The changing nature of the earth’s vegetation is a crucial dynamics in ecosystem processes. Here, we investigate the changes in global dryland vegetation over the past two decades and the effects of surface moisture and atmospheric water demand on greening and browning. Our results show that significant greening occurred across approximately 21

Evaluating the true efficacy of nature-based conservation interventions in climate-sensitive ecosystems is fundamentally confounded by macroclimatic noise. To resolve this, we constructed a counterfactual attribution framework across the Changtang Plateau (2001–2020) to effectively decouple the independent effects of grazing exclusion from climate variability across five core ecosystem functions. Our findings unravel two ecological mechanisms dictating functional recovery. First, fencing benefits are governed by an aridity-dependent time lag, while humid alpine meadows exhibit rapid functional saturation, resource-limited arid steppes require prolonged accumulation to safely surpass intrinsic restoration thresholds. Second, ecosystem functions displayed distinctly asynchronous recovery trajectories. While aboveground primary production initially surged and stabilized, subsequent belowground carbon sequestration systematically attenuated, reflecting delayed vegetation-soil feedbacks. Furthermore, we identified a key intensity-spatial decoupling across the regional landscape. Although fencing acts as an effective local amplifier for biologically driven processes (e.g., productivity and soil retention), its spatial footprint gradually contracts along the aridity gradient. Conversely, hydrological and aerodynamic regulations remained primarily disjointed from fencing interventions, highlighting a critical boundary where localized engineering is structurally insufficient to override broad-scale macroclimatic forcing. Ultimately, conflating these complex spatiotemporal dynamics under traditional static evaluations risks notable climate-driven false positives and protection-driven false negatives. By exposing these evaluation traps, this study underscores the need for an alternative paradigm shift from conventional one-size-fits-all policies toward spatially explicit, climate-adaptive management frameworks, establishing an evidence-based basis for optimizing global protected area networks in fragile cold-arid biomes.

Climate change-driven conservation strategies commonly project habitat availability but may not account for local adaptation among populations of the same species, which can influence prediction accuracy. Using the giant panda (Ailuropoda melanoleuca) as a case study, we developed a regional-scale species distribution model (SDM) and 33 population-specific local models to assess niche divergence and climate-induced habitat shifts (current vs. 2080-2100, SSP2-4.5). Comparisons between the two model scales, validated against observed habitat distributions, revealed clear differences in predicted habitat range, area, quality, and fragmentation among local populations. Specifically, regional-scale models predicted lower climate threats for 15 local populations, higher threats for 10, and did not identify suitable habitats for 8 populations, particularly those that were smaller and more isolated. These findings highlight the importance of incorporating population-specific climatic niche differentiation into conservation planning to improve the reliability of climate impact assessments and to guide population-level strategies for biodiversity conservation under future climate change.

Greening plays a key role in surface-atmosphere energy exchanges, exerting feedback on the climate. While several studies have examined greening feedback in forested regions, the effect on temperature feedbacks in global drylands remain poorly understood. In this study, we used a combination of satellite and reanalysis data to examine the global dryland greening trends over the past two decades and assessed the changes in the biophysical properties. Results show that the greening-induced changes in surface albedo affected net shortwave radiation; however, evapotranspiration linked to changes in soil moisture dominated the temperature feedback, accounting for approximately 54-83% of the feedback signal. This dominance was most pronounced for daytime surface temperature, where the evapotranspiration effect was up 66% more than that of surface albedo. Significant greening was associated with a decrease in daytime surface temperature of 0.53 and 0.8 °C/decade, while significant browning was associated with an increase of 0.86 and 1.32 °C/ decade. Although greening was significant over the global drylands, soil moisture availability strongly controlled evapotranspiration and its contribution to temperature feedback. As dryland continues to warm, moisture availability is important for plant functioning, which impacts the regional climate through surface and atmosphere feedback and, therefore, the sustainability of drylands.

Climate change affects biodiversity through multidimensional impacts, influencing not only shifts in habitat range but also changes in habitat quality. In this context, habitat area and bioclimatic velocity have become critical metrics for assessing species-specific vulnerabilities to climate change. Here, we assessed the extinction risk and exposure risk of giant pandas (Ailuropoda melanoleuca) based on habitat area and bioclimatic velocity, respectively, and examined the differences between these two risks to inform climate-adaptive conservation strategies. Our findings indicate that under the SSP2-4.5 scenario, degraded giant panda habitats are projected to total 13846.1 km2, with the Qinling (QL), Liangshan (LS), and Daxiangling (DXL) populations experiencing substantial habitat loss of 3790.4, 2722.8, and 1135.4 km2, respectively. Bioclimatic velocities across different populations range from -0.468 to 0.309 km year-1, with higher velocities observed in southeastern Minshan (MS) and Qionglaishan (QLS) and Liangshan (LS) regions, suggesting potential declines in habitat suitability and substantial challenges to population survival. Our results also reveal that while most populations exhibit consistent risk patterns when assessed by both habitat area and bioclimatic velocity, notable discrepancies remain. Populations with high extinction risk generally face high exposure risk; however, some populations with low extinction risk may encounter substantial exposure risk (e.g., DXL_A and MS_K). These findings highlight the limitations of relying on single-dimensional assessments of species' vulnerability to climate change, as evidenced by the variability in risk assessment outcomes. Therefore, integrating changes in both habitat area and bioclimatic velocity provides a more comprehensive understanding of species' vulnerability, reveals local adaptation mechanisms, and offers a robust scientific basis for formulating targeted climate-resilient conservation strategies.

In the face of accelerating biodiversity loss and ecosystem degradation, enhancing the effectiveness of ecological restoration under limited resources has become an urgent challenge. This study employed the snow leopard (Panthera uncia) as a case study to prioritize its restorable habitats through enhanced cost-effectiveness in the Sanjiangyuan region of the Qinghai-Tibet Plateau. A multi-scale framework was developed to identify restoration priority areas by integrating habitat suitability, vegetation degradation, and vegetation restoration potential (VRP). The aim was to scientifically identify areas of high ecological value for restoration while minimizing costs. We used species distribution models to predict snow leopard habitats and applied the Mann-Kendall test and Sen’s Slope trend analysis to assess vegetation change trends. To estimate vegetation’s ideal ecological state, the counterfactual matching was used to calculate VRP by comparing it with current conditions. Systematic conservation planning was then employed to delineate core and general restoration areas. Results show 17.8% of suitable habitats exhibit significant vegetation degradation, with 68.6% located in medium to high suitability areas. High-VRP areas are mainly concentrated in the central and southern parts of the habitat, indicating high restoration value. The identified restoration priority areas cover 30% of the study area, with core areas comprising 64% of this subset. The proposed framework achieves over 60% coverage of areas with high suitability, severe degradation, and high VRP, demonstrating its potential for maximizing ecological benefits and optimizing resource allocation. This study provides a scientific foundation and practical framework for formulating targeted restoration strategies and enhancing endangered species conservation.

Assessing the carbon sequestration capacity of regional ecosystems is essential for achieving carbon neutrality goals. However, existing research often fails to comprehensively evaluate the spatial-temporal dynamic changes of ecosystem carbon emission and absorption. This study introduces a Carbon Benefit Index (CBI) to assess carbon neutrality potential and classifies Shandong's 16 cities into four regions based on their carbon storage and emission profiles. We conducted an in-depth analysis of ecosystem carbon benefit in Shandong Province from 2000 to 2020 using the multimodel random forest ensemble method, which enhances the accuracy of carbon sink simulations across terrestrial ecosystems. Our results showed that from 2000 to 2020, Shandong's carbon emission increased by 1.45 x 108 tons (a 203.8% rise), while carbon storage decreased by 3.40 x 107 tons (a 2.05% decline). Compared to previous studies, our findings underscore the significance of both above-ground and below-ground carbon storage. Grey correlation analysis of land use, anthropogenic CO2 emission, and ecosystem carbon storage revealed that cultivated and forest lands were most significantly correlated with carbon storage, whereas built-up areas were most closely linked to carbon emission. The CBI analysis and classification of the 16 cities into four categories highlights the spatial-temporal heterogeneous of the carbon efficiency, and diverse roles cities play in the province's overall carbon balance, informing city-specific, targeted carbon reduction strategies. The study emphasizes the need for spatially differentiated, comprehensive carbon accounting to improve carbon efficiency. Based on these findings, we propose tailored low-carbon improvement strategies for different regions. This research not only contributes to existing literature by incorporating below-ground carbon storage but also provides valuable insights for policy and land management, with practical implications for promoting sustainable development and advancing efforts toward carbon neutrality.

Islands are critically important but inherently fragile due to their isolation and limited size. Climate change poses escalating threats to island biodiversity, comprehensive and spatially explicit assessments are still limited, hindering the development of targeted adaptation strategies. Here, we integrated species distribution modeling, inundation modeling, extinction risk analysis, and spatial prioritization to assess the risks to China's island biodiversity from climate change. We applied two scenarios—SSP2-4.5 and SSP5-8.5—over 2041–2060, 2061–2080, and 2081–2100 to evaluate the risks to five key taxa, namely amphibians, birds, mammals, reptiles, and vascular plants. Future climate change might result in an extinction rate of 11.6%, 5.5%, 11.9%, 12.0% and 11.9% for amphibians, birds, mammals, reptiles, and vascular plants respectively under the SSP2-4.5 climate change scenario and 19.0%, 8.9%, 20.6%, 19.6% and 20.6% respectively under the SSP5-8.5 scenario. Additionally, 60 and 97 islands under SSP2-4.5 and SSP5-8.5, respectively, are projected to lose at least one major taxonomic group by 2081–2100. High-risk zones, such as the islands near the Pearl River Delta and the Yangtze River Delta, are likely to face greater vulnerability than other islands in China. Our species- and island-specific results provide a scientific basis for developing targeted adaptation technologies, tailored to local island characteristics and species habitat dynamics. Recommended technologies include enhancing coastal engineering and restoring coastal shelter forest for island protection, expanding protected area networks for habitat preservation, and designating target habitat islands to support species relocation for high-risk species. Advanced monitoring technologies, such as AI-driven ecological sensors, are also critical for managing data-deficient and dynamic islands.

Climate change-induced range shifts in species pose a profound challenge to biodiversity conservation. China has recently updated its list of key protected species, encompassing 980 wildlife species and 455 plant species. However, the potential impacts of climate change on the distribution patterns of these species remain unclear, inevitably hindering the formulation of effective and adaptive conservation strategies. This study combines species distribution models with gap analysis to examine the negative and positive impacts of climate change on 1023 key protected species. We assessed species extinction risks, identified conservation gaps and effectiveness, and proposed adaptive strategies to mitigate the impacts of climate change. Our findings indicate that under the SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios, 5, 28, and 83 species, respectively, would face high extinction risks with universal dispersal by the end of the 21st century. Plants generally exhibit higher habitat loss rates and extinction risks than animals. Among animal taxa, amphibians exhibit the highest extinction risks and habitat loss rates, with notably lower habitat gain and habitat remain rates compared to other groups. Geographically, species in Central China and Northeast China are at the highest risk of extinction, whereas the Qinghai‒Xizang Plateau, Northwest China, and South China experience relatively lower risks. Although the current protected area network provides adequate coverage for the majority of target species, a notable conservation gap (>25%) persists for 115 species. Under the SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios with universal dispersal, the average species turnover rates within protected areas are 36.29%, 43.29%, and 51.10%, respectively, by the end of the 21st century. This study highlights the need for dynamic conservation and adaptation strategies in the context of climate change, offering essential insights for achieving the 30 × 30 conservation target and developing long-term effective adaptation strategies.

The development of climate-adaptive migration corridors has emerged as a key strategy for biodiversity conservation. However, most existing studies focus on the migration patterns and adaptability of a few species and barely pay attention to the design of migration corridors that address multispecies needs at a national scale under climate change. In this study, we analysed 1023 nationally protected wildlife species in China to predict their potential distributions under current climatic conditions and the SSP2-4.5 scenario using the maximum entropy model. The projections were used as a base to conduct hotspot analysis to identify areas with declining, stable or increasing habitat selection rates (HSRs), which were designated as ecological sources. These areas correspond to regions likely to experience species emigration, retention or immigration. Using circuit theory and the minimum cumulative resistance model, we employed the Linkage Mapper tool to construct climate-resilient conservation corridors and identify critical ecological nodes. We identified 49 ecological sources, including 19 ecological sources with declining HSRs, 13 ecological sources with stable HSRs and 17 ecological sources with increasing HSRs. These HSRs collectively covered over 90% of the studied species and demonstrated a strong conservation representativeness. We also mapped 108 migration corridors, including 49 supporting species movement from areas with declining HSRs and 59 enhancing connectivity and species exchange. In addition, we identified 978 ecological pinch points and 203 barrier points, which are critical priorities for future corridor planning. A novel framework for the design of multispecies conservation corridors that support climate change adaptation, which contributes to China's efforts to achieve the Kunming-Montreal Biodiversity Framework targets and improve ecosystem connectivity.

The diurnal temperature range (DTR) has generally decreased at global scale according to the IPCC reports and previous publications. Here, we examined the change in DTR within the past four decades and found that DTR decreased in wet areas, but it increased in dry areas. This is because the daily minimum air temperature (Tmin) increased at a slower rate than the daily maximum air temperature (Tmax) in the dry areas, while it increased more rapidly in the wet areas. The changes in cloud cover, water vapour and moisture flux were observed to have contributed to the observed change in DTR. The change in moisture flux largely contributed to this change through the indirect effect on the change in water vapour across the aridity gradient. The significant impact of moisture flux is linked to the differential change in moisture flux between the dry and wet zones. No significant change in diurnal temperature range in recent decades over global land areas as a whole. Significant increase in diurnal temperature range in the dry areas but decrease in the wet areas. Changes in cloud cover, water vapour and moisture flux influence the decrease in diurnal temperature range along the aridity gradient.image

Limiting global warming to within 1.5 °C might require large-scale deployment of premature negative emission technologies with potentially adverse effects on the key sustainable development goals. Biochar has been proposed as an established technology for carbon sequestration with co-benefits in terms of soil quality and crop yield. However, the considerable uncertainties that exist in the potential, cost, and deployment strategies of biochar systems at national level prevent its deployment in China. Here, we conduct a spatially explicit analysis to investigate the negative emission potential, economics, and priority deployment sites of biochar derived from multiple feedstocks in China. Results show that biochar has negative emission potential of up to 0.92 billion tons of CO 2 per year with an average net cost of US$90 per ton of CO 2 in a sustainable manner, which could satisfy the negative emission demands in most mitigation scenarios compatible with China’s target of carbon neutrality by 2060.

Plant diversity plays a crucial role in maintaining the functionality of a community and providing essential ecosystem services. Studying the plant diversity and its response to environmental factors in the Yellow River Delta, China, as a newly formed coastal land, is beneficial for protecting plant diversity in coastal areas and maintaining ecosystem stability. In this study, 56 sites were sampled to investigate the diversity of shrubs and herbaceous plant community and its response to environmental factors. The results indicate that the plants growing in the Yellow River Delta are predominantly from the Poaceae and Asteraceae families, with dominant communities consisting of species such as Suaeda salsa, Phragmites australis, Setaria viridis, Imperata cylindrica, and Tamarix chinensis. The Shannon–Wiener index, Simpson diversity index, and Pielou’s evenness index exhibit average values of 0.34, 0.21, and 0.25, respectively, within the Yellow River Delta. These values collectively indicate a low diversity in the vegetation community, reflecting a relatively uncomplicated ecological structure in this area. Additionally, there were no significant differences in biodiversity indices under different soil formation times, but under different land cover types, the biodiversity index of cropland was significantly higher than that of impervious land. Soil salinity index exhibited a significant negative correlation with plant diversity (R2 = 0.279, p < 0.001) in the Yellow River Delta. Moreover, elevation (R2 = 0.247, p < 0.001) and temperature (R2 = 0.219, p < 0.001) showed significant positive effects on plant diversity. Regarding the ecological stoichiometry of plant elements, soil organic carbon exhibited a negative effect on the biodiversity index, while litter carbon showed a positive effect. This may be attributed to the unique topographical conditions and soil salinization in the Yellow River Delta. Our findings provide important references for the sustainable management of wetlands in the Yellow River Delta under conditions of soil salinization.

Building ecological networks can effectively enhance the quality and stability of ecosystems and better conserve biodiversity. Previous studies mainly determined ecological corridors based on selecting ecological sources at a regional scale (e.g., an administrative area), without considering the bioclimatic heterogeneity within the study area. Here, we propose a novel integrating approach involving bioclimatic zoning and selecting ecological sources from various bioclimatic zones to design ecological corridors. Taking Xi’an City, China, as an example, key bioclimatic variables were first chosen, and we partitioned the study area based on its bioclimatic characteristics through a combination of K-means clustering and variance inflation factor (VIF). Ecological sources were then identified from the combination of ecosystem services and habitats of 36 endangered species. Subsequently, the minimum cumulative resistance (MCR) model was used to build ecological networks within different bioclimatic zones and across the entire region. We found the following: (1) In Xi’an city, a total of 49 source areas and 117 corridors were identified. The identified network can protect 97.77% of species, facilitating connectivity between 30.50% of ecosystems and 35.5% of species-rich areas. (2) The integrating approach protects 12.26% more species richness and 10.95% more ecosystem services than the average value of the regional and bioregional approaches. Compared to regional and bioregional methods, integrating approaches demonstrate greater advantages in preserving species richness and ecosystem services. This study introduces a novel approach to constructing regional ecological networks, which integrates the impact of bioclimatic zoning into the process of network construction to improve ecosystem services and protect species habitats.

China's forests play a vital role in the global carbon cycle through the absorption of atmospheric CO2 to mitigate climate change caused by the increase of anthropogenic CO2. It is essential to evaluate the carbon sink potential (CSP) of China's forest ecosystem. Combining NDVI, field-investigated, and vegetation and soil carbon density data modeled by process-based models, we developed the state-of-the-art learning ensembles model of process-based models (the multi-model random forest ensemble (MMRFE) model) to evaluate the carbon stocks of China's forest ecosystem in historical (1982-2021) and future (2022-2081, without NDVI-driven data) periods. Meanwhile, we proposed a new carbon sink index (CSindex) to scientifically and accurately evaluate carbon sink status and identify carbon sink intensity zones, reducing the probability of random misjudgments as a carbon sink. The new MMRFE models showed good simulation results in simulating forest vegetation and soil carbon density in China (significant positive correlation with the observed values, r = 0.94, P < 0.001). The modeled results show that a cumulative increase of 1.33 Pg C in historical carbon stocks of forest ecosystem is equivalent to 48.62 Bt CO2, which is approximately 52.03% of the cumulative increased CO2 emissions in China from 1959 to 2018. In the next 60 years, China's forest ecosystem will absorb annually 1.69 (RCP45 scenario) to 1.85 (RCP85 scenario) Bt CO2. Compared with the carbon stock in the historical period, the cumulative absorption of CO2 by China's forest ecosystem in 2032-2036, 2062-2066, and 2077-2081 are approximately 11.25-39.68, 110.66-121.49 and 101.31-111.11 Bt CO2, respectively. In historical and future periods, the medium and strong carbon sink intensity regions identified by the historical CSindex covered 65% of the total forest area, cumulative absorbing approximately 31.60 and 65.83-72.22 Bt CO2, respectively. In the future, China's forest ecosystem has a large CSP with a non-continuous increasing trend. However, the CSP should not be underestimated. Notably, the medium carbon sink intensity region should be the priority for natural carbon sequestration action. This study not only provides an important methodological basis for accurately estimating the future CSP of forest ecosystem but also provides important decision support for future forest ecosystem carbon sequestration action.

Quantifying the contribution of natural ecosystems on air quality regulation can help to lay the foundation for ecological construction, and to promote the sustainable development of natural ecosystems. To identify the spatio-temporal dynamic changes of natural vegetation regulation on SO2 absorption and the underlying mechanism of these changes in Qinghai Province, an important ecological barrier and the unique natural ecosystems, the Biome-BGC model was improved to simulate the canopy conductance to SO2 and leaf area index (LAI) on the daily scale, and then the SO2 absorption by vegetation was estimated coupling SO2 concentration from satellite data. Our results showed that the annual average SO2 absorption of the natural ecosystems in Qinghai Province was 4.74 × 104 tons yr−1 from 2005 to 2018, accounting for about 40% of the total emissions. Spatially, the ecosystem service of SO2 absorption gradually decreased from southeast to northwest, and varied from 0 in Haixi state to 14.37 kg SO2 ha−1 yr−1 in Haibei state. The annual average SO2 absorption in unit area was 0.68 kg SO2 ha−1 yr−1, and significantly higher SO2 absorption was observed in summer with 0.45 kg SO2 ha−1 quarterly. The canopy conductance and LAI controlled by climate variables would be the dominant driving factors for the variation of SO2 absorption for natural ecosystems. The sensitivity analysis showed that SO2 concentration contributed more to the uncertainties of SO2 absorption than the conductance in this study. Our results could provide powerful supports for realistic eco-environmental policy and decision making.

The snow leopard (Panthera uncia) lives in alpine ecosystems in Central Asia, where it could face intensive climate change and is thus a major conservation concern. We compiled a dataset of 406 GPS-located occurrences based on field surveys, literature, and the GBIF database. We used Random Forest to build different species distribution models with a maximum of 27 explanatory variables, including climatic, topographical, and human impact variables, to predict potential distribution for the snow leopard and make climate change projections. We estimated the potential range shifts of the snow leopard under two global climate models for different representative concentration pathways for 2050 and 2070. We found the distribution center of the snow leopard may move northwest by about 200 km and may move upward in elevation by about 100 m by 2070. Unlike previous studies on the range shifts of the snow leopard, we highlighted that upward rather than northward range shifts are the main pathways for the snow leopard in the changing climate, since the landform of their habitat allows an upward shift, whereas mountains and valleys would block northward movement. Conservation of the snow leopard should therefore prioritize protecting its current habitat over making movement corridors.

Background Climate change has altered global hydrological cycles mainly due to changes in temperature and precipitation, which may exacerbate the global and regional water shortage issues, especially in the countries along the Belt and Road (B&R). Methods In this paper, we assessed water supply, demand, and stress under three climate change scenarios in the major countries along the Belt and Road. We ensembled ten Global Climate Model (GCM) runoff data and downscaled it to a finer resolution of 0.1° × 0.1° by the random forest model. Results Our results showed that the GCM runoff was highly correlated with the FAO renewable water resources and thus could be used to estimate water supply. Climate change would increase water supply by 4.85%, 5.18%, 8.16% and water demand by 1.45%, 1.68%, 2.36% under RCP 2.6, 4.5, and 8.5 scenarios by 2050s, respectively. As a result, climate change will, in general, have little impact on water stress in the B&R countries as a whole. However, climate change will make future water resources more unevenly distributed among the B&R countries and regions, exacerbating water stress in some countries, especially in Central Asia and West Asia. Our results are informative for water resource managers and policymakers in the B&R countries to make sustainable water management strategies under future climate change.

The Tibetan Plateau (TP) has a variety of vegetation types that range from alpine tundra to tropic evergreen forest, which play an important role in the global carbon (C) cycle and is extremely vulnerable to climate change. The vegetation C uptake is crucial to the ecosystem C sequestration. Moreover, net reduction in vegetation C uptake (NRVCU) will strongly affect the C balance of terrestrial ecosystem. Until now, there is limited knowledge on the recovery process of vegetation net C uptake and the spatial-temporal patterns of NRVCU after the disturbance that caused by climate change and human activities. Here, we used the MODIS-derived net primary production to characterize the spatial-temporal patterns of NRVCU. We further explored the influence factors of the net reduction rate in vegetation C uptake (NRRVCU) and recovery processes of vegetation net C uptake across a unique gradient zone on the TP. Results showed that the total net reduction amount of vegetation C uptake gradually decreased from 2000 to 2015 on the TP (Slope = -0.002, P < 0.05). Specifically, an increasing gradient zone of multi-year average of net reduction rate in vegetation carbon uptake (MYANRRVCU) from east to west was observed. In addition, we found that the recovery of vegetation net C uptake after the disturbance caused by climate change and anthropogenic disturbance in the gradient zone were primarily dominated by precipitation and temperature. The findings revealed that the effects of climate change on MYANRRVCU and vegetation net C uptake recovery differed significantly across geographical space and vegetation types. Our results highlight that the biogeographic characteristics of the TP should be considered for combating future climate change.

It is the goal of protected area management to make full use of limited resources to better protect biodiversity. Currently, the main tasks of developing national park system in China are to combine conservation features, optimize the spatial network of protected areas, and identify the prio-rity conservation areas of national parks effectively. In this study, we assessed the spatial distribution of key ecosystem services (carbon sequestration, oxygen release, hydrological regulation, water resources, and soil retention) using ecological model, and simulated the distribution of suitable habitats for 37 endangered species by MaxEnt in Lishui City, Zhejiang Province. The irreplaceability index of each planning unit in Lishui was calculated on the 0.4 km×0.4 km grid using the systema-tic conservation planning model (MARXAN), setting key ecosystem services and endangered species as the conservation objects. Combined with the local management needs, the priority protection areas of national parks were identified comprehensively. The results showed that during 2005 to 2015, the annual carbon storage, oxygen release, hydrological regulation, water resource, and soil retention in the study area was 0.05 kg C·m-2·a-1, 0.13 kg O2·m-2·a-1, 83.25×108 m3·a-1, 803 mm·a-1, and 95.53 t·hm-2·a-1, respectively. The irreplaceability index of different land use types was significantly different. The irreplaceability index of forest, river and reservoir, garden, cultivated land, residential land was 50-100, 60-100, 30-50, 15-35, 0-25, respectively. The priority conservation areas accounted for 11.8% of the study area. This study put forward a systematic conservation planning idea combining biodiversity and ecosystem services, which could provide a useful framework and technical support for optimizing the network layout of protected areas and priority conservation areas of national parks, and help to enhance the overall effectiveness of the establishment of the protected areas system with national parks as its main type in China.