Climate change has become inevitable, and appropriate measures and strategies must be adopted to adapt to its impacts (Intergovernmental Panel on Climate Change IPCC, 2023a). Against the backdrop of intensifying global climate change, urban, as major concentrations of population and economic activity, are experiencing increasingly pronounced climate vulnerability. Adaptation refers to reducing the adverse impacts and potential risks of climate change by strengthening risk identification and management within natural ecosystems, as well as economic and social systems, and implementing adjustment measures to optimize favorable factors while minimizing unfavorable ones (Intergovernmental Panel on Climate Change IPCC, 2023a; Intergovernmental Panel on Climate Change IPCC, 2023b). Climate adaptation focuses on reducing the vulnerability of urban systems to climate-related hazards, emphasizing both incremental adjustments and transformative actions to address actual or anticipated climate impacts. Its core objectives are to enhance adaptive capacity, flexibility, and equity, as reflected in reductions in heat-related mortality, the protection of ecosystem services, and the equitable access to resources for vulnerable groups. Urban development, by contrast, is typically guided by objectives of economic growth, infrastructure expansion, and social welfare improvement, with a primary focus on enhancing productivity and overall competitiveness. While climate adaptation and urban development can be synergistic, adaptation places greater emphasis on long-term risk governance and the management of uncertainty. This orientation requires prioritizing disaster prevention capabilities, institutional learning, and forward-looking planning, even when such measures yield limited or no immediate economic returns. The IPCC framework emphasizes adaptive capacity as a fundamental component for reducing climate vulnerability and achieving long-term adaptive capacity (Intergovernmental Panel on Climate Change (IPCC), 2023b). Several strategies have been proposed to enhance adaptive capacity (Shariatzadeh et al., 2021; Abawiera Wongnaa et al., 2024; Fernandez-Perez et al., 2024). Since the release of the National Strategy for Adaptation to Climate Change 2035, various regions have developed and implemented their own regional action programs. The policy framework for climate change adaptation has been progressively refined, and the capacity for monitoring, early warning, and risk management has been consistently strengthened. As a result, the adaptive capacity of key areas and regions has significantly improved, yielding notable progress in climate adaptation efforts.

Cities consume approximately 70% of the world’s energy and account for a substantial proportion of global greenhouse gas emissions, positioning them not only as major contributors to climate change but also as areas most directly affected by its consequences. The implementation of climate adaptation policies depends on specific administrative units, with cities serving as the smallest effective governance units for policy execution. The unique vulnerability and complexity of urban systems stem from pronounced spatial heterogeneity within a single city, factors such as topography, building density, and the exposure levels of different social groups can vary substantially. The high concentration of population and economic activity in cities means that extreme climate events can trigger cascading socioeconomic losses, necessitating targeted and context-specific assessment and analysis. Global- or regional-scale climate models and adaptation frameworks presented in IPCC reports often fail to provide direct guidance for concrete local actions. Adaptation measures must therefore be localized, requiring cities to design tailor-made solutions based on local data, resource endowments, and socio-cultural and institutional contexts. The vast majority of existing research focuses on climate adaptation at the city level, with comparatively limited understanding of processes operating at the metropolitan scale. Climate change impacts—such as flooding, heat islands, and water scarcity—frequently transcend administrative boundaries, necessitating coordinated governance at the metropolitan regional level (Clerc, 2021) However, current adaptation assessment and planning systems are predominantly structured around administrative units, making it difficult to capture interactions and risk transmission among cities within metropolitan areas (Gori Nocentini, 2024). Effective adaptation therefore requires strengthening coordination across administrative boundaries in metropolitan areas; yet existing governance frameworks often lack the institutional arrangements necessary to support such coordination.

Examining climate change at the urban scale reveals several concrete challenges. These include difficulties in data acquisition and scale matching; challenges in managing multisystem coupling and complexity; dilemmas related to governance structures and interest coordination; trade-offs between long-term adaptation and short-term development; and decision-making risks under uncertainty. At present, high-resolution data at the urban level remain scarce. The spatial resolution of regional climate models is often too coarse to capture urban microclimates, while the sparse distribution of local meteorological stations hampers fine-grained assessment. Urban infrastructure systems are highly interconnected, with energy, water, and transportation networks exhibiting strong interdependencies, such that failure at a single node can propagate throughout the system. Socio-ecological-technical systems are deeply coupled, requiring adaptation measures to simultaneously account for technical feasibility, ecological constraints, and social equity. Furthermore, climate adaptation involves multiple sectors, yet traditional administrative systems often suffer from fragmented responsibilities and conflicting objectives. Differing priorities among residents, businesses, and governments regarding adaptation measures, combined with urban-scale decision-making that is frequently influenced by electoral cycles and economic growth pressures, often lead to insufficient long-term investment. Significant uncertainties in urban climate projections, when contrasted with the long investment cycles of major infrastructure projects, can create fundamental dilemmas for climate adaptation.

The Chengdu–Chongqing region, western China’s premier economic hub, faces significant climate change challenges, including rising temperatures, altered precipitation regimes, and an increasing frequency of extreme weather events (Pickson and He, 2021; Zhao et al., 2020; Yin et al., 2024). The unique topographic conditions of Chengdu and Chongqing give rise to distinctly different patterns of climate exposure and mechanisms of vulnerability formation. The Chengdu Plain, characterized by low-lying terrain and dense hydrological networks, is primarily exposed to climate risks such as urban flooding, intensified urban heat island effects, and water-quality-related resource shortages. In contrast, Chongqing, a unique mountainous city, exhibits pronounced vertical climatic differentiation, frequent geological hazards, and strong compound effects of heatwaves and droughts, while infrastructure deployment is significantly constrained by topographic conditions. Existing generic assessment frameworks (e.g., global or national-scale models) often overlook these topography-driven regional variations, thereby failing to capture fundamental differences in dominant risk types and feasible adaptation pathways—such as plains relying on pipeline network expansion versus mountainous areas requiring enhanced ecological slope protection and vertical evacuation routes. They also fail to reflect the pronounced spatial heterogeneity in infrastructure vulnerability, particularly the heightened susceptibility of slope-built structures and road networks in Chongqing to extreme rainfall events. The two municipalities have overcome administrative barriers by signing multiple cooperation agreements on climate change adaptation, thereby laying a solid foundation for joint action. Furthermore, the launch of the “Ten Major Joint Actions” centered on the dual carbon goals, together with the establishment of platforms such as “Carbon Benefit Tianfu” and “Carbon Benefit Connect,” signifies an evolution in climate governance from singular environmental objectives toward comprehensive, incentive-driven policy frameworks. With respect to infrastructure and risk management, in response to pronounced risks posed by extreme heat, the region has strengthened meteorological monitoring, early-warning systems, and emergency response coordination, while also implementing timely adaptive measures to protect human health, such as designating cooling zones in subway systems and opening civil air-defense shelters. Concurrently, a series of major water conservancy projects aimed at ensuring long-term water security, which including the “Diverting Water from the Dadu River to the Min River” project, the West Chongqing Water Resources Allocation project, and climate-adaptive pilot projects which have entered the planning or construction phase. Efforts to enhance power grid interconnectivity and structural optimization have also commenced to address supply pressures during extreme weather events. In the realm of public health and livelihoods, people-centered adaptation practices are gradually diversifying through measures such as establishing service stations for outdoor workers, adjusting working hours during periods of extreme heat, and promoting stress-resistant crop varieties.

Despite these achievements, the Chengdu–Chongqing region continues to face significant challenges in building a comprehensive, balanced, and efficient climate adaptation system. Current cooperation predominantly focuses on the coordinated improvement of environmental quality rather than comprehensive climate adaptation. A fully integrated regional system for climate risk monitoring and assessment, data sharing, and joint accounting of greenhouse gas emissions and carbon sinks has yet to be established. The absence of an overarching “Regional Climate Adaptation Joint Action Plan” reflects a persistent path dependency that prioritizes mitigation over adaptation objectives. Second, the capacity of key systems to cope with extreme and compound climate risks remains fragile. The infrastructure systems of Chengdu and Chongqing have not yet fully kept pace with the rapidly evolving demands of climate adaptation. When confronted with novel compound hazards, such as flash droughts following floods and droughts occurring during flood seasons, water resource and energy security systems experience considerable strain. Existing practices demonstrate that Chengdu–Chongqing collaboration offers a novel governance approach for addressing transboundary climate risks and has achieved substantive progress in specific sectors. Nevertheless, significant gaps persist and require urgent attention, particularly with respect to the depth of collaborative governance, the robustness of infrastructure adaptive capacity, and the breadth and precision of adaptation measures. This case is capable of revealing the complexity and multidimensional characteristics of urban-scale climate adaptation, spanning climate impacts, vulnerability, and adaptation actions.

Developing an indicator system that reflects region-specific risks and adaptation pathways is not only a methodological necessity for localized assessment but also a prerequisite for scientifically elucidating the coupling mechanisms between regional climatic processes and urban systems. Therefore, this study innovatively operationalizes a regional scale-oriented linkage from the global adaptative capacity framework to national macro-level strategies and urban-level practices, and constructs an urban climate change adaptative capacity assessment index system applicable to regional collaborative governance contexts. This framework enables the systematic evaluation of urban adaptative capacity and the quantitative identification of strengths and weaknesses. It provides a scientific basis and decision-support for adaptative governance in the Chengdu–Chongqing region and other comparable urban agglomerations.

The Chengdu-Chongqing region, located in southwestern China, is a vital economic and geographic area encompassing the Sichuan Basin. It spans a vast area of Chengdu, the capital of Sichuan Province, and Chongqing, a major municipality. Known for its diverse natural conditions, the region experiences a subtropical monsoon climate, characterized by warm winters, hot summers, and abundant rainfall, which support its rich biodiversity and agricultural productivity. Economically, the Chengdu-Chongqing region serves as a growth engine for China, hosting thriving industries such as electronics and logistics, along with a rapidly expanding high-tech sector (Figure 1). The region is also a transportation hub, linking western China to global markets, underscoring its strategic importance in national development initiatives such as the Belt and Road Initiative and the Yangtze River Economic Belt. However, the region faces significant challenges due to climate change, including an increased frequency of extreme weather events such as heatwaves, droughts, and flooding, which threaten its urban infrastructure, agricultural outputs, and ecosystems. These vulnerabilities necessitate comprehensive adaptation and mitigation strategies to sustain the region’s socio-economic development and enhance its environmental adaptive capacity. The selection of Chongqing and Chengdu as case studies is not solely based on their status as major economic centres. More importantly, they serve as representative exemplars of two extreme and archetypal urban climate risk profiles. Chongqing exemplifies the high-complexity, cascading risk profile characteristic of mountainous river-valley cities, while Chengdu represents the highly systemic and diffuse risk profile of a plains city. Investigating how these cities build adaptive capacity under their respective risk constraints can not only deepen understanding of adaptation practices in the climatically sensitive region of Southwest China, but also provide a validated and differentiated toolbox of adaptation strategies, along with a decision-making framework, for other megacities and large cities worldwide confronting analogous topographic and climatic challenges.

This study is grounded in the risk and adaptation management framework proposed by the IPCC Sixth Assessment Report (AR6). To operationalize this macro-level framework at the urban scale in China, we systematically aligned it with the key urban development tasks outlined in the “National Climate Change Adaptation Strategy.” The strategy explicitly mandates enhancing urban climate risk assessment, developing climate risk maps, optimizing urban functional layouts, ensuring the safe operation of infrastructure, improving ecosystem services, strengthening flood defense and water supply security, and enhancing climate risk response capacities. These national-level directives provide a fundamental basis for translating the global conceptual triad of “impacts–exposure–vulnerability” into actionable and measurable urban indicators. Building upon this alignment, the study further localized, translated, and refined the indicator system. For example, urban heat-island intensity and urban waterlogging were incorporated as key exposure indicators. With respect to adaptive capacity (hard measures), the strategic requirement of ensuring the safe operation of infrastructure was operationalized through quantifiable indicators, including metro line length, total societal electricity consumption, the length of water supply pipelines in urban built-up areas, the length of drainage pipelines in urban built-up areas, urban gas coverage rate, park green area, and the green coverage rate of urban built-up areas. In accordance with the strategic requirement to enhance risk response capacities, indicators such as agricultural disaster prevention and mitigation expenditure, forestry disaster prevention and mitigation expenditure, flood control expenditure, geological disaster prevention and control expenditure, and the number of national-level disaster prevention and mitigation communities were incorporated.

Adaptive capacity is generally defined as the ability of a system—whether social, ecological, or socio-ecological—to anticipate, prepare for, respond to, and recover from climate-related impacts. This study utilizes the climate change adaptive capacity index system for the Chengdu-Chongqing region. To assess the climate adaptation capacity of Chengdu and Chongqing, ten primary indicators were selected to construct a comprehensive evaluation system that is consistent with the national strategic framework and closely aligned with the two cities’ distinct natural geographic characteristics, development trajectories, and key vulnerabilities. The ten primary indicators are designed to comprehensively capture urban climate adaptation capacity. Climate Change Impacts on Cities (A1) assess the overall extent to which Chongqing and Chengdu are affected by climate change, supporting decision-making in evaluating adaptation urgency and aligning with the IPCC risk assessment framework’s initial step of identifying key risks. Meteorological Disasters (A2) reflect direct climatic stress, recognizing that Chongqing is primarily characterized by extreme heat, whereas Chengdu is predominantly affected by heavy rainfall and droughts. Infrastructure (A3) corresponds to the core task of ensuring the safe operation of urban infrastructure outlined in the “National Climate Change Adaptation Strategy,” making the evaluation of infrastructural adaptive capacity critically important. Agriculture (A4), despite high urbanization levels, remains a foundation of regional ecological security and social stability, rendering agricultural adaptive capacity essential to food security. Forestry (A5) functions as a vital ecological barrier and carbon sink, directly supporting the enhancement of urban ecosystem service functions. Water Resources (A6), as the lifeline of urban development, address key adaptation objectives, with Chongqing facing water scarcity and uneven distribution, and Chengdu constrained by limited water availability and high exploitation intensity. Geological Disasters (A7) provide a targeted response to climate-sensitive geographical vulnerabilities, particularly in Chongqing, highlighting cascading and amplifying climate effects. Human Health (A8) captures climate-related health risks and evaluates public health systems’ capacity for early warning, response, and community protection, embodying a people-centered adaptation approach. Capacity to Invest in Adaptation Funding (A9) reflects the economic foundations and institutional soft power necessary to sustain long-term adaptation pathways. Disaster Prevention and Mitigation (A10) represent the most direct mechanism for reducing disaster-related losses and enhancing urban adaptive capacity. Together, these indicators comprehensively cover the key areas emphasized in the “National Climate Change Adaptation Strategy” and are closely integrated with the distinct municipal contexts of Chongqing and Chengdu, which form a logical closed loop encompassing risk drivers, sectoral exposure, and adaptive capacity, achieving precise localization of the adaptation framework and enabling a systematic diagnosis of adaptation strengths and weaknesses in Chongqing and Chengdu.

The framework proposed by the IPCC provides guidance for constructing the indicator system used in this study, which focuses primarily on the evaluation and analysis of urban adaptive capacity. Consequently, 37 indicators (as shown in Table 1) were comprehensively considered in evaluating adaptive capacity. A judgment matrix was constructed to determine the weights of each indicator factor presented in Supplementary Table S5, and “+” means positive indicator, “-” means negative indicator.

To ensure scientific rigor and methodological robustness in weight allocation, this study employed the Analytic Hierarchy Process (AHP) in combination with the Delphi method. The specific procedures were as follows: (1) Expert Panel Formation: Experts were invited from the fields of climate change research, urban planning, municipal engineering, public administration, and emergency management. All experts possessed relevant professional experience, ensuring the reliability of their judgments. Consultations were conducted with eight experts representing the fields of climate change research, municipal infrastructure, water resource conservation, ecological and environmental protection, agricultural cultivation research, geological disaster prevention and mitigation, and climate change adaptation research. Consistency tests were first performed on each expert’s judgment matrix, with all consistency ratios satisfying the criterion of CR <0.1. Subsequently, to evaluate the level of consensus within the expert group, the median, interquartile range (IQR), and coefficient of variation (CV) were calculated for the relative importance assigned to each indicator. Results indicated that, after the second round of expert consultation, the IQR for all primary indicators was ≤1. Compared with the first round, more than 85% of the indicators exhibited a reduction in CV exceeding 30%, demonstrating sufficient convergence and an acceptable level of expert consensus. Based on these results, the validated individual judgment matrices were synthesized using the geometric mean method to construct a group judgment matrix, from which the final aggregated indicator weights were derived. (2) Judgment Matrix Construction and Solicitation: Based on the established hierarchical structure, a questionnaire adopting the Saaty 1–9 scale was designed. Through two rounds of anonymous Delphi consultations, experts were asked to conduct pairwise comparisons of the relative importance of indicators at each hierarchical level. (3) Weight Calculation and Consistency Check: After questionnaire collection, the principal eigenvector method was applied to calculate the weight vector for each judgment matrix. Consistency checks were rigorously performed, and all matrices exhibited consistency ratios satisfying the accepted consistency threshold. (4) Aggregation of Group Decisions: The geometric mean method was used to aggregate individual expert judgments into a group judgment matrix, from which the final indicator weights were subsequently derived.

The geographic data in this article is derived from https://data.stats.gov.cn/, the data for the 37 indicators in this paper are mainly obtained from publicly available online sources, such as Statistical Yearbooks, Bulletin of Soil And Water Conservation, Statistical Bulletin on National Economic and Social Development, Bulletin of Water Resources and news reports (Supplementary Table S5). It should be noted that the 2025 Chongqing Statistical Yearbook, which contains critical data of 2024 for this investigation, has not yet been officially released. Consequently, the temporal scope of the current analysis has been delineated as 2017–2023 based on the most recent available datasets. While the 2024 data remain inaccessible at this stage, the research team will systematically update the database upon publication of subsequent statistical compilations. Importantly, this temporal limitation does not compromise the methodological rigor or the validity of the current findings, nor does it affect the timely submission of this manuscript.

Since positive and negative indicators represent different meanings. Therefore, different algorithms are adopted for data normalization. The formula for the standardization of indicators are as Equations 1, 2. B i = a i − a i , ⁡ min a i , ⁡ max − a i , ⁡ min (1) B i = a i , ⁡ max − a i a i , ⁡ max − a i , ⁡ min (2)

B _i represents the normalized value; a _i represents the observed value of the indicator; a _i,min represents the minimum observed value of the indicator; a _i,max represents the maximum observed value of the indicator.

For the whole evaluation system, the Climate Change Adaptive Capacity Index (CCAI) is as Equation 3. C C A I = ∑ i = 1 n = 37 B i × W i (3)

CCAI represents the Climate Change Adaptive Capacity Index; B _i is the normalized indicator value; W _i is the weight of each indicator.

For A1, the Climate Change Adaptive Capacity Index is as Equation 4. C C A I A 1 = ∑ i = 1 n = 2 B i × W i (4)

CCAI(A1) represents the Climate Change Adaptive Capacity Index of A1; B _i is the normalized indicator value; W _i is the weight of each indicator.

Similarly, for A2, the Climate Change Adaptive Capacity Index is as CCAI(A2), Equation 5. C C A I A 2 = ∑ i = 3 n = 6 B i × W i (5)

With the climate change adaptive capacity index system in Chengdu-Chongqing region, we evaluated the weight of driving factors of the first layer (A1∼A10) with judgment matrix (Franek and Kresta, 2014). 2,4,6,8 are values for compromise in judgement of importance between 1 and 3, 3 and 5, 5 and 7, 7 and 9 respectively. This judgment matrix and the weight are presented in Table 2. Similarly, the weight of driving factors of the second layer evaluated with the judgment matrix.

The specific division is shown in Table 3.

For the first indicator level, the contribution analysis method is as Equation 6. C A i = C C A I A i C C A I × 100 % (6)

C _Ai represents the contribution rate of each indicator of the first indicator level.

For the second indicator level, the contribution analysis method is as Equation 7. C i = W i × B i ∑ n = 1 37 W i × B i × 100 % (7)

C _i represents the contribution rate of each indicator; W _i represents the weight of the indicator; B _i is the standardized value of the indicator.

The adaptive capacity in Chengdu and Chongqing is quantified using an adaptive capacity index system, as shown in Table 4. Overall, the Climate Change Adaptive Capacity Index of both cities increased, with Chongqing experiencing a slightly greater increase than Chengdu.

The CCAI in the Chengdu-Chongqing region demonstrates an overall upward trend from 2017 to 2023, with greater volatility observed in Chongqing. The CCAI of both regions was relatively low in 2017 but gradually increased, with Chongqing reaching the highest value in 2023. This trend may be attributed to differing policy measures adopted by the two cities to address climate change and reflects disparities in resource allocation, technological application, and environmental governance.

In the summer of 2022, Chongqing experienced its most severe compound heatwave–drought event since the establishment of comprehensive meteorological records in 1961, posing a severe test of the city’s adaptive capacity. Nevertheless, the Adaptive Capacity Index for Chongqing continued to increase relative to previous years. This outcome can be explained by three interrelated mechanisms. First, the “stock release effect” of accumulated infrastructure adaptation was evident. Empirical results indicate that infrastructure (A3) accounted for 30.03% of total contribution in 2022. This increase was not incidental but reflected the delayed and concentrated release of effectiveness from long-term hardware investments, including riverbank improvement projects, slope stabilization, and aging pipeline upgrades. During the heatwave–drought response period, emergency measures included repairing and activating drought-resistant pumping stations, installing temporary pumping facilities, and extending more than 2,300 km of pipelines. These emergency response capabilities were fundamentally underpinned by cumulative prior infrastructure investments. Second, the “resource mobilization effect” of Adaptive Funding Capacity (A9) played a decisive role. The contribution of A9 increased to 23.10% in 2022, highlighting the central role of fiscal resources in emergency response. The allocation of 10 million yuan in drought relief funding directly demonstrated the performance of adaptive funding capacity under extreme temperature stress. Third, the “institutional response effect” of the emergency management system was fully activated. Daily drought assessments and disaster reporting mechanisms were implemented, 26 task forces were dispatched across 36 districts and counties, and nearly 900,000 officials and residents were mobilized for water delivery operations. This institutional capacity—captured by the disaster prevention and mitigation indicator (A10)—translated into basic livelihood security for 269,000 residents experiencing water shortages. Therefore, the Adaptation Capacity Index reflects cumulative, year-long capacity accumulation rather than short-term emergency response performance. The increase in the 2022 Adaptation Capacity Index thus coexisted with the occurrence of a historic extreme heat event. The former captures the concentrated release of accumulated infrastructure adaptation, fiscal mobilization capacity, and institutional effectiveness, whereas the latter reflects the unprecedented magnitude of climatic hazard intensity. It was precisely the sustained enhancement of adaptive capacity that prevented this record-breaking extreme heat event from escalating into a humanitarian catastrophe of comparable scale.

Chongqing’s CCAI surpassed that of Chengdu in 2020, a shift that coincided with the global outbreak of the COVID-19 pandemic. In 2020, Chengdu’s tertiary industry value-added growth rate declined sharply to 2.3%, down from 8.6% in 2019. Contact-intensive service sectors—including accommodation and catering, as well as cultural, sports, and entertainment industries—were hit particularly hard, with both recording negative growth. Tourism revenue fell by approximately 40% year-on-year. In contrast, Chongqing’s secondary industry demonstrated greater adaptation, with value-added growth reaching 4.9% in 2020, exceeding the growth of its tertiary sector (2.9%). This divergence indicates that Chengdu’s service-oriented economic structure was more vulnerable to short-term systemic shocks than Chongqing’s industry-based economy. Fiscal performance mirrored these structural differences. Chengdu’s general public budget revenue declined by 2.9% year-on-year (or stagnated at best), potentially increasing pressure on expenditure restructuring, particularly for disaster prevention and emergency management. By contrast, Chongqing maintained positive fiscal revenue growth of approximately 1%, reflecting relatively stable fiscal fundamentals. As local fiscal resources constitute the primary funding source for climate adaptation investments—such as water conservancy projects, sponge city construction, and environmental monitoring networks—these contrasting fiscal trajectories are highly consequential. Chengdu’s sharper economic downturn likely constrained its short-term fiscal flexibility, forcing the government to prioritize rigid expenditures such as livelihood support and emergency relief, while postponing or scaling back medium-to long-term adaptation capacity-building investments—the very dimensions captured by the CCAI. Chongqing’s more resilient fiscal base, supported by its industrial economy, enabled greater continuity in adaptation-related investment. Within the indicator system, variables closely tied to fiscal allocation—such as B9 (disaster prevention and mitigation expenditure as a proportion of fiscal spending) and B10 (science and technology expenditure supporting climate-related)—are particularly sensitive to annual budget adjustments. In 2020, Chengdu experienced a noticeable slowdown or decline in these indicators, whereas Chongqing’s corresponding metrics remained stable or exhibited modest growth. Overall, this divergence illustrates that macroeconomic adaptation constitutes a critical foundation for sustaining long-term climate adaptation investment. Short-term systemic socioeconomic shocks can interrupt the accumulation of adaptive capacity through fiscal transmission channels, thereby shaping inter-city differences in observed adaptation trajectories.

This study evaluates the annual contribution rates of key adaptive capacity drivers in Chengdu and Chongqing from 2017 to 2023, identifying trends and the underlying factors influencing these changes in Figure 2.

According to the research, A6 (Water Resources) and A8 (Human Health) are identified as the key factors influencing adaptive capacity in Chengdu and Chongqing. Based on the analysis of A6 (Water Resources) from 2017 to 2023 in Chengdu and Chongqing, it was observed that total water resources exhibited significant volatility and an overall declining trend, primarily influenced by natural factors (e.g., precipitation and temperature changes) and anthropogenic factors (e.g., urbanization, population growth, economic development, and water resource management). The significant decline in total water resources in 2023 is likely closely related to dry weather conditions and reduced precipitation during that year. This indirectly indicates that climate change will adversely impact the total amount of water resources, subsequently affecting the evaluation of cities’ adaptive capacity to climate change (Ricart et al., 2021). A city with well-developed emergency response mechanisms and adequate medical resources is more effective in minimizing human casualties and controlling the spread of disease during natural disasters triggered by climate change. The health infrastructure (e.g., hospitals, clinics, CDCs) and service capacity (e.g., accessibility of medical resources, emergency response capacity) are crucial in addressing the health risks posed by climate change (Cordiner et al., 2024). The adaptive capacity of the health system, including the ability to provide timely warnings and responses to climate-related disease outbreaks, as well as to offer effective medical assistance to affected populations, directly impacts the evaluation of the city’s overall capacity to adapt to climate change. A3 (Infrastructure) and A10 (Disaster Prevention and Mitigation) also significantly influence the evaluation results of adaptive capacity in Chengdu and Chongqing. Effective disaster prevention measures can reduce losses during disasters and enhance the adaptive capacity and robustness of cities, thereby improving their capacity to adapt to climate change. The performance of a city’s infrastructure (e.g., transportation, energy, communications) during extreme weather events directly influences its capacity to adapt to climate change. Cities with strong disaster prevention and mitigation capabilities typically reinforce and upgrade their infrastructure to ensure the maintenance of basic functions during a disaster (Ferdowsi et al., 2024; Fernandez-Perez et al., 2024; Romanach et al., 2024; Zhang et al., 2024). These measures not only enhance the city’s disaster prevention and mitigation capacity, but also strengthen its overall capacity to adapt to climate change.

Also, this study evaluates the contribution rates of 37 factors to Chengdu and Chongqing’s adaptive capacity to climate change from 2017 to 2023 (Figure 3).

Among the secondary indicators, B10 (Green coverage rate of urban built-up area), B15 (Cultivated land area), B22 (Total water resources), B28 (Number of health technicians per 1,000 people), and B32 (Adaptation financial input capacity) are found to significantly contribute to the capacity to adapt to climate change. B10 (Green coverage rate of urban built-up area) is strongly correlated with cities’ ability to adapt to climate change. High green coverage lowers urban temperatures through transpiration, mitigating the urban heat island effect and decreasing the risk of extreme heat. Vegetation absorbs pollutants, releases oxygen, improves air quality, and mitigates health risks to residents. Ecosystem services, such as green spaces and wetlands, enhance biodiversity conservation and rainwater regulation, reduce the risk of waterlogging and flooding, and strengthen cities’ overall capacity to adapt to climate change. B15 (Cultivated land area) is strongly associated with cities’ capacity to adapt to climate change. Cultivated land, as a vital ecosystem, provides ecological services, including climate regulation, water conservation, and soil preservation (Maher et al., 2025; Schröder et al., 2024; Effiong et al., 2024). A larger area of arable land can mitigate the urban heat island effect and reduce the impact of extreme heat events through vegetation cover and soil moisture regulation. Additionally, cropland can augment the green infrastructure of cities, improve their ecological adaptive capacity, and enhance their capacity to withstand natural disasters. Furthermore, arable land ensures food security for cities and mitigates the risk of food supply disruptions due to climate change, thereby enhancing cities’ overall capacity to adapt to climate change. On one hand, declining water resources can exacerbate the risk of urban water scarcity, threatening the stability and security of water supply systems and diminishing cities’ ability to adapt to climate change. For instance, droughts can deplete water sources, compromising the water supply to cities, while floods may pollute water sources, endangering the security of water supplies. On the other hand, changes in water resources can also impact urban ecosystems, economic development, and social stability, further diminishing cities’ capacity to adapt to climate change. Therefore, cities must strengthen water resource management, enhance the efficiency of water utilization, and bolster the adaptive capacity of water supply systems to improve their capacity to adapt to climate change. Sufficient health technicians can bolster the city’s capacity to provide medical care and increase its adaptive capacity to health risks induced by climate change. Health technicians help mitigate the negative impacts of climate change on public health, thereby enhancing cities’ ability to adapt to climate change. The availability of health technicians reflects a city’s ability to deploy resources and manage emergencies in response to climate change and is a crucial component of its adaptive capacity. Sufficient funding enables cities to undertake climate change adaptation projects in infrastructure development, ecological restoration, and technology research and development, thereby enhancing adaptive capacity. These funds can be allocated to retrofitting drainage systems and constructing sponge cities to manage extreme rainfall, or to vegetation restoration and wetland protection to regulate local climates and safeguard biodiversity.

The evolution of climate adaptation capabilities in Chongqing and Chengdu reveals distinct trajectories while simultaneously reflecting strong regional linkage dynamics. The two cities have gradually established an emerging regional climate governance framework, shifting from isolated self-improvement toward coordinated advancement.

Although the Analytic Hierarchy Process (AHP) offers notable advantages in integrating expert knowledge with localized cognitive insights, its reliance on subjective judgment also constitutes a fundamental methodological limitation. The composition of the expert panel exerts a direct influence on weighting outcomes. When experts share similar disciplinary backgrounds or institutional experiences, the resulting weights may amplify shared cognitive biases, thereby underrepresent certain dimensions of adaptive capacity and limit the framework’s ability to capture its multidimensional value structure comprehensively. While the construction of judgment matrices incorporates consistency tests (CR < 0.1) to ensure internal logical coherence, such procedures cannot fully resolve the inherent tension between individual cognition and collective consensus. This limitation is particularly salient in the context of adaptive governance, where unavoidable normative trade-offs—such as equity versus efficiency or short-term responsiveness versus long-term adaptation—are embedded within expert judgments. In these cases, weighting outcomes may implicitly reflect experts’ value orientations rather than purely objective assessments. Accordingly, AHP-derived weights should be interpreted as a form of “negotiated knowledge output” shaped by specific local contexts, institutional settings, and expert perceptions. Their scientific contribution lies not solely in the numerical weight values, but more importantly in their capacity to illuminate underlying tensions and priority structures among adaptation dimensions within localized decision-making environments. From this perspective, AHP functions less as a definitive optimization tool and more as a structured deliberative mechanism, providing a transparent point of departure for subsequent participatory planning, iterative evaluation, and adaptive policy adjustment.