The integration of large-scale Low Earth Orbit satellite constellations (hereinafter referred to as "LEO constellations") and artificial intelligence (AI) technology presents a historic opportunity for a paradigm shift in geography research, heralding a new era for geography to evolve from qualitative geography, quantitative geography, and digital geography into the "LEO constellation-AI-driven Geography". Under this framework, future geographic research can rely on the high spatio-temporal resolution monitoring data provided by LEO constellations to accurately capture the high-frequency dynamic changes of geographic elements at multiple scales, particularly at the global scale. By coupling physical models with AI, it becomes feasible to conduct simulation experiments on the complex interactions between natural and human elements, system states, and interface changes. This will facilitate a deeper understanding of core geographic issues such as variable coupling, multi-process cascading effects, and teleconnection mechanisms. To propel "LEO constellation-AI-driven Geography", there is an urgent need to establish a new-generation data acquisition and sharing platform relying on LEO constellation, seamlessly creating a "dynamic map" of global geographic resources and elements. Additionally, a geographic process simulator that couples physical models and AI needs to be developed to intelligently simulate and predict changes and impacts of geographic elements and landscapes.
Vegetation on the Qinghai-Xizang Plateau exhibits high-altitude and vertical zonation distribution characteristics, which pose significant challenges for fine-scale vegetation classification based on remote sensing. A major issue is the limited separability of remote sensing features among certain vegetation types, necessitating the effective integration of additional non-remote sensing features to improve separability. To address this problem, the present study developed a novel method for fine-scale vegetation remote sensing classification by progressively incorporating coarse spatial resolution vegetation and environmental features. This approach aims to improve both the accuracy and precision of classification. The new method comprises three primary components. First, vegetation and environmental features that substantially enhance vegetation classification and exhibit distinct feature differences are selected. These features are then used to calculate the prior probabilities for each class through a generalized additive model. Concurrently, machine learning classification with remote sensing features is employed to obtain the posterior probabilities for each class. Finally, by applying the Bayesian algorithm, the prior probabilities derived from coarse spatial resolution data are employed to adjust the posterior probabilities obtained from high spatial resolution data, resulting in refined classification outcomes. The method was rigorously tested and applied to the Qilian Mountains, Yellow River Source Area, and Hengduan Mountains on the Qinghai-Xizang Plateau. Sentinel-2 remote sensing data with a spatial resolution of 10 m, vegetation and environmental data with spatial resolutions ranging from 90 m to 10000 m, and ground survey data were utilized. The fine-scale vegetation classification results with a spatial resolution of 10 m were achieved. Compared to using only remote sensing features, the new method improved classification accuracy by 8% to 24%. This new classification method provides effective technical support for improving the accuracy and precision of vegetation classification and offers significant reference value for fine-scale vegetation classification on the Qinghai-Xizang Plateau and similar regions.
Runoff observation uncertainty is one of the key unresolved issues in the field of hydrology. Current studies mainly focus on the uncertainty sources and their impacts, but the effects of observation uncertainty on changes of entire flow regime characteristics are still rare. This study collects daily runoff observation series from 1971 to 2020 at five hydrological stations in the water conservation zone of the Yellow River (Huangheyan, Tangnaihai, and Lanzhou stations in the Yellow River Source Region, Xianyang Station in the Weihe River, and Heishiguan Station in the Yiluo River). Changes in 16 metrics from five main flow regime characteristics (magnitude, frequency of events, variability, duration, and timing) are detected by the trend tests, and the effects of observation uncertainty on trends of flow regime metrics are evaluated by adopting a normal distribution error model and some metrics, i.e., the uncertainty width, significance change rate of slopes, coefficient of variation, and degree of deviation. Results showed that: (1) At all the stations, flow regimes showed significant increases in the low flow magnitude, and significant decreases in the high and average flow magnitude, variability and duration. At the Heishiguan Station, the magnitude, variability and duration metrics significantly decreased, while the frequency metrics significantly increased. At the Xianyang Station, the low flow magnitude and timing metrics significantly increased, while the high flow magnitude, frequency and variability metrics significantly decreased metrics. In the Yellow River Source Region, the low flow magnitude and high flow timing metrics significantly increased, while the low flow frequency, high flow magnitude and variability metrics significantly decreased. (2) Observation uncertainty considerably affected the trend changes of 28.75% of total flow regime metrics at all the five stations. Among these, the trends of 11.25% of total metrics changed from significance to insignificance, and those of 17.5% of total metrics changed from insignificance to significance. For the rest metrics, the trends remained the same, i.e., significant trends (18.75%) and insignificant trends (52.50%). (3) Observation uncertainty had the greatest impacts on frequency metrics, especially at the Xianyang Station, followed by the duration, variability and timing metrics. The magnitude metrics were impacted least.
Projecting changes in flow regime under climate change and reservoir operations is crucial for safeguarding water security, preserving biodiversity and restoring aquatic ecosystems of a basin. Leveraging bias-corrected climate model data, this study utilized the Soil and Water Assessment Tool (SWAT) with an improved reservoir module and Indicators of Hydrologic Alteration to project the impacts of climate change and reservoir operations on the flow regime of the Yangtze River basin in the 21st century. The results showed that: (1) The improved SWAT model outperformed the original version, enhancing the simulation accuracy of daily streamflow and hydrological indicators at representative stations. The NSE (Nash-Sutcliffe efficiency coefficient) and KGE (Kling-Gupta efficiency coefficient) of simulation and observation of daily streamflow increased by 0.01-0.26 and 0.01-0.08, respectively. (2) Climate change is projected to increase streamflow, in which the streamflow increases significantly from January to April (the maximum increase is 17%). Meanwhile, climate change increases the variability of pre-flood streamflow (the maximum increase is 18%), the frequency of extremely high pulse, the duration of extremely high pulse and the rise rate of flow. (3) Reservoir operations effectively replenish dry-season flows (the maximum increase is 67%), reduce the frequency of extremely high pulse (the maximum decrease is 16%), and shorten the duration of both extremely high pulse (by -10% to -4%) and extremely low pulse (by -51% to -38%). Reservoir operations improve the stability of flow. While upstream reservoir operations can partially counteract climate-induced streamflow changes, their regulatory capacity is constrained by operational rules and total storage capacity. Therefore, adjusting reservoir operation strategies in response to the impacts of climate change on streamflow becomes imperative.
Water provisioning in the Asian Water Tower (AWT) is crucial to meet the demand of economic development in basin countries, whereas both climate change and economic development are altering the balance of water supply and demand. Previous researches mostly focus on the problem of how the availability of upstream water resources affects downstream uses, but the benefits to economic development are yet addressed. On the basis of hydrological runoff data in the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP), we developed an index representing the degree of water stress mitigation by the AWT, and established a production-oriented model using Adaptive Regional Input-Output (ARIO), for the evaluation of economic benefit and its relationship with the degree of water stress mitigation. The results show that the water stress mitigation index of basin countries exhibits dependency on geographical location, and the Central Asian countries display as a core area with higher mitigation index and temporal variation. Meanwhile, the overall economic benefit of the AWT to mitigate water stress reaches as high as 13.06% of Gross Domestic Product (GDP) of all basin countries. The benefit exhibits a significant spatial heterogeneity, with the beneficiary countries mainly concentrated in Central and South Asia, exceeding up to 80% of GDP, such as Bhutan, Nepal, Tajikistan and Kyrgyzstan. Importantly, the benefit not only depends on the degree of water stress mitigation, but also on economic structure, notably demonstrated by Bangladesh. Furthermore, the marginal economic benefit of water stress mitigation by the AWT increases with the degree of mitigation. This implies that the reduced water supply will have a much greater impact on the countries highly benefited from the AWT, especially in the low developing countries. The disparity between countries that benefit the most and least, respectively, can reach up to 2.63 times. As mentioned above, the low developing countries receive the greatest economic benefits when there is an increasing water supply of the AWT. Therefore, the greater significance of AWT is illustrated by its role, not only in mitigating water stress, but also in advancing economic development in the basin countries. In this regard, the basin countries are suggested to take joint actions, so as to safeguard the AWT, and establish a long-term comprehensive water resources management that takes into account the balance of supply and demand. This study can provide evidences for promoting sustainable development of the economic circles around the AWT, especially in low developing countries.
As a conceptual framework for understanding contemporary sustainability challenges, the metacoupling framework emphasizes the dynamic interactions and multi-regional connections within human-environment systems. Utilizing this framework to investigate the interactions between water supply and demand within and between regions can provide theoretical support for the sustainable management of water resources. In this study, we applied the environmentally extended input-output method and the Wang, Wei, and Zhu (WWZ) value-added decomposition method to quantify the metacoupling intensity and connections in the sending-receiving and spillover systems of China's virtual water trade. We also assessed the impact of virtual water trade on various regions. The results indicate the following: (1) Internal consumption and trade dynamics (2012-2017): The average internal consumption of virtual water within a province accounted for 78.05% of the total virtual water volume. Provinces primarily engaged in long-distance virtual water trade, which was on average 4.96 times the volume of peripheral virtual water trade. The manufacturing and agricultural sectors were the largest consumers of virtual water. (2) Spillover systems and trends: Upstream and downstream virtual water spillovers associated with the spillover systems exhibited an increasing trend, averaging 46.28% of the total traded virtual water volume. Jiangsu and Jilin produced the highest upstream virtual water spillover, whereas Xinjiang and Heilongjiang produced the highest downstream virtual water spillover. (3) Sectoral drivers and spillovers (2017): In 2017, the manufacturing and construction industries were the primary drivers of the upstream virtual water spillover, whereas the manufacturing and agricultural sectors were the primary drivers of the downstream virtual water spillovers. Long-distance trade generated a greater amount of both upstream and downstream virtual water spillovers. (4) Contribution to SDG 6.4: Virtual water trade contributed to a 5.75% improvement in achieving China's SDG 6.4 target, with long-distance trade contributing more significantly than peripheral trade. The virtual water trade was the most beneficial in alleviating water stress in economically developed provinces with poor water resource endowments, but it had negative impacts on some economically underdeveloped provinces.
Tropical cyclone activity has undergone significant changes under the impact of the global warming since the 20th century. However, the characteristic and trend changes of landing tropical cyclones over China still need to be further clarified. We conducted an analysis of the spatiotemporal characteristics and trends of landing tropical cyclones over China from 1949 to 2022 using the dataset of the best tracks of tropical cyclones from the China Meteorological Administration. Additionally, we explored the influences of El Niño-Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO) on landing tropical cyclone activities. The results indicate that: (1) The annual average number of landing tropical cyclones over China is approximately 8.85, showing a significant decreasing trend, and the decreasing range becomes larger with lower latitude overall. However, both the proportion of landing tropical cyclones to the total number and the percentage of higher intensity tropical cyclones increase. The landing tropical cyclones from July to September account for 76.3% of the total, in which August experiences the most. (2) The landing positions of tropical cyclones in China are mainly concentrated between 18°N and 26°N, accounting for approximately 88.2% of the total, and the landing frequency shows a sharp decline in the regions north of 30°N. The central landing position of tropical cyclones has shifted significantly northwestward, moving closer to China. Compared to 1949-1969, the central generating position from 2010 to 2022 shifted 4.5 degrees westward and 2.0 degrees northward. (3) There is a correlation between ENSO and the frequency variation of generating tropical cyclones in the Northwest Pacific and landing over China. El Niño promotes the formation of strong tropical cyclones and leads to a more southeastern bias in the generating positions of landing tropical cyclones, while La Niña has an opposite effect. The PDO also affects the tropical cyclones to a certain extent. During the warm phase of PDO, the generating position of tropical cyclones is westward and the number is less than that in the cold phase. This study further clarifies the changing trends and characteristics of the landing tropical cyclones over China since 1949. It also highlights the impacts of ENSO and PDO on tropical cyclone activities. The findings can serve as a scientific basis for conducting simulations and assessments of tropical cyclones and for disaster prevention and mitigation efforts.
The East Asian monsoon plays a crucial role in linking the climate systems of the Northern and Southern Hemispheres, affecting the lives of hundreds of millions of people. There, understanding the evolutionary traits from previous warm periods, such as the deep-sea oxygen isotope stage 5 (MIS 5, 130-70 ka BP), is crucial to comprehending the pattern of climate change in Asia and the mechanisms causing it. The MIS 5, with its lengthy duration and significant amplitude of climatic changes, is the warmer period closer to us and has been extensively utilized in studies of Late Pleistocene climate change. Here, we report new rock magnetic, environmental magnetic, and diffuse reflectance spectroscopy from the Huanxian loess-paleosol sequence from 155 to 20 ka BP, north-central Loess Plateau to study the relationship between monsoon precipitation and temperature changes during MIS 5. Rock magnetics results indicate that the magnetic minerals of the loess-paleosol sequence in Huanxian are dominated by pedogenic fine-grained magnetite/maghemite, with grain sizes ranging from superparamagnetic (SP)/single-domain (SD) to small pseudo-single domain (PSD) sizes (i.e., from <30 nm to ~1000 nm). The environmental magnetic and diffuse reflectance spectroscopy results indicated that the Loess Plateau had higher precipitation and temperature during MIS 5 than MIS 2-4, and MIS 6, suggesting a coupled relationship between them in the geologic period on the orbital (glacial-interglacial) time scale. During MIS 5, it is discovered that there was a decreasing trend in temperature but not in precipitation, indicating that the trend of precipitation and temperature in the north-central of Chinese Loess Plateau were decoupled. Our research demonstrates that temperature and precipitation in the study area have always had a coupling and decoupling relationship. Combining a detailed land-sea comparison, we suggest that both monsoon precipitation and temperature changes are influenced by ice volume and atmospheric CO2 concentration in the Northern Hemisphere. Additionally, monsoon precipitation is also influenced by the 20000-year variation in summer solar radiation.
The valley roads at the northern foothills of the Qinling Mountains deeply cut the main chain of the Qinling Mountains, and are the "natural window" to gain insight into the geology, geomorphology, ecology and hydrological evolution of the Mountains. In this paper, the spatial distribution pattern and development characteristics of the valley roads at the northern foothills of the Qinling Mountains (Shaanxi section) were obtained by using Google Earth remote sensing image data, DEM elevation data and water system vector data, combined with ArcGIS, GlobalMapper multimodal spatial data identification and interpretation. The inheritance relationship between the valley roads at the northern foothills of the Qinling Mountains and the Guanzhong river system was explored. The following understandings were obtained: (1) According to the geomorphological characteristics of the valley roads, the length, pedestrian and water channel, the statistical standard of the valley roads was put forward. Combined with contour terrain and digital elevation model, the zoning interpretation is carried out, and the actual number of valley roads is 206. (2) Based on the visibility of remote sensing interpretation and the endowment water system in the valley roads, 76 representative valley roads were selected for statistical analysis. According to the length (≥15 km, 5-15 km, <5 km), the large, medium and small valley roads were divided. The Xi'an section is the most densely developed mainly for medium valley road (44.74%). (3) The direction of the valley roads is mostly N0°-NE30°, approximately perpendicular to the strike of the Fenwei Fault. The elevation of Yukou is normally distributed, concentrated between 500-800 m. The spacing between the valley roads is the rhythmic distribution of "small valley roads in big valley roads". (4) The elevation of the middle line of the valley roads is mainly distributed at 600-1500 m, the slope is mainly concentrated at 0°-20°, the width of the valley entrances are mainly at 30-90 m, and the shape of the valley entrances are mainly U-shaped, V-shaped and inverted trapezoidal. (5) The valley roads at the northern foothills of the Qinling Mountains and Weihe water system have a close inheritance relationship. Six of the "Eight River Systems in Chang'an" originated from the northern foothills of the Qinling Mountains, characterized by multi-source confluence. The purpose of this study is to provide detailed background data for the study of the most iconic geological and geographical labels in the Qinling Mountains with valley roads.
The degradation of surface water resources, particularly in lakes and wetlands of arid regions, poses the risk of salt and dust storms. However, the impacts of the diminution of surface water on the spatiotemporal dynamics of regional aerosol patterns, especially in northwest China in the context of global environmental change is still unclear. Utilizing MODIS surface reflectance data (MOD09A1), this study introduces a dynamic threshold-based methodology for water body identification using multiple remote sensing indices. This framework is coupled with Multiangle Implementation of Atmospheric Correction (MAIAC) Aerosol Optical Depth (AOD) product to systematically assess the temporal dynamics of both surface water bodies and AOD. Employing a partial least squares path modeling (PLSPM) approach, this study comprehensively evaluates the mechanisms of surface water degradation on salt/sand and dust aerosols across various spatiotemporal scales. The results indicate that: (1) Surface water in the study area has exhibited a consistent upward trend from 2000 to 2020, with over 98% of resources localized in Qinghai and Xinjiang. (2) AOD levels in the northwest region demonstrate marked seasonality, peaking in spring and registering annual lows in autumn. High AOD concentrations are primarily in the Taklamakan Desert and the Loess Plateau. (3) In northwest China, as well as the typical regions of Qinghai Lake and Ebinur Lake featured by natural landscapes, fluctuations in surface water area directly and significantly correlate with aerosol growth. These findings have substantial implications for understanding salt and dust aerosol release and transmission mechanisms in arid regions while providing a scientific basis for resource management and ecological conservation.
The characteristics of paleochannel sediments in the Zoige basin of the northeastern Tibetan Plateau offer rich information regarding regional environmental evolution. This paper examined the representative paleochannel profile of the Yellow River Oxbow lake (NYQ-B) using systematic environmental magnetic analyses, and integrated with optical luminescence dating data to investigate the environmental evolution recorded in the paleochannel sedimentary profile. The primary magnetic minerals identified in the sediments of the profile NYQ-B were pseudo-single domain (PSD) and multi-domain (MD) magnetite. Magnetic minerals in the sediments of the profile NYQ-B were mainly influenced by weathering mechanisms, pedogenesis, and early diagenesis. The lowest concentrations occurred in the paleoflood and paleochannel deposits and the highest concentrations were found in the modern soil layer. The paleo-oxbow lake (A) deposits were formed from (3.24±0.26) to (2.96±0.24) ka, characterized by warm and humid climatic conditions. The higher χ and SIRM values and χARM/χ and χARM/SIRM values of these sediments indicate that they contain abundant fine sediment formed by strong weathering. The magnetic characteristics of the paleo-oxbow lake (B) deposits indicated that the period from (2.87±0.27) to (1.84±0.20) ka had dry and cold climatic features. The two paleoflood events recorded in the profile NYQ-B occurred at (2.96±0.24) to (2.87±0.27) ka and (1.84±0.20) to (1.70±0.16) ka, which correspond to the Holocene Climatic Optimum during the late dry period and Dark Age Cold Period, respectively. During these two periods, the climate was relatively cold and dry, leading to the general advancement of mountain glaciers on the Tibetan Plateau. We concluded that the contribution of ice and snow meltwater was relatively limited, while major flood events primarily resulted from heavy rainfall triggered by atmospheric circulation anomalies and interactions between middle- and low-latitude systems during climatic transition and abrupt change. These results enrich the application of environmental magnetism in the study of paleochannel sedimentary facies and contribute to a better understanding of the hydrological characteristics of the Yellow River source and their response patterns to global climate change.
Enhancing the carbon sequestration capabilities of terrestrial ecosystems by optimizing territorial spatial layout is a feasible approach to achieving the "carbon neutrality" goals. However, accurately identifying and characterizing the spatial differentiation of terrestrial ecosystem carbon sinks on a regional scale remains a challenge. As a result, effective territorial spatial management measures have yet to be established. To address this issue, this study uses Zhejiang province as an example, employing the Integrated Biosphere Simulator (IBIS) to simulate terrestrial ecosystem carbon sinks at 1-km grid resolution. Based on simulation results under different future climate change scenarios, we identify and delineate carbon sink conservation areas with adaptive features in response to climate change. The analysis results show that terrestrial ecosystems within carbon sink conservation areas exhibit a significant carbon sequestration advantage. Within these areas, the average Net Ecosystem Productivity (NEP) per unit area of forests, shrubs, and grasslands exceeds approximately 15%, 10%, and 5%, respectively. Under medium to high social vulnerability and radiative forcing scenarios, the carbon sequestration advantage of the terrestrial ecosystems within carbon sink conservation areas is further amplified. Additionally, by contrasting the carbon sink conservation areas with the existing Ecological Conservation Redline (ECR) in the study region, it is discovered that more than half of the areas have not yet been incorporated into the ECR. In comparison to the ECR, it is estimated that the total Net Ecosystem Productivity (NEP) within the carbon sink conservation areas is over 7 MtC annually, with the average carbon sequestration efficiency of forested regions surpassing the ECR by 15-22 g C/m2. This study underscores the importance and necessity of integrating and enhancing the consideration of ecosystem carbon sink function in optimizing territorial spatial patterns and existing ECR systems from a quantitative standpoint. Furthermore, we put forward corresponding policy recommendations.
The domestication of animals has provided humans with a stable food source. The wild yak is a unique species on the Qinghai-Xizang Plateau and plays an irreplaceable role in daily life of the Tibetan people. This study collected 291 localities of wild yaks on the Qinghai-Xizang Plateau and modeled the spatiotemporal evolution of wild yaks on the plateau every 500 years since the Holocene (11.0-3.0 ka BP) by Maxent. The accuracy of the simulation result reached above 0.9. The results indicated that the main environmental factors affecting the suitable habitat of wild yaks on the Qinghai-Xizang Plateau include the average temperature of the coldest quarter, the coefficient variation of precipitation, the precipitation of the wettest quarter, and the accumuation of the annual snowfall. From a spatiotemporal perspective, the distribution area was the smallest in the early Holocene (11.0-9.0 ka BP), mainly distributed in areas like Hala Lake, the eastern part of the northern Tibetan Plateau, the highlands and valleys in the southern Tibetan Plateau, and the highlands in the western Kunlun Mountains. During the early period of the middle Holocene (9.0-6.0 ka BP), the distribution area was located between that of the early and late Holocene, mainly in areas such as the Qilian Mountains, the western part of the northern Tibetan Plateau, and the southern Qinghai Plateau. In the middle to late Holocene (6.0-3.0 ka BP), the distribution area reached its maximum, mainly in areas such as the Qinghai Lake basin, the northern Tibetan Plateau, the Qiangtang Plateau, and the upper reaches of the Yangtze River and Lancang River. In addition, since the Holocene, the habitat of the wild yak has gradually expanded, possibly due to climate and environmental changes. Comprehensive analysis of the Human Living Environment Index and the Wild Yak Habitat Index revealed that the earliest potential domestication time of the wild yak is around 4.5-4.0 ka BP, with potential domestication areas including the Qinghai Lake basin, the upper reaches of the Yangtze River, Lancang River, and the Hengduan Mountains.
Ecological justice is an important element in realizing the harmonious coexistence of man and nature, and it is an effective path for promoting the coordination of ecological protection and livelihood development in natural-based tourism destinations. This paper constructs a theoretical framework and index system for the impact of ecological justice on farmers' livelihood resilience in nature-based tourism destinations. Taking Huangshan district of Anhui province as a case study, the dynamic impacts and threshold effects of ecological justice on farmers' livelihood resilience from 2000 to 2021 are systematically explored in terms of the three dimensions of ecological resource allocation justice, ecological rights enjoyment justice, and ecological responsibility sharing justice by using a time-varying parameter-vector autoregressive model and threshold regression model. The results show that: First, the medium-term positive impact of ecological justice on farmers' livelihood resilience is significant, and the short- and long-term impacts show "inverted V-shaped" time-varying pattern of alternating positive and negative impacts. The short-term positive impact of ecological resource allocation justice is the strongest, the medium-term positive impact of ecological rights enjoyment justice is the strongest, and the impact intensity of ecological responsibility sharing justice has the characteristic of diminishing marginal efficiency over time. Second, the moderating effects of ecological justice and ecological responsibility sharing justice on farmers' livelihood resilience are more affected by changes in the ecological construction environment, and ecological resource allocation justice is more sensitive to shocks from changes in the tourism development environment. Third, ecological justice, ecological resource allocation justice, and ecological rights enjoyment justice have a single threshold effect on farmers' livelihood resilience, with threshold values of 33.021, 51.377, and 33.021, respectively. Ecological responsibility sharing justice has a double threshold effect, with the most significant positive impact on livelihood resilience when the level of ecological justice crosses the second threshold value of 43.250.
The poverty among elder adults in rural areas of China poses a significant challenge to the achievement of shared prosperity. The concept of development resilience offers a novel perspective for examining livelihood and poverty, with scientific implications for quantitatively analyzing "silver-haired poverty" and proposing countermeasures of alleviating poverty among the rural elderly. This paper constructs a comprehensive framework for analyzing the relative poverty risk and development resilience of the rural elder adults, using a case study conducted in Shizhu, a mountainous county of Chongqing. Here, the paper evaluates the relative poverty risk and development resilience of the rural elder adults from 2010 to 2020, revealing their heterogeneity and spatiotemporal change characteristics. Moreover, the paper elucidates the relationship and feedback paths between relative poverty risk and development resilience, while explaining the determinants of development resilience. The results indicate that: (1) The relative poverty risk increased by 19.3% from 2010 to 2020. Health and natural disaster risks emerged as primary concerns, with market risk and health risk showing a noticeable upward trend. The rural elder adults engaged in traditional and guaranteed livelihoods faced a relatively high risk of poverty. The spatial distribution of relative poverty risk showed heterogeneity, with middle-altitude mountain areas experiencing a more pronounced increase in this risk. (2) The development resilience varied significantly among the elder adults across different livelihood types, as did the extent to which it increased. Rural areas witnessed the highest level of resilience development among elder adults engaged in multi-crop farming and casual labor. The spatial differentiation of development resilience in the study area was clearly evident, presenting a multi-gradient distribution pattern characterized by alternating high and low values from northwest to southeast. (3) Relative poverty risk coupled with development resilience given the dynamics status of the rural elderly, shaping both positive and negative feedback loops. (4) The development resilience of the rural elder adults was influenced by multiple factors at different levels. The community environment exerted an impact on the historical accumulation and subsequent development of development resilience. Policy formulation provided a safety baseline to withstand risk shocks. Intergenerational support played the role of a risk-sharing mechanism. The productive activities and psychological preferences of the elderly were the intrinsic driving forces for the formation of development resilience.
