- Email: [email protected]
Delineating the ecological conservation redline based on the persistence of key species: Giant pandas (Ailuropoda melanoleuca) inhabiting the Qinling Mountains
Ecological Modelling 345 (2017) 56–62
Contents lists available at ScienceDirect
Ecological Modelling journal homepage: www.elsevier.com/locate/ecolmodel
Delineating the ecological conservation redline based on the persistence of key species: Giant pandas (Ailuropoda melanoleuca) inhabiting the Qinling Mountains Minghao Gong a,∗ , Zhiyong Fan b , Junyan Wang a , Gang Liu a , Chin Lin c a
Research Institute of Wetland, Beijing Key Laboratory of Wetland Services and Restoration, Chinese Academy of Forestry, Beijing 100091, China WWF-China, Baiwanzhuang Street of West district, Beijing, 100037, China c Academy of Forestry Inventory and Planning, State Forestry Administration, Beijing 100714, China b
a r t i c l e
i n f o
Article history: Received 14 September 2016 Received in revised form 18 November 2016 Accepted 20 November 2016 Keywords: Ecological safety Ecological conservation redline Delineating framework Species Giant panda Qinling mountains
a b s t r a c t To effectively resolve conflicts between natural resources and environmental protection and guarantee ecological safety, the Chinese government has proposed a national strategy to delineate ecological conservation redlines (ECRs). The ECR is defined as the least amount of area needed to guarantee the national and regional ecological safety of ecosystem services and implementation of strict mandatory protection policy. Because this was piloted by government and theoretical study has lagged, there remains no fully accepted ECR delineation framework. Being neglected in the current delineating guidelines for ECR, we focused on ECR delineation of the Qinling Mountains based on the sustainable survival needs of giant pandas (Ailuropoda melanoleuca) to explore the ECR delineation approach with species persistence. We define the concepts of basic, current and future ECR, and set the principals and procedures for ECR delineation based on the historical and current giant panda population range, current and future habitat modeling with data from national giant panda surveys, and the impacts of climate change. Our results indicate that the basic ECR is 369,531 ha; the current ECR is 422,149 ha with 33,498 ha of suitable and sub-suitable habitat from the basic ECR covering 67% of current giant panda reserves. The future ECR in 2050 is 516,838 ha with 109,990 ha future suitable and sub-suitable habitat based on current ECR and covers 84% of current reserves. As the foundation of an ecosystem, species deserve to be an important basis when delineating ECR. Concentrating on the needs of long-term species survival, rich and powerful study of target species’ biology makes ECR delineation feasible and operational, while strengthening theoretical and scientific support at better temporal and spatial scales. The approach and index system employed here is a scientific framework for ECR delineation, especially given the temporal scale of population and habitat, and the main driving factors for future habitat dynamics. This method avoids the complicated process of ecological assessment. We hope our methods and reasoning can be incorporated into national guidelines and applied to ECR delineation across China. © 2016 Elsevier B.V. All rights reserved.
1. Introduction To address the conflict between natural resources and environmental protection, and to resolve spatial mismatches and management gaps in current ecological conservation policy, the Chinese government has proposed a national environmental policy based on ecological conservation redlines (ECRs) (Gao, 2015; Xu et al., 2015; Zheng and Ouyang, 2014). The ECR is defined as the least amount of area needed to guarantee the national and regional eco-
∗ Corresponding author. E-mail address: [email protected] (M. Gong). http://dx.doi.org/10.1016/j.ecolmodel.2016.11.011 0304-3800/© 2016 Elsevier B.V. All rights reserved.
logical safety of ecosystem services and implementation of strict mandatory protection policy. Due to its significance to economic and social sustainable development, this policy was approved by the Central Committee of the Communist Party of China in 2013 and included in revised environmental protection law in 2014 (Bai et al., 2015; Liu et al., 2015; Yang et al., 2014). As a policy for ecological protection, the ECR needs to be delineated according to specific areas and boundaries before it can be put into conservation practice. The Chinese Ministry of Environmental Protection, the key ministry sponsoring ECR policy, released a recommended standard technical guide (hereafter, the guide) for identifying ECRs based on previous zoning of eco-function areas, eco-fragile hotspots and biodiversity hotspots in 2015 (Chinese Ministry of Environmental
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
Protection, 2015). With methodological support, some provinces and cities launched a delineation of their ECR, including Hunan, Jilin and Jiangsu (Lin et al., 2016; Liao et al., 2015; Yan et al., 2014). The ECR is a policy piloted by government and remains ahead of the current state of ecological research. There is a lack, and lag, of theoretical and technical support, and this has resulted in operational challenges when implementing ECR delineation. The main theories initially reflected in ECR delineation were those around landscape ecology and system ecology (Lin et al., 2016; Si et al., 2013; Xu et al., 2015) such as hotspots, gap analysis and systematic conservation planning (Jørgensen, 2001; Jantke et al., 2011). Other theories, especially population ecology and conservation ecology, are seldom incorporated into the recommended techniques for operational implementation. The assessment of ecological value is an important process in ECR delineation, but the valuation models in national guidelines require a great number of parameter settings and scenario assumptions. Without clear identification of the main function for a given area in the guide, confusion about which function can be used to evaluate ecological value and parameters persists (e.g. water storage, biodiversity protection or landscape value) (Lin et al., 2016), and it remains difficult to construct a more convenient model for ECR. Thus we now have a situation where ecological value assessments are only completed at provincial scales during ECR delineation (e.g. in Jilin and Jiangsu). In addition, the ECR guidelines rely heavily on the results of previous zoning of ecological function, major function and conservation networks, and lack response strategies to faults in this zoning (Lin et al., 2016). All zoning has defects caused by topographic isolation, landscape fragmentation and administrative boundaries that remain unresolved (Xu et al., 2015). Further, the main basis for ECR delineation was based on the current requirements of economic development projects at the expense of ecological function and system dynamics, and the historical and future conditions affecting ecological conservation needs are rarely included. In combination, these factors have impeded the development of a fully accepted technical, scientific and index system for delineating ECR in China. Species form the basic component of biodiversity, ecosystems and ecological services, but have been neglected from the ECR process. The current ECR definition and guidelines have biodiversity as the main target, but the sustainable survival needs of species are not reflected. The theories of population ecology and conservation ecology, closely related to species, have not been incorporated into the scientific basis of ECR and these omissions represent an important technical gap in the delineation of ECR. In order to address this gap, here we focused on ECR delineation based on the survival of one species. We used the Qinling Mountains as a case study and delineated its ECR according to the sustainable needs of giant pandas (Ailuropoda melanoleuca). Giant pandas are a global icon of biodiversity conservation and the Qinling Mountains form their northernmost stronghold (State Forestry Administration, 2014). This case study provides an opportunity (1) to explore ECR delineation based on species sustainability and survival; (2) strengthen and build theoretical support of ECR; and (3) develop a case study for ECR delineation using giant pandas and their long-term survival based on biological characteristics, and current and future habitat dynamics. Our overall aim is that a species approach is incorporated into ECR delineation in order to facilitate better zoning and ECR implementation across China.
2. Materials and methods 2.1. Study area The Qinling Mountains run west to east across Shanxi in central China. The highest peak is Taibai Mountain at 3767 m above
57
sea level in southern Shanxi (Fig. 1). The Qinling Mountains are of great importance to China because they form a watershed between the Yangtze and Yellow Rivers, and a natural climatic and cultural boundary between southern and northern China. This area is a biodiversity hotspot and a global conservation priority area (Mittermeier et al., 2009; Olson and Dinerstein, 1998) containing many rare, endangered and endemic species such giant panda, golden monkey (Rhinopithecus), takin (Budorcas taxicolor) and ibis (Nipponia nippon). Fourteen nature reserves have been established here (Fig. 1). The Qinling Mountains are a stronghold for giant pandas and contain 18.6% of all giant pandas in China according to the Fourth National Giant Panda Survey (FNGPS) conducted by the State Forestry Administration of China. The Qinling Mountains are the focus of nature forest protection projects and provide water for Beijing and Xian. The mountains have been designated an important ecological function area and area of water resource conservation by the Chinese Ministry of Environmental Protection, and a major function-oriented area by the Chinese National Development and Reform Commission (Fan et al., 2010; Yue et al., 2012). Delineating ECR is of real significance to the eco-functioning and ecoservices of the Qinling Mountains. 2.2. Giant panda presence and habitat data The State Forestry Administration conducted national surveys of giant pandas in 2000 and 2012. During each survey, line transect methods were applied within pre-defined 2 km2 survey grid cells (the side length of a cell is close to the average radius of a giant panda’s home range) (State Forestry Administration 2006, 2014) that covered the entire giant panda range and potential habitat. All signs for the presence of giant pandas such as feces, dens, sleeping sites and footprints were collected; latitude, longitude, elevation, slope and vegetation at each sign point were recorded. We derived elevation and slope data of habitat and GIS layers from a digital elevation model based on 1:50,000 topographic maps obtained from the Chinese Academy of Sciences (www.gscloud.cn). In addition, vegetation and bamboo data and GIS layers were obtained from the two surveys and satellite images from Landsat 5 in 2000 and Spot5 in 2012 using the maximum likelihood classification algorithm in supervised classification by Erdas 8.7 (Leica Geosystems GIS and Mapping, 2003, LLC, Atlanta, GA, USA). All data were obtained and approved by the State Forestry Administration. Given global change, we also included climatic variables as habitat factors using current and future bioclimatic variables at a 30 s resolution from the WorldClim database (WorldClim.com) (Hijmans et al., 2005). As the main human-induced threat to giant panda habitat (Zhu et al., 2013), the latest road data, including national roads, highways and high-speed railways were taken from previous studies and field surveys (Fig. 1). All geospatial data were based on the UTM WGS 84 coordinate system. The raster data resolution was 30 × 30 m and data were analyzed using ArcGIS10.0 (Esri, Redlands, USA). 2.3. ECR modeling 2.3.1. The principles and definition of ECR To identify a scientific ECR for the Qinling Mountains based on the long-term survival of giant pandas we set the principal criteria for ECR delineation and modeling as follows: (1) ECR should include the area that current giant pandas possibly inhabit, taking historical distributions into consideration; (2) ECR should be dynamic and include current and future high quality habitat to ensure species long term survival; (3) ECR should meet the integrity and connectivity of landscapes for ecosystem function and biodiversity conservation needs referring to expert knowledge; and (4) the size of ECR should be appropriate considering land resource
58
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
Fig. 1. Giant panda population range from the results of the fourth national survey and nature reserves in the Qinling Mountains, China.
shortages and giant panda habitat usage, adjusted according to habitat and population range dynamics, and exclude areas used by people. To reflect temporal and spatial ECR dimensions we defined the area of ECR as: BECR = CPR + HPR
(1)
CECR = BECR + CHR − HUA
(2)
FECR = CECR + FHR − HUA
(3)
where, BECR: basic ECR; CECR: current ECR, FECR: future ECR; CPR: current population range; HPR: historical population range; CHR: current habitat range; FHR: future habitat range; and HUA: human used area, such as community land, agriculture land and road network. 2.3.2. Population range Using these principles we needed to identify the population distribution range and current and future habitat suitability. As a solitary animal (Schaller, 1985), the giant panda has a relatively exclusive home range and signs (fecal droppings, dens and sleeping sites) are effective indicators of activity. The mean diameter of the giant panda home range in the Qinling Mountains is 3.6 km, ranging up to 10.62 km2 (Pan et al., 2014). Thus, giant panda signs were used to denote the center of a circle and produce circles with a radius of 3.6 km using the buffer function in Arcgis 10.0. These circles represented the most likely area that a giant panda may utilize and reach from all field signs. Then, polygons were created based on merging these circles to establish the population distribution range and formed the area of BECR. Data collected during the giant panda surveys in 2000 and 2012 were used as the baseline for basic ECR calculation. 2.3.3. Climate change To determine habitat and its dynamics, maximum entropy modeling (Maxent) was applied to assess current and future habitat based on variables such as vegetation, bamboo, elevation, slope, resident community and road disturbances (Phillips and Dudík, 2008; Qi et al., 2011). Climate variables were also incorporated into this model and set as trigger factors to changes in habitat in the
future (Fan et al., 2014; Li et al., 2014; Songer et al., 2012). There are several climatic variables, general circulation models (GCMs), periods and CO2 emission scenarios which need to be chosen. In order to have an actual contribution to giant panda conservation practices we set 2050 as the time range for habitat projection. Based on previous work on the impacts of climate change on giant pandas we employed eight climatic variables (Bio 2, Bio 4, Bio 10, Bio 11, Bio 15, Bio 17, Bio 18 and Bio 19; Table 1) to represent climate conditions (Tuanmu et al., 2013), and selected BCC-CSM1-1, CCSM4, HadGEM2-ES, and MIROC5 as the GCMs for future climate projection (Li et al., 2014; Liu et al., 2016). To avoid uncertainty of future climate projection by different GCMs we averaged the value of the eight climatic variables under the four selected GCMs using ARCGIS 10.0, and used the average value to construct future climate models. Given our study area’s designation as an ecological function zone and bans on large scale industrial projects (Jie et al., 2012; Yue et al., 2012), the development model with low CO2 emissions was chosen. We selected representative concentration pathways (RCP) of 2.6 as future emissions scenarios (Hijmans et al., 2005; Parry et al., 2011).
2.3.4. Habitat suitability and ECR determination We used the above habitat variables and Maxent model to determine current and future giant panda habitat suitability. With suitability indices, habitat was classified as: suitable habitat 0.5–1, sub-suitable habitat 0.2–0.5, and ordinary habitat 0–0.2, matching 70%, 20% and 10% of giant panda presence, based on expert knowledge and previous work (Jian et al., 2013; Liu et al., 2016; Store and Kangas, 2001). To reveal dynamics of habitat under climate change we calculated the total area of suitable and sub-suitable habitat within the study area for the current and future scenarios, and the latitude, longitude and elevation of centroid for all habitat patches employing Arcgis10.0. Then, we compared the change of habitat in size and spatial pattern from now to 2050. With the results of current habitat and BECR, we can arrive at CECR by delineating current suitable habitats outside the BECR into the area of ECR. Then, we determine the boundary of FECR based on the CECR and future habitat using the same method. To delineate a specific ECR, we set 3.6 km as the scope to catch suitable habitat
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
59
Table 1 Bioclimatic variables, their contribution and percent permutation importance reported by Maxent. Variables are in order of highest to lowest permutation importance. Habitat variables
Description
Variable contribution (%)
Permutation importance (%)
Bio4 Bio15 Bio11 Bio17 Slop Bio18 Bio2 Elevation Vegetation Bamboo Bio19 Road Aspect Bio10 Resident
Temperature seasonality Precipitation seasonality Mean temperature of coldest quarter Precipitation of driest quarter Topographic characteristic Precipitation of warmest quarter Mean diurnal range Topographic characteristic Vegetation formation group Food resource Precipitation of coldest quarter Anthropogenic disturbance of transportation Topographic characteristic Mean temperature of warmest quarter Anthropogenic disturbance of human activity
22.9 20.4 5.4 7.4 6.8 14.1 9.0 2.1 1.5 1.2 0.1 3.3 0.8 4.5 0.5
25.0 23.0 16.3 8.4 5.8 4.9 4.2 3.7 2.4 1.9 1.3 1.1 0.8 0.7 0.5
Fig. 2. The giant panda population range based on home ranges in 2000 and 2012, and the area of the BECR in the Qinling Mountains.
patches out of the population range according to the diameter of giant panda home ranges in the Qinling Mountains (we doubled the area of likely giant panda occurrence). 2.3.5. Optimizing ECR Based on the baseline and the current government guide for ECR we removed areas impossible for giant pandas to utilize, such as (1) areas above 3100 m and under 1200 m (based on elevation selection from the results of the national survey in 2012); and (2) areas occupied by people, agriculture and road networks. Last, areas close to the integrity and connectivity of landscapes, ecological services and biodiversity conservation were included in the scope of ECR based on expert knowledge and consensuses. 3. Results
habitat is 246,082 ha and 247,998 ha, accounting for 27.4% of the study area, Besides the growth in total suitability, the suitable habitat increased compared to the loss in sub-suitable habitat in 2050. Compared with the centroid location of current and future habitat, habitat in 2050 showed a spatial shift northward, eastward and upward including suitable habitat (6760 m, 4882 m, 33 m) and sub-suitable habitat (12,386 m, 2351 m, 75 m) (Fig. 3). Both the training AUC (0.91) and test AUC (0.87) indicate that the results and the performance of the model were reliable (Rebelo et al., 2010). The most important climate variable based on permutation importance was temperature seasonality (Bio4, 25%), followed by precipitation seasonality (Bio15, 23%). Slope was the most important habitat variable (5.8%) among abiotic and biotic habitat factors after climate factors. The disturbance to giant panda habitat from transportation is much higher than disturbance arising from human activity (Table 1).
3.1. Habitat suitability and dynamics, model validation with variable importance
3.2. Current and historical population range
With our suitability classification, current suitable and subsuitable habitat is 140,200 ha and 298,718 ha, accounting for 24.4% of the 1,800,565 ha study area. Future suitable and sub-suitable
Surveys in the Qinling Mountains detected 1800 giant panda signs in 2000 and 1854 in 2012. Based on population range modeling the area of current and historical giant panda population
60
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
Fig. 3. Present and 2050 giant panda habitat, suitability, and the area of the CECR and FECR in the Qinling Mountains.
Table 2 The area and suitability of three ECR for the Qinling Mountains when giant pandas are considered. ECR types
Habitat types, area (ha) and percentage (%)
Suitable habitat BECR CECR FECR
109123.6 118504.4 204863.0
29.5% 28.1% 39.6%
Sub-suitable habitat
Ordinary habitat
139521.0 170645.6 144857.5
120886.4 132999.0 167117.5
37.8% 40.4% 28.0%
distribution (CPR, HPR) was 312,198 ha (2000) and 283,786 ha (2012). After merging the two population ranges in 2000 and 2012 the total area was 369,531 ha covering 56.6% of total current suitable and sub-suitable habitat, and this was used as BECR and baseline for ECR modeling (Fig. 2, Table 2).
32.7% 31.5% 32.3%
Real area of ECR(ha)
Proportion of suitable and sub-suitable habitat in its total size of study area
369531.0 422149.0 516838.0
56.6% 65.9% 70.8%
3.3. Current ECR and future ECR based on population range and habitat Using the framework and ECR criteria we delineated the scope of CECR and FECR with calculated habitat, topographic features and giant panda characteristics. We incorporated 33,498 ha of suitable
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
and sub-suitable habitat into the scope of the ECR and identified the area of current ECR as 416,719 ha. Considering the importance to ecological service and biodiversity of areas above 3100 m and land needs in area at low elevations, we excluded the 3400 ha area below 1200 m out of the ECR which giant pandas unlikely use and added the 13,941 ha area above 3100 m into the scope of the ECR. After leaving 5111 ha of area within above ECR likely needed by people according to a national guide, the real scope of CECR for the Qinling Mountains was 422,149 ha with two patches covering 67% of the area of the current giant panda reserve network. The CECR includes 56.6% of total current suitable and sub-suitable habitat and includes 94% of the current population range and 99% of field signs. Considering habitat shift due to climate change in 2050 and based on the scope of the CECR, the future ECR of the Qinling Mountains was determined to be 526,709 ha, including 109,990 ha of future suitable habitat and sub-suitable habitat outside the CECR. The future ECR covered 84% of the area of the current giant panda reserve network, population and field signs. Due to climate change, the area above 3100 m will become giant panda habitat in 2050 and is included in the scope of the ECR (Fig. 3). Removing 9871 ha occupied by people and leaving future land needs, the real FECR for the Qinling Mountains was 516,838 ha comprising 70.8% of total future suitable and sub-suitable habitat (Fig. 3, Table 2)
4. Discussion Species are the basic unit of ecosystems and form the characteristics of an ecosystem via species biodiversity and proportional relationships (Balvanera et al., 2006; Blower, 1982; Loreau et al., 2001). All ecosystem functions and services are achieved by species, including water storage, oxygen production, carbon storage and soil conservation (Maitre et al., 2008; Sessment, 2005). Furthermore, species are direct and sensitive indicators of anthropogenic and natural disturbances via population size, spatial patterns and habitat usage behavior. Species deserve to be an important basis when delineating ECR. According to the needs of species survival, conservation programs including habitat restoration, connection and human disturbance control facilitate the improvement of landscape integrity and ecosystem structure and achieve effective conservation for an ecosystem and biodiversity. A well-known example involves biodiversity and ecosystems of the upper-middle reaches of the Yangtze River that have benefited from giant panda conservation programs (State Forestry Administration, 2014; WWF, 2015). Other examples are conservation of the Komodo dragon (Varanus komodoensis) for the restoration of ecosystems across eastern Indonesia (Ariefiandy et al., 2014), the Asian elephant (Elephas maximus) for the conservation of landscapes in Aceh, Indonesia, and the flying fox (Pteropus voeltzkowi) for forest protection on Pemba Island, Tanzania (BowenJones and Entwistle, 2002). We name species used to delineate ECR as target species. To ensure the significance of ECR we suggest that target species should be representative, typical and indicative to an area and include flagship species, key species, dominant species or surrogate species (Xu et al., 2014), and criteria for target species selection should be established. Over evolution, species gradually develop unique spatial spanning topography, food resources, habitat, behavior, home range, migration and predator avoidance (Allee and Schmidt, 1950). According to the requirements of environmental variables by biological characteristics, we can determine a specific and suitable area to guarantee long-term survival of a species that includes the current population and habitat and future habitat affected by the climate or other factors (Boitani et al., 2008; Hirzel et al., 2006; Hole et al., 2009). Using the area required for species long-term survival makes the ECR approach and policy more acceptable and
61
convincing, and solves the isolation and fragmentation issues in previous conservation zoning with species population development and safety. Further, species’ biology (especially some rare and endangered animals and plants) is often been well studied and can provide theoretical responses to ECR delineation. The persistence of a species depends on the integrity and connectivity of the ecosystem where it is located, and a species is unlikely to survive long term in a fragmented, poorly suitable environment (Hanski and Ovaskainen, 2000). The integrity and connectivity of landscape effectively guarantees the results of the ECR by providing an efficient and competent ecosystem for the long-term survival of a species. To maintain the integrity and connectivity of ECR, we removed western patches in the basic ECR that did not contain suitable giant panda habitat and included the perforation of the population range into the scope of CECR (Fig. 2); this area between the southeast and northeast of future suitable habitat was delineated into the scope of FECR too. Considering shrinking habitat in the central Qinling Mountains and shortages in land resources, the future ECR may remove this area located to the south of Changqing and Foping reserve (Fig. 3). We also noted that the northeastern part of the future ECR of the Qinling Mountains may become high quality habitat in 2050 but is currently outside the reserve network. This area should be included in future conservation planning by designing a new reserve or a dispersal corridor to link future giant panda habitat. With the increase in the proportion by suitable and sub-suitable habitat in the BECR, CECR and FECR, the delineated ECR covered the main high quality habitat for giant pandas and set a strong foundation for its persistence until 2050. It also shows that our ECR achieved its original aim and our method is of robust reliability. Technically, due to differences among the biological characteristic of species, the model of population range and habitat suitability in our study is not likely suitable for other animals. A franchised model and parameters based on the characteristic and study results of different targeted species during ECR modeling are needed. A similar situation lies in determining the driving factors of habitat dynamics. We also note that one limitation is that our method focused on species’ spatial requirements and less on ecosystem and landscape requirement. Therefore, the last step in optimizing the ECR based on expert knowledge and ecological needs will be important to improving the accuracy of ECR and should be emphasized during delineation. Species are not the aim of ECR policy and methodology, but are the outputs of the management of ecosystems. Based on long-term survival of giant pandas in the Qinling Mountains our study provides a framework for delineating a scientific ECR using population and habitat at different scales and primary habitat variables. Our method is ahead of the current national guideline in that (1) we expanded the temporal and spatial scale of ECR from the present to the future, and population to habitat while considering ecological and landscape requirements; (2) we concentrated on the long-term survival of a species and avoided the complicated process of ecosystem assessment; and (3) we demonstrated that ECR delineation is feasible and operational while improving theoretical support. We hope our method and reasoning can be incorporated into national guidelines and applied to ECR delineation across the whole of China.
Acknowledgements This study was funded by the Fourth National Giant Panda Survey initiated by the Chinese State Forestry Administration and WWF-China (No. 10003187). We are thankful for support from the Department of Nature Reserve Governance and Wildlife Protection, Chinese State Forestry Administration. We also acknowledge all
62
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
staff involved in the fourth national giant panda survey for their hard work, wisdom and dedication in 2012. References Allee, W.C., Schmidt, K.P., 1950. Book reviews: principles of animal ecology. Science 111, 424–425. Ariefiandy, A., Purwandana, D., Seno, A., Chrismiawati, M., Ciofi, C., Jessop, T.S., 2014. Evaluation of three field monitoring-density estimation protocols and their relevance to Komodo dragon conservation. Biodivers. Conserv. 23, 2473–2490. Bai, Y., Jiang, B., Wang, M., Li, H., Alatalo, J.M., Huang, S., 2015. New ecological redline policy (ERP) to secure ecosystem services in China. Land Use Policy 25, 348–351. Balvanera, P., Pfisterer, A.B., Buchmann, N., He, J.S., Nakashizuka, T., Raffaelli, D., Schmid, B., 2006. Quantifying the evidence for biodiversity effects on ecosystem functioning and services. Ecol. Lett. 9, 1146–1156. Blower, J.G., 1982. Population ecology: a unified study of animals and plants, third edition. J. Ecol. 70, 226–229. Boitani, L., Sinibaldi, I., Corsi, F., Biase, A.D., Carranza, I.D.I., Ravagli, M., Reggiani, G., Rondinini, C., Trapanese, P., 2008. Distribution of medium- to large-sized African mammals based on habitat suitability models. Biodivers. Conserv. 17, 605–621. BowenJones, Evan, Entwistle, Abigail, 2002. Identifying appropriate flagship species: the importance of culture and local contexts. Oryx 36, 189–195. Fan, J., Tao, A., Qing, Ren, 2010. On the historical background, scientific intentions, goal orientation, and policy framework of major function-Oriented zone planning in China. J. Res. Ecol., 289–299. Fan, J., Li, J., Xia, R., Hu, L., Wu, X., Li, G., 2014. Assessing the impact of climate change on the habitat distribution of the giant panda in the Qinling Mountains of China. Ecol. Modell. 274, 12–20. Gao, J., 2015. Exploring the delineation and supervision of ecological protection redlines in China. Biodivers. Sci. 23, 705–707. Hanski, I., Ovaskainen, O., 2000. The metapopulation capacity of a fragmented landscape. Nature 404, 755–758. Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G., Jarvis, A., 2005. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 25, 1965–1978. Hirzel, A.H., Lay, G.L., Helfer, V., Randin, C., Guisan, A., 2006. Evaluating the ability of habitat suitability models to predict species presences. Ecol. Modell. 199, 142–152. Hole, D.G., Willis, S.G., Pain, D.J., Fishpool, L.D., Butchart, S.H.M., Collingham, Y.C., Rahbek, C., Huntley, B., 2009. Projected impacts of climate change on a continent-wide protected area network. Ecol. Lett. 12, 420–431. Jørgensen, S.E., 2001. Recent Developments in System Ecology. Springer Berlin Heidelberg, pp. 155–169. Jantke, K., Schleupner, C., Schneider, U.A., 2011. Gap analysis of European wetland species: priority regions for expanding the Natura 2000 network. Biodiver. Conserv. 20, 581–605. Jian, J., Jiang, H., Jiang, Z., Zhou, G., Yu, S., Peng, S., Liu, S., Liu, S., Wang, J., 2013. Predicting giant panda habitat with climate data and calculated habitat suitability index (HSI) map. Meteorol. Appl. 21, 210–217. Jie, F., Wei, S., Kan, Z., Dong, C., 2012. Major Function Oriented Zone: new method of spatial regulation for reshaping regional development pattern in China. Chin. Geog. Sci. 22, 196–209. Li, R., Xu, M., Wong, M.H.G., Qiu, S., Sheng, Q., Li, X., Song, Z., 2014. Climate change-induced decline in bamboo habitats and species diversity: implications for giant panda conservation. Divers. Distrib. 21, 379–391. Liao, J.X., Zhou, Z., Department, B., 2015. Delineation and implementation of ecological protection redline in hunan province. J. Cent. South Univ. For. Technol. 21, 281–282. Lin, Y., Fan, J., Wen, Q., Liu, S., Li, B., 2016. Primary exploration of ecological theories and technologies for delineation of ecological redline zone. Acta Ecol. Sin. 36 (5), 1244–1252. Liu, D., Lin, N., Zou, C., You, G., 2015. Development of foreign ecological protected areas and linkages to ecological protection redline delineation and management in China. Biodivers. Sci. 23, 708–715.
Liu, G., Guan, T., Dai, Q., Li, H., Gong, M., 2016. Impacts of Temperature on Giant Panda Habitat in the North Minshan Mountains. Ecology and Evolution. Loreau, M., Naeem, S., Inchausti, P., Bengtsson, J., Grime, J.P., Hector, A., Hooper, D.U., Huston, M.A., Raffaelli, D., Schmid, B., 2001. Biodiversity and ecosystem functioning: current knowledge and future challenges. Science 294, 804–808. Maitre, D.C.L., Milton, S.J., Jarmain, C., Colvin, C.A., Saayman, I., Vlok, J.H., 2008. Linking ecosystem services and water resources: landscape-scale hydrology of the Little Karoo. Front. Ecol. Environ. 5, 261–270. Mittermeier, R.A., Myers, N., Mittermeier, C.G., Robles Gil, P., 2009. Hotspots: earth’s biologically richest and most endangered terrestrial ecoregions. Libr. J., 237–238. Olson, D.M., Dinerstein, E., 1998. The global 200: a representation approach to conserving the earth’s most biologically valuable ecoregions. Conserv. Biol. 12, 502–515. Pan, W., Mcshea, W.J., Garshelis, D., Wang, D., Harris, R.B., 2014. A Chance for Lasting Survival. Parry, M.L., Canziani, O.F., Palutikof, J.P., Linden, P.J.V.D., Hanson, C.E., 2011. Climate change 2007: impacts, adaptation and vulnerability: contribution of working group II to the fourth assessment report of the intergovernmental panel on climate change. Summary Policymakers. Astron. Nachr. 332, 697–705. Phillips, S.J., Dudík, M., 2008. Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography 31, 161–175. Qi, D., Zhang, S., Zhang, Z., Hu, Y., Yang, X., Wang, H., Wei, F., 2011. Different habitat preferences of male and female giant pandas. J. Zool. 285, 205–214. Rebelo, H., Tarroso, P., Jones, G., 2010. Predicted impact of climate change on European bats in relation to their biogeographic patterns. Global Change Biol. 16, 561–576. Schaller, G.B., 1985. Giant pandas of wolong. Bioscience 36. Sessment, M., 2005. Ecosystems and human well-being. Eco. Human Well-being Framework Assess. 42, 77–101. Si, Q., Li, M.C., Zhang, G.Y., Liang, S.X., Sun, Z.C., 2013. Pilot study on delimited method of marine regional ecological red line. Adv. Mater. Res. 807 (807–809), 728–731. Songer, M., Delion, M., Biggs, A., Huang, Q., 2012. Modeling impacts of climate change on giant panda habitat. Int. J. Ecol. 2012, 1–12. State Forestry Administration, 2006. The Third National Survey Report on Giant Panda in China. Science Press, Beijing (In Chinese). State Forestry Administration, 2014. The Fourth National Survey Report for Giant Panda in China (Announced but not published In Chinese). Store, R., Kangas, J., 2001. Integrating spatial multi-criteria evaluation and expert knowledge for GIS-based habitat suitability modelling. Landscape Urban Plann. 55, 79–93. ˜ A., Winkler, J.A., Li, Y., Xu, W., Ouyang, Z., Liu, J., 2013. Tuanmu, M.N., Vina, Climate-change impacts on understorey bamboo species and giant pandas in China/’s Qinling Mountains. Nat. Clim. Change 3, 249–253. WWF, 2015. Understanding Biodiversity in the Conservation of Flagship Species of Giant Pandas. http://mp.weixin.qq.com/s? biz=MjM5NTkxMTIyMQ==&mid=205378773&idx=1&sn=44d1e265 bae41b394733697dfff6f3f8. ˜ A., Qi, Z., Ouyang, Z., Liu, J., Wei, L., Hui, W., 2014. Evaluating Xu, W., Vina, conservation effectiveness of nature reserves established for surrogate species: case of a giant panda nature reserve in qinling mountains, China. Chin. Geog. Sci. 24, 60–70. Xu, D., Zou, C., Xu, M., You, G., Wu, D., 2015. Ecological security pattern construction based on ecological protection redlines. Biodivers. Sci. 23, 740–746. Yan, S.G., Lin, N.F., Shen, W.S., 2014. Delineation and protection of ecological red lines in jiangsu province. J. Ecol. Rural Environ. 30 (3), 294–299. Yang, B., Gao, J., Zou, C., 2014. The Strategic Significance of Drawing the Ecological Protection Red Line. China Development. Yue, X.M., Long, C.J., Long, G.J., Qian, Y.E., 2012. Study on ecological compensation model for major function oriented zones. Chin. J. Eco-Agric. 20, 1404–1408. Zheng, H., Ouyang, Z., 2014. Practice and Consideration for Ecological Redlining. Bulletin of Chinese Academy of Sciences. Zhu, L.F., Hu, Y.B., Zhang, Z.J., Wei, F.W., 2013. Effect of China’s rapid development on its iconic giant panda. Chin. Sci. Bull. 58, 2134–2139.
Contents lists available at ScienceDirect
Ecological Modelling journal homepage: www.elsevier.com/locate/ecolmodel
Delineating the ecological conservation redline based on the persistence of key species: Giant pandas (Ailuropoda melanoleuca) inhabiting the Qinling Mountains Minghao Gong a,∗ , Zhiyong Fan b , Junyan Wang a , Gang Liu a , Chin Lin c a
Research Institute of Wetland, Beijing Key Laboratory of Wetland Services and Restoration, Chinese Academy of Forestry, Beijing 100091, China WWF-China, Baiwanzhuang Street of West district, Beijing, 100037, China c Academy of Forestry Inventory and Planning, State Forestry Administration, Beijing 100714, China b
a r t i c l e
i n f o
Article history: Received 14 September 2016 Received in revised form 18 November 2016 Accepted 20 November 2016 Keywords: Ecological safety Ecological conservation redline Delineating framework Species Giant panda Qinling mountains
a b s t r a c t To effectively resolve conflicts between natural resources and environmental protection and guarantee ecological safety, the Chinese government has proposed a national strategy to delineate ecological conservation redlines (ECRs). The ECR is defined as the least amount of area needed to guarantee the national and regional ecological safety of ecosystem services and implementation of strict mandatory protection policy. Because this was piloted by government and theoretical study has lagged, there remains no fully accepted ECR delineation framework. Being neglected in the current delineating guidelines for ECR, we focused on ECR delineation of the Qinling Mountains based on the sustainable survival needs of giant pandas (Ailuropoda melanoleuca) to explore the ECR delineation approach with species persistence. We define the concepts of basic, current and future ECR, and set the principals and procedures for ECR delineation based on the historical and current giant panda population range, current and future habitat modeling with data from national giant panda surveys, and the impacts of climate change. Our results indicate that the basic ECR is 369,531 ha; the current ECR is 422,149 ha with 33,498 ha of suitable and sub-suitable habitat from the basic ECR covering 67% of current giant panda reserves. The future ECR in 2050 is 516,838 ha with 109,990 ha future suitable and sub-suitable habitat based on current ECR and covers 84% of current reserves. As the foundation of an ecosystem, species deserve to be an important basis when delineating ECR. Concentrating on the needs of long-term species survival, rich and powerful study of target species’ biology makes ECR delineation feasible and operational, while strengthening theoretical and scientific support at better temporal and spatial scales. The approach and index system employed here is a scientific framework for ECR delineation, especially given the temporal scale of population and habitat, and the main driving factors for future habitat dynamics. This method avoids the complicated process of ecological assessment. We hope our methods and reasoning can be incorporated into national guidelines and applied to ECR delineation across China. © 2016 Elsevier B.V. All rights reserved.
1. Introduction To address the conflict between natural resources and environmental protection, and to resolve spatial mismatches and management gaps in current ecological conservation policy, the Chinese government has proposed a national environmental policy based on ecological conservation redlines (ECRs) (Gao, 2015; Xu et al., 2015; Zheng and Ouyang, 2014). The ECR is defined as the least amount of area needed to guarantee the national and regional eco-
∗ Corresponding author. E-mail address: [email protected] (M. Gong). http://dx.doi.org/10.1016/j.ecolmodel.2016.11.011 0304-3800/© 2016 Elsevier B.V. All rights reserved.
logical safety of ecosystem services and implementation of strict mandatory protection policy. Due to its significance to economic and social sustainable development, this policy was approved by the Central Committee of the Communist Party of China in 2013 and included in revised environmental protection law in 2014 (Bai et al., 2015; Liu et al., 2015; Yang et al., 2014). As a policy for ecological protection, the ECR needs to be delineated according to specific areas and boundaries before it can be put into conservation practice. The Chinese Ministry of Environmental Protection, the key ministry sponsoring ECR policy, released a recommended standard technical guide (hereafter, the guide) for identifying ECRs based on previous zoning of eco-function areas, eco-fragile hotspots and biodiversity hotspots in 2015 (Chinese Ministry of Environmental
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
Protection, 2015). With methodological support, some provinces and cities launched a delineation of their ECR, including Hunan, Jilin and Jiangsu (Lin et al., 2016; Liao et al., 2015; Yan et al., 2014). The ECR is a policy piloted by government and remains ahead of the current state of ecological research. There is a lack, and lag, of theoretical and technical support, and this has resulted in operational challenges when implementing ECR delineation. The main theories initially reflected in ECR delineation were those around landscape ecology and system ecology (Lin et al., 2016; Si et al., 2013; Xu et al., 2015) such as hotspots, gap analysis and systematic conservation planning (Jørgensen, 2001; Jantke et al., 2011). Other theories, especially population ecology and conservation ecology, are seldom incorporated into the recommended techniques for operational implementation. The assessment of ecological value is an important process in ECR delineation, but the valuation models in national guidelines require a great number of parameter settings and scenario assumptions. Without clear identification of the main function for a given area in the guide, confusion about which function can be used to evaluate ecological value and parameters persists (e.g. water storage, biodiversity protection or landscape value) (Lin et al., 2016), and it remains difficult to construct a more convenient model for ECR. Thus we now have a situation where ecological value assessments are only completed at provincial scales during ECR delineation (e.g. in Jilin and Jiangsu). In addition, the ECR guidelines rely heavily on the results of previous zoning of ecological function, major function and conservation networks, and lack response strategies to faults in this zoning (Lin et al., 2016). All zoning has defects caused by topographic isolation, landscape fragmentation and administrative boundaries that remain unresolved (Xu et al., 2015). Further, the main basis for ECR delineation was based on the current requirements of economic development projects at the expense of ecological function and system dynamics, and the historical and future conditions affecting ecological conservation needs are rarely included. In combination, these factors have impeded the development of a fully accepted technical, scientific and index system for delineating ECR in China. Species form the basic component of biodiversity, ecosystems and ecological services, but have been neglected from the ECR process. The current ECR definition and guidelines have biodiversity as the main target, but the sustainable survival needs of species are not reflected. The theories of population ecology and conservation ecology, closely related to species, have not been incorporated into the scientific basis of ECR and these omissions represent an important technical gap in the delineation of ECR. In order to address this gap, here we focused on ECR delineation based on the survival of one species. We used the Qinling Mountains as a case study and delineated its ECR according to the sustainable needs of giant pandas (Ailuropoda melanoleuca). Giant pandas are a global icon of biodiversity conservation and the Qinling Mountains form their northernmost stronghold (State Forestry Administration, 2014). This case study provides an opportunity (1) to explore ECR delineation based on species sustainability and survival; (2) strengthen and build theoretical support of ECR; and (3) develop a case study for ECR delineation using giant pandas and their long-term survival based on biological characteristics, and current and future habitat dynamics. Our overall aim is that a species approach is incorporated into ECR delineation in order to facilitate better zoning and ECR implementation across China.
2. Materials and methods 2.1. Study area The Qinling Mountains run west to east across Shanxi in central China. The highest peak is Taibai Mountain at 3767 m above
57
sea level in southern Shanxi (Fig. 1). The Qinling Mountains are of great importance to China because they form a watershed between the Yangtze and Yellow Rivers, and a natural climatic and cultural boundary between southern and northern China. This area is a biodiversity hotspot and a global conservation priority area (Mittermeier et al., 2009; Olson and Dinerstein, 1998) containing many rare, endangered and endemic species such giant panda, golden monkey (Rhinopithecus), takin (Budorcas taxicolor) and ibis (Nipponia nippon). Fourteen nature reserves have been established here (Fig. 1). The Qinling Mountains are a stronghold for giant pandas and contain 18.6% of all giant pandas in China according to the Fourth National Giant Panda Survey (FNGPS) conducted by the State Forestry Administration of China. The Qinling Mountains are the focus of nature forest protection projects and provide water for Beijing and Xian. The mountains have been designated an important ecological function area and area of water resource conservation by the Chinese Ministry of Environmental Protection, and a major function-oriented area by the Chinese National Development and Reform Commission (Fan et al., 2010; Yue et al., 2012). Delineating ECR is of real significance to the eco-functioning and ecoservices of the Qinling Mountains. 2.2. Giant panda presence and habitat data The State Forestry Administration conducted national surveys of giant pandas in 2000 and 2012. During each survey, line transect methods were applied within pre-defined 2 km2 survey grid cells (the side length of a cell is close to the average radius of a giant panda’s home range) (State Forestry Administration 2006, 2014) that covered the entire giant panda range and potential habitat. All signs for the presence of giant pandas such as feces, dens, sleeping sites and footprints were collected; latitude, longitude, elevation, slope and vegetation at each sign point were recorded. We derived elevation and slope data of habitat and GIS layers from a digital elevation model based on 1:50,000 topographic maps obtained from the Chinese Academy of Sciences (www.gscloud.cn). In addition, vegetation and bamboo data and GIS layers were obtained from the two surveys and satellite images from Landsat 5 in 2000 and Spot5 in 2012 using the maximum likelihood classification algorithm in supervised classification by Erdas 8.7 (Leica Geosystems GIS and Mapping, 2003, LLC, Atlanta, GA, USA). All data were obtained and approved by the State Forestry Administration. Given global change, we also included climatic variables as habitat factors using current and future bioclimatic variables at a 30 s resolution from the WorldClim database (WorldClim.com) (Hijmans et al., 2005). As the main human-induced threat to giant panda habitat (Zhu et al., 2013), the latest road data, including national roads, highways and high-speed railways were taken from previous studies and field surveys (Fig. 1). All geospatial data were based on the UTM WGS 84 coordinate system. The raster data resolution was 30 × 30 m and data were analyzed using ArcGIS10.0 (Esri, Redlands, USA). 2.3. ECR modeling 2.3.1. The principles and definition of ECR To identify a scientific ECR for the Qinling Mountains based on the long-term survival of giant pandas we set the principal criteria for ECR delineation and modeling as follows: (1) ECR should include the area that current giant pandas possibly inhabit, taking historical distributions into consideration; (2) ECR should be dynamic and include current and future high quality habitat to ensure species long term survival; (3) ECR should meet the integrity and connectivity of landscapes for ecosystem function and biodiversity conservation needs referring to expert knowledge; and (4) the size of ECR should be appropriate considering land resource
58
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
Fig. 1. Giant panda population range from the results of the fourth national survey and nature reserves in the Qinling Mountains, China.
shortages and giant panda habitat usage, adjusted according to habitat and population range dynamics, and exclude areas used by people. To reflect temporal and spatial ECR dimensions we defined the area of ECR as: BECR = CPR + HPR
(1)
CECR = BECR + CHR − HUA
(2)
FECR = CECR + FHR − HUA
(3)
where, BECR: basic ECR; CECR: current ECR, FECR: future ECR; CPR: current population range; HPR: historical population range; CHR: current habitat range; FHR: future habitat range; and HUA: human used area, such as community land, agriculture land and road network. 2.3.2. Population range Using these principles we needed to identify the population distribution range and current and future habitat suitability. As a solitary animal (Schaller, 1985), the giant panda has a relatively exclusive home range and signs (fecal droppings, dens and sleeping sites) are effective indicators of activity. The mean diameter of the giant panda home range in the Qinling Mountains is 3.6 km, ranging up to 10.62 km2 (Pan et al., 2014). Thus, giant panda signs were used to denote the center of a circle and produce circles with a radius of 3.6 km using the buffer function in Arcgis 10.0. These circles represented the most likely area that a giant panda may utilize and reach from all field signs. Then, polygons were created based on merging these circles to establish the population distribution range and formed the area of BECR. Data collected during the giant panda surveys in 2000 and 2012 were used as the baseline for basic ECR calculation. 2.3.3. Climate change To determine habitat and its dynamics, maximum entropy modeling (Maxent) was applied to assess current and future habitat based on variables such as vegetation, bamboo, elevation, slope, resident community and road disturbances (Phillips and Dudík, 2008; Qi et al., 2011). Climate variables were also incorporated into this model and set as trigger factors to changes in habitat in the
future (Fan et al., 2014; Li et al., 2014; Songer et al., 2012). There are several climatic variables, general circulation models (GCMs), periods and CO2 emission scenarios which need to be chosen. In order to have an actual contribution to giant panda conservation practices we set 2050 as the time range for habitat projection. Based on previous work on the impacts of climate change on giant pandas we employed eight climatic variables (Bio 2, Bio 4, Bio 10, Bio 11, Bio 15, Bio 17, Bio 18 and Bio 19; Table 1) to represent climate conditions (Tuanmu et al., 2013), and selected BCC-CSM1-1, CCSM4, HadGEM2-ES, and MIROC5 as the GCMs for future climate projection (Li et al., 2014; Liu et al., 2016). To avoid uncertainty of future climate projection by different GCMs we averaged the value of the eight climatic variables under the four selected GCMs using ARCGIS 10.0, and used the average value to construct future climate models. Given our study area’s designation as an ecological function zone and bans on large scale industrial projects (Jie et al., 2012; Yue et al., 2012), the development model with low CO2 emissions was chosen. We selected representative concentration pathways (RCP) of 2.6 as future emissions scenarios (Hijmans et al., 2005; Parry et al., 2011).
2.3.4. Habitat suitability and ECR determination We used the above habitat variables and Maxent model to determine current and future giant panda habitat suitability. With suitability indices, habitat was classified as: suitable habitat 0.5–1, sub-suitable habitat 0.2–0.5, and ordinary habitat 0–0.2, matching 70%, 20% and 10% of giant panda presence, based on expert knowledge and previous work (Jian et al., 2013; Liu et al., 2016; Store and Kangas, 2001). To reveal dynamics of habitat under climate change we calculated the total area of suitable and sub-suitable habitat within the study area for the current and future scenarios, and the latitude, longitude and elevation of centroid for all habitat patches employing Arcgis10.0. Then, we compared the change of habitat in size and spatial pattern from now to 2050. With the results of current habitat and BECR, we can arrive at CECR by delineating current suitable habitats outside the BECR into the area of ECR. Then, we determine the boundary of FECR based on the CECR and future habitat using the same method. To delineate a specific ECR, we set 3.6 km as the scope to catch suitable habitat
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
59
Table 1 Bioclimatic variables, their contribution and percent permutation importance reported by Maxent. Variables are in order of highest to lowest permutation importance. Habitat variables
Description
Variable contribution (%)
Permutation importance (%)
Bio4 Bio15 Bio11 Bio17 Slop Bio18 Bio2 Elevation Vegetation Bamboo Bio19 Road Aspect Bio10 Resident
Temperature seasonality Precipitation seasonality Mean temperature of coldest quarter Precipitation of driest quarter Topographic characteristic Precipitation of warmest quarter Mean diurnal range Topographic characteristic Vegetation formation group Food resource Precipitation of coldest quarter Anthropogenic disturbance of transportation Topographic characteristic Mean temperature of warmest quarter Anthropogenic disturbance of human activity
22.9 20.4 5.4 7.4 6.8 14.1 9.0 2.1 1.5 1.2 0.1 3.3 0.8 4.5 0.5
25.0 23.0 16.3 8.4 5.8 4.9 4.2 3.7 2.4 1.9 1.3 1.1 0.8 0.7 0.5
Fig. 2. The giant panda population range based on home ranges in 2000 and 2012, and the area of the BECR in the Qinling Mountains.
patches out of the population range according to the diameter of giant panda home ranges in the Qinling Mountains (we doubled the area of likely giant panda occurrence). 2.3.5. Optimizing ECR Based on the baseline and the current government guide for ECR we removed areas impossible for giant pandas to utilize, such as (1) areas above 3100 m and under 1200 m (based on elevation selection from the results of the national survey in 2012); and (2) areas occupied by people, agriculture and road networks. Last, areas close to the integrity and connectivity of landscapes, ecological services and biodiversity conservation were included in the scope of ECR based on expert knowledge and consensuses. 3. Results
habitat is 246,082 ha and 247,998 ha, accounting for 27.4% of the study area, Besides the growth in total suitability, the suitable habitat increased compared to the loss in sub-suitable habitat in 2050. Compared with the centroid location of current and future habitat, habitat in 2050 showed a spatial shift northward, eastward and upward including suitable habitat (6760 m, 4882 m, 33 m) and sub-suitable habitat (12,386 m, 2351 m, 75 m) (Fig. 3). Both the training AUC (0.91) and test AUC (0.87) indicate that the results and the performance of the model were reliable (Rebelo et al., 2010). The most important climate variable based on permutation importance was temperature seasonality (Bio4, 25%), followed by precipitation seasonality (Bio15, 23%). Slope was the most important habitat variable (5.8%) among abiotic and biotic habitat factors after climate factors. The disturbance to giant panda habitat from transportation is much higher than disturbance arising from human activity (Table 1).
3.1. Habitat suitability and dynamics, model validation with variable importance
3.2. Current and historical population range
With our suitability classification, current suitable and subsuitable habitat is 140,200 ha and 298,718 ha, accounting for 24.4% of the 1,800,565 ha study area. Future suitable and sub-suitable
Surveys in the Qinling Mountains detected 1800 giant panda signs in 2000 and 1854 in 2012. Based on population range modeling the area of current and historical giant panda population
60
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
Fig. 3. Present and 2050 giant panda habitat, suitability, and the area of the CECR and FECR in the Qinling Mountains.
Table 2 The area and suitability of three ECR for the Qinling Mountains when giant pandas are considered. ECR types
Habitat types, area (ha) and percentage (%)
Suitable habitat BECR CECR FECR
109123.6 118504.4 204863.0
29.5% 28.1% 39.6%
Sub-suitable habitat
Ordinary habitat
139521.0 170645.6 144857.5
120886.4 132999.0 167117.5
37.8% 40.4% 28.0%
distribution (CPR, HPR) was 312,198 ha (2000) and 283,786 ha (2012). After merging the two population ranges in 2000 and 2012 the total area was 369,531 ha covering 56.6% of total current suitable and sub-suitable habitat, and this was used as BECR and baseline for ECR modeling (Fig. 2, Table 2).
32.7% 31.5% 32.3%
Real area of ECR(ha)
Proportion of suitable and sub-suitable habitat in its total size of study area
369531.0 422149.0 516838.0
56.6% 65.9% 70.8%
3.3. Current ECR and future ECR based on population range and habitat Using the framework and ECR criteria we delineated the scope of CECR and FECR with calculated habitat, topographic features and giant panda characteristics. We incorporated 33,498 ha of suitable
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
and sub-suitable habitat into the scope of the ECR and identified the area of current ECR as 416,719 ha. Considering the importance to ecological service and biodiversity of areas above 3100 m and land needs in area at low elevations, we excluded the 3400 ha area below 1200 m out of the ECR which giant pandas unlikely use and added the 13,941 ha area above 3100 m into the scope of the ECR. After leaving 5111 ha of area within above ECR likely needed by people according to a national guide, the real scope of CECR for the Qinling Mountains was 422,149 ha with two patches covering 67% of the area of the current giant panda reserve network. The CECR includes 56.6% of total current suitable and sub-suitable habitat and includes 94% of the current population range and 99% of field signs. Considering habitat shift due to climate change in 2050 and based on the scope of the CECR, the future ECR of the Qinling Mountains was determined to be 526,709 ha, including 109,990 ha of future suitable habitat and sub-suitable habitat outside the CECR. The future ECR covered 84% of the area of the current giant panda reserve network, population and field signs. Due to climate change, the area above 3100 m will become giant panda habitat in 2050 and is included in the scope of the ECR (Fig. 3). Removing 9871 ha occupied by people and leaving future land needs, the real FECR for the Qinling Mountains was 516,838 ha comprising 70.8% of total future suitable and sub-suitable habitat (Fig. 3, Table 2)
4. Discussion Species are the basic unit of ecosystems and form the characteristics of an ecosystem via species biodiversity and proportional relationships (Balvanera et al., 2006; Blower, 1982; Loreau et al., 2001). All ecosystem functions and services are achieved by species, including water storage, oxygen production, carbon storage and soil conservation (Maitre et al., 2008; Sessment, 2005). Furthermore, species are direct and sensitive indicators of anthropogenic and natural disturbances via population size, spatial patterns and habitat usage behavior. Species deserve to be an important basis when delineating ECR. According to the needs of species survival, conservation programs including habitat restoration, connection and human disturbance control facilitate the improvement of landscape integrity and ecosystem structure and achieve effective conservation for an ecosystem and biodiversity. A well-known example involves biodiversity and ecosystems of the upper-middle reaches of the Yangtze River that have benefited from giant panda conservation programs (State Forestry Administration, 2014; WWF, 2015). Other examples are conservation of the Komodo dragon (Varanus komodoensis) for the restoration of ecosystems across eastern Indonesia (Ariefiandy et al., 2014), the Asian elephant (Elephas maximus) for the conservation of landscapes in Aceh, Indonesia, and the flying fox (Pteropus voeltzkowi) for forest protection on Pemba Island, Tanzania (BowenJones and Entwistle, 2002). We name species used to delineate ECR as target species. To ensure the significance of ECR we suggest that target species should be representative, typical and indicative to an area and include flagship species, key species, dominant species or surrogate species (Xu et al., 2014), and criteria for target species selection should be established. Over evolution, species gradually develop unique spatial spanning topography, food resources, habitat, behavior, home range, migration and predator avoidance (Allee and Schmidt, 1950). According to the requirements of environmental variables by biological characteristics, we can determine a specific and suitable area to guarantee long-term survival of a species that includes the current population and habitat and future habitat affected by the climate or other factors (Boitani et al., 2008; Hirzel et al., 2006; Hole et al., 2009). Using the area required for species long-term survival makes the ECR approach and policy more acceptable and
61
convincing, and solves the isolation and fragmentation issues in previous conservation zoning with species population development and safety. Further, species’ biology (especially some rare and endangered animals and plants) is often been well studied and can provide theoretical responses to ECR delineation. The persistence of a species depends on the integrity and connectivity of the ecosystem where it is located, and a species is unlikely to survive long term in a fragmented, poorly suitable environment (Hanski and Ovaskainen, 2000). The integrity and connectivity of landscape effectively guarantees the results of the ECR by providing an efficient and competent ecosystem for the long-term survival of a species. To maintain the integrity and connectivity of ECR, we removed western patches in the basic ECR that did not contain suitable giant panda habitat and included the perforation of the population range into the scope of CECR (Fig. 2); this area between the southeast and northeast of future suitable habitat was delineated into the scope of FECR too. Considering shrinking habitat in the central Qinling Mountains and shortages in land resources, the future ECR may remove this area located to the south of Changqing and Foping reserve (Fig. 3). We also noted that the northeastern part of the future ECR of the Qinling Mountains may become high quality habitat in 2050 but is currently outside the reserve network. This area should be included in future conservation planning by designing a new reserve or a dispersal corridor to link future giant panda habitat. With the increase in the proportion by suitable and sub-suitable habitat in the BECR, CECR and FECR, the delineated ECR covered the main high quality habitat for giant pandas and set a strong foundation for its persistence until 2050. It also shows that our ECR achieved its original aim and our method is of robust reliability. Technically, due to differences among the biological characteristic of species, the model of population range and habitat suitability in our study is not likely suitable for other animals. A franchised model and parameters based on the characteristic and study results of different targeted species during ECR modeling are needed. A similar situation lies in determining the driving factors of habitat dynamics. We also note that one limitation is that our method focused on species’ spatial requirements and less on ecosystem and landscape requirement. Therefore, the last step in optimizing the ECR based on expert knowledge and ecological needs will be important to improving the accuracy of ECR and should be emphasized during delineation. Species are not the aim of ECR policy and methodology, but are the outputs of the management of ecosystems. Based on long-term survival of giant pandas in the Qinling Mountains our study provides a framework for delineating a scientific ECR using population and habitat at different scales and primary habitat variables. Our method is ahead of the current national guideline in that (1) we expanded the temporal and spatial scale of ECR from the present to the future, and population to habitat while considering ecological and landscape requirements; (2) we concentrated on the long-term survival of a species and avoided the complicated process of ecosystem assessment; and (3) we demonstrated that ECR delineation is feasible and operational while improving theoretical support. We hope our method and reasoning can be incorporated into national guidelines and applied to ECR delineation across the whole of China.
Acknowledgements This study was funded by the Fourth National Giant Panda Survey initiated by the Chinese State Forestry Administration and WWF-China (No. 10003187). We are thankful for support from the Department of Nature Reserve Governance and Wildlife Protection, Chinese State Forestry Administration. We also acknowledge all
62
M. Gong et al. / Ecological Modelling 345 (2017) 56–62
staff involved in the fourth national giant panda survey for their hard work, wisdom and dedication in 2012. References Allee, W.C., Schmidt, K.P., 1950. Book reviews: principles of animal ecology. Science 111, 424–425. Ariefiandy, A., Purwandana, D., Seno, A., Chrismiawati, M., Ciofi, C., Jessop, T.S., 2014. Evaluation of three field monitoring-density estimation protocols and their relevance to Komodo dragon conservation. Biodivers. Conserv. 23, 2473–2490. Bai, Y., Jiang, B., Wang, M., Li, H., Alatalo, J.M., Huang, S., 2015. New ecological redline policy (ERP) to secure ecosystem services in China. Land Use Policy 25, 348–351. Balvanera, P., Pfisterer, A.B., Buchmann, N., He, J.S., Nakashizuka, T., Raffaelli, D., Schmid, B., 2006. Quantifying the evidence for biodiversity effects on ecosystem functioning and services. Ecol. Lett. 9, 1146–1156. Blower, J.G., 1982. Population ecology: a unified study of animals and plants, third edition. J. Ecol. 70, 226–229. Boitani, L., Sinibaldi, I., Corsi, F., Biase, A.D., Carranza, I.D.I., Ravagli, M., Reggiani, G., Rondinini, C., Trapanese, P., 2008. Distribution of medium- to large-sized African mammals based on habitat suitability models. Biodivers. Conserv. 17, 605–621. BowenJones, Evan, Entwistle, Abigail, 2002. Identifying appropriate flagship species: the importance of culture and local contexts. Oryx 36, 189–195. Fan, J., Tao, A., Qing, Ren, 2010. On the historical background, scientific intentions, goal orientation, and policy framework of major function-Oriented zone planning in China. J. Res. Ecol., 289–299. Fan, J., Li, J., Xia, R., Hu, L., Wu, X., Li, G., 2014. Assessing the impact of climate change on the habitat distribution of the giant panda in the Qinling Mountains of China. Ecol. Modell. 274, 12–20. Gao, J., 2015. Exploring the delineation and supervision of ecological protection redlines in China. Biodivers. Sci. 23, 705–707. Hanski, I., Ovaskainen, O., 2000. The metapopulation capacity of a fragmented landscape. Nature 404, 755–758. Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G., Jarvis, A., 2005. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 25, 1965–1978. Hirzel, A.H., Lay, G.L., Helfer, V., Randin, C., Guisan, A., 2006. Evaluating the ability of habitat suitability models to predict species presences. Ecol. Modell. 199, 142–152. Hole, D.G., Willis, S.G., Pain, D.J., Fishpool, L.D., Butchart, S.H.M., Collingham, Y.C., Rahbek, C., Huntley, B., 2009. Projected impacts of climate change on a continent-wide protected area network. Ecol. Lett. 12, 420–431. Jørgensen, S.E., 2001. Recent Developments in System Ecology. Springer Berlin Heidelberg, pp. 155–169. Jantke, K., Schleupner, C., Schneider, U.A., 2011. Gap analysis of European wetland species: priority regions for expanding the Natura 2000 network. Biodiver. Conserv. 20, 581–605. Jian, J., Jiang, H., Jiang, Z., Zhou, G., Yu, S., Peng, S., Liu, S., Liu, S., Wang, J., 2013. Predicting giant panda habitat with climate data and calculated habitat suitability index (HSI) map. Meteorol. Appl. 21, 210–217. Jie, F., Wei, S., Kan, Z., Dong, C., 2012. Major Function Oriented Zone: new method of spatial regulation for reshaping regional development pattern in China. Chin. Geog. Sci. 22, 196–209. Li, R., Xu, M., Wong, M.H.G., Qiu, S., Sheng, Q., Li, X., Song, Z., 2014. Climate change-induced decline in bamboo habitats and species diversity: implications for giant panda conservation. Divers. Distrib. 21, 379–391. Liao, J.X., Zhou, Z., Department, B., 2015. Delineation and implementation of ecological protection redline in hunan province. J. Cent. South Univ. For. Technol. 21, 281–282. Lin, Y., Fan, J., Wen, Q., Liu, S., Li, B., 2016. Primary exploration of ecological theories and technologies for delineation of ecological redline zone. Acta Ecol. Sin. 36 (5), 1244–1252. Liu, D., Lin, N., Zou, C., You, G., 2015. Development of foreign ecological protected areas and linkages to ecological protection redline delineation and management in China. Biodivers. Sci. 23, 708–715.
Liu, G., Guan, T., Dai, Q., Li, H., Gong, M., 2016. Impacts of Temperature on Giant Panda Habitat in the North Minshan Mountains. Ecology and Evolution. Loreau, M., Naeem, S., Inchausti, P., Bengtsson, J., Grime, J.P., Hector, A., Hooper, D.U., Huston, M.A., Raffaelli, D., Schmid, B., 2001. Biodiversity and ecosystem functioning: current knowledge and future challenges. Science 294, 804–808. Maitre, D.C.L., Milton, S.J., Jarmain, C., Colvin, C.A., Saayman, I., Vlok, J.H., 2008. Linking ecosystem services and water resources: landscape-scale hydrology of the Little Karoo. Front. Ecol. Environ. 5, 261–270. Mittermeier, R.A., Myers, N., Mittermeier, C.G., Robles Gil, P., 2009. Hotspots: earth’s biologically richest and most endangered terrestrial ecoregions. Libr. J., 237–238. Olson, D.M., Dinerstein, E., 1998. The global 200: a representation approach to conserving the earth’s most biologically valuable ecoregions. Conserv. Biol. 12, 502–515. Pan, W., Mcshea, W.J., Garshelis, D., Wang, D., Harris, R.B., 2014. A Chance for Lasting Survival. Parry, M.L., Canziani, O.F., Palutikof, J.P., Linden, P.J.V.D., Hanson, C.E., 2011. Climate change 2007: impacts, adaptation and vulnerability: contribution of working group II to the fourth assessment report of the intergovernmental panel on climate change. Summary Policymakers. Astron. Nachr. 332, 697–705. Phillips, S.J., Dudík, M., 2008. Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography 31, 161–175. Qi, D., Zhang, S., Zhang, Z., Hu, Y., Yang, X., Wang, H., Wei, F., 2011. Different habitat preferences of male and female giant pandas. J. Zool. 285, 205–214. Rebelo, H., Tarroso, P., Jones, G., 2010. Predicted impact of climate change on European bats in relation to their biogeographic patterns. Global Change Biol. 16, 561–576. Schaller, G.B., 1985. Giant pandas of wolong. Bioscience 36. Sessment, M., 2005. Ecosystems and human well-being. Eco. Human Well-being Framework Assess. 42, 77–101. Si, Q., Li, M.C., Zhang, G.Y., Liang, S.X., Sun, Z.C., 2013. Pilot study on delimited method of marine regional ecological red line. Adv. Mater. Res. 807 (807–809), 728–731. Songer, M., Delion, M., Biggs, A., Huang, Q., 2012. Modeling impacts of climate change on giant panda habitat. Int. J. Ecol. 2012, 1–12. State Forestry Administration, 2006. The Third National Survey Report on Giant Panda in China. Science Press, Beijing (In Chinese). State Forestry Administration, 2014. The Fourth National Survey Report for Giant Panda in China (Announced but not published In Chinese). Store, R., Kangas, J., 2001. Integrating spatial multi-criteria evaluation and expert knowledge for GIS-based habitat suitability modelling. Landscape Urban Plann. 55, 79–93. ˜ A., Winkler, J.A., Li, Y., Xu, W., Ouyang, Z., Liu, J., 2013. Tuanmu, M.N., Vina, Climate-change impacts on understorey bamboo species and giant pandas in China/’s Qinling Mountains. Nat. Clim. Change 3, 249–253. WWF, 2015. Understanding Biodiversity in the Conservation of Flagship Species of Giant Pandas. http://mp.weixin.qq.com/s? biz=MjM5NTkxMTIyMQ==&mid=205378773&idx=1&sn=44d1e265 bae41b394733697dfff6f3f8. ˜ A., Qi, Z., Ouyang, Z., Liu, J., Wei, L., Hui, W., 2014. Evaluating Xu, W., Vina, conservation effectiveness of nature reserves established for surrogate species: case of a giant panda nature reserve in qinling mountains, China. Chin. Geog. Sci. 24, 60–70. Xu, D., Zou, C., Xu, M., You, G., Wu, D., 2015. Ecological security pattern construction based on ecological protection redlines. Biodivers. Sci. 23, 740–746. Yan, S.G., Lin, N.F., Shen, W.S., 2014. Delineation and protection of ecological red lines in jiangsu province. J. Ecol. Rural Environ. 30 (3), 294–299. Yang, B., Gao, J., Zou, C., 2014. The Strategic Significance of Drawing the Ecological Protection Red Line. China Development. Yue, X.M., Long, C.J., Long, G.J., Qian, Y.E., 2012. Study on ecological compensation model for major function oriented zones. Chin. J. Eco-Agric. 20, 1404–1408. Zheng, H., Ouyang, Z., 2014. Practice and Consideration for Ecological Redlining. Bulletin of Chinese Academy of Sciences. Zhu, L.F., Hu, Y.B., Zhang, Z.J., Wei, F.W., 2013. Effect of China’s rapid development on its iconic giant panda. Chin. Sci. Bull. 58, 2134–2139.