. Carbon-intensive activity, the study suggests, matters most for destinations seeing low to medium volumes of arrivals, while at the top of the distribution other forces dominate.

The ecological footprint tells a much darker and more uniform story. Unlike carbon emissions, the ecological footprint is negative and statistically significant across every quantile examined. This means that general environmental degradation systematically suppresses international visitation regardless of whether a destination currently receives few tourists or millions of them. The implication is sobering for tourism-dependent economies: while the carbon-tourism link may weaken as destinations mature, the broader erosion of the natural capital that attracts visitors in the first place, clean beaches, healthy reefs, intact forests, breathable air, exacts a penalty everywhere. Environmental degradation, in other words, is not a problem that afflicts only struggling destinations; it is a structural threat that follows an economy up the entire arrivals ladder.

Trade openness emerges as the single most powerful positive driver in the analysis. Its effect is statistically significant across the distribution and peaks at the median quantile, confirming the role of commercial connectivity as a critical engine of cross-border tourism mobility. This finding aligns with a long tradition in the tourism economics literature, including gravity-model studies showing that transport infrastructure and economic linkages facilitate international visitor flows. The mechanism is intuitive: economies deeply embedded in global trade networks accumulate the airports, airlines, business travel, familiarity, and visa-friendly relationships that make leisure travel easier. Business trips seed future holidays, and trade agreements quietly double as tourism agreements.

GDP, by contrast, shows a positive association with arrivals throughout the distribution, but the marginal effect shrinks at higher quantiles. Economic growth fuels tourism most powerfully in destinations that are still building their visitor base; among the region’s tourism giants, additional GDP buys progressively fewer incremental arrivals. This pattern echoes the tourism-induced environmental Kuznets curve hypothesis explored in prior literature, in which the environmental and economic dynamics of tourism shift character as destinations move through different stages of development. It also carries a warning for mature destinations: growth alone will not sustain arrivals if the environmental foundations of attractiveness are allowed to decay.

Methodologically, the study is careful about the pitfalls that plague panel data in environmental economics. The analysis builds on a battery of diagnostic tests drawn from the modern econometrics literature, including panel unit root tests that account for cross-sectional dependence, slope homogeneity tests in the tradition of Pesaran and Yamagata, and cointegration testing to establish long-run relationships among the variables. The Method of Moments Quantile Regression framework then allows slope coefficients to vary across quantiles while handling the heterogeneity that ordinary least squares would average away. The author reports that the distributional differences are substantial enough that a single mean-based estimate would obscure, and in places invert, the true picture.

The policy implications are where the study becomes genuinely provocative. Because the climate-tourism relationship varies systematically across the arrivals distribution, Wu argues that environmental and tourism policy should be differentially designed according to a country’s relative position in the global tourism hierarchy. Emerging destinations with low to moderate arrivals, where carbon emissions and tourism still move together, face a window of opportunity to decouple the two before carbon-intensive lock-in takes hold. Mature, high-volume destinations, where the carbon effect fades but ecological footprint damage persists, should prioritize environmental restoration and carrying-capacity management over sheer growth. And for economies in the middle of the distribution, where trade openness exerts its strongest pull, the evidence suggests that commercial integration and environmental protection are complements rather than rivals: the connectivity that draws visitors is only worth as much as the environment they come to see.

The broader stakes are hard to overstate. Tourism contributes substantially to global carbon emissions, with recent research in Nature Communications estimating the drivers of tourism’s climate footprint, and the sector is simultaneously among the most vulnerable to climate impacts, from coral bleaching to heat extremes. The new findings add a distributional dimension to this feedback loop: environmental degradation does not merely threaten tourism in the abstract, it threatens it unevenly, punishing destinations at every level of development while the carbon-tourism coupling that fuels the problem is strongest precisely where destinations can least afford to manage it. For the ten Asia-Pacific economies studied, and for tourism-dependent nations everywhere, the message is that averages lie. The future of tourism will be written not in a single regression coefficient but across an entire distribution, and the destinations that read their own position carefully may be the ones that keep their beaches, forests, and visitors intact.

Subject of Research: Distributional analysis of how carbon emissions, ecological footprint, trade openness, and GDP affect international tourism arrivals in Asia-Pacific economies from 2001 to 2024

Article Title: Carbon, carrying capacity, and connectivity: Distributional evidence on the climate change-tourism nexus in Asia–Pacific economies

Article References: Wu, C. (2026). Carbon, carrying capacity, and connectivity: Distributional evidence on the climate change-tourism nexus in Asia–Pacific economies. Theoretical and Applied Climatology, 157(10), Article 681. https://doi.org/10.1007/s00704-026-06612-1

Image Credits: AI Generated

DOI: 10.1007/s00704-026-06612-1

Keywords: climate change, tourism economics, carbon emissions, ecological footprint, quantile regression, trade openness, Asia-Pacific, international tourism arrivals, environmental degradation, GDP, sustainable tourism, econometrics