Plant conservation priorities of Xinjiang region, China

As an important region in the Silk Road, Xinjiang is getting a good chance of developing economy. However at the same time, its natural environment is facing a big challenge. To better protect the plant diversity, it is urgent to make a thorough conservation plan. With a full database of vascular and medicinal plant distributions and nature reserve plant lists and boundaries in Xinjiang of China, we analysed the plant diversity hotspots, protection gaps and proposed the plant conservation priorities of this region. Differed from the widely accepted viewpoints that lots of plants were not included in nature reserves, we found that most of the plants ( > 90%) were actually included in the current nature reserves. We believe that compared with establishing more nature reserves, improving the management of the existing ones is also important. Furthermore, the very few unprotected plants ( < 10%) were distributed mostly in the regions of Aletai, Tacheng, Zhaosu, Manasi, Qitai and Hetian which could be the future conservation priorities.


Biodiversity hotspots and conservation
Biodiversity is a great treasure for human being.While with the fast economic development, biodiversity is destroyed severely by resource over-exploitation, habitat destruction and other human activities.One good way to protect biodiversity is to build nature reserves.Prior to this, a scientific planning is important, e.g., to have a comprehensive plan on where to build nature reserves and how big the nature reserves should be.In order to support most biodiversity with relatively least cost, nature reserves are better built in regions with high biodiversity, i.e. in biodiversity hotspots (Myers, 1988;Myers et al., 2000).
With the development of GIS (Geographic Information System) method, biodiversity hotspots are more easily defined to guide the nature conservation.Through the intensive species distribution data collection, via the application of GIS, people could define the high biodiversity regions fast and reliably.GAP (Geographic Approach to Protection of biodiversity, Burley, 1988;Scott et al., 1993) analysis is also a broadly used method in biodiversity conservation.A fundamental assumption of the method is that the best time to decrease the probability of a species' extinction due to human activities is well before its population is diminished to the point of endangerment (Jennings, 2000).So GAP analysis tries to find species that are not well protected whether the species are meanwhile endangered or not.Through all these methods, the destination of conservation is to keep common species always common.
With some newly developed methods in macroecology and some good quality regional plant checklists, distribution patterns of vascular plants were being acquired, e.g., in Xinjiang, China (Li et al., 2011(Li et al., , 2013)).With these data available, combined with the nature reserve data, currently it would be possible to investigate the biodiversity hotspots, conservation gaps and conservation priorities.

Medicinal plant conservation in Xinjiang
Species having economic values is usually over-used which makes huge biodiversity loss.Medicinal plants are among the species facing over-exploitation.In 1970s, paclitaxel in Taxus chinensis var.mairei was found an active constituent of cancer-combating, and then this species was severely logged to a very low stock.It was then listed as endangered species in China Plant Red Data Book (Fu & Jin, 1991).Sinopodophyllum hexandrum has limited use in traditional Chinese medicines, and it has always been a common species without many disturbances.But all of a sudden, when podophyllotoxin was found high in this plant, it was very soon severely depleted and listed as a rare species in China Plant Red Data Book (Fu & Jin, 1991;Li et al., 2005).If there was basic research on normal species, when they were even not threatened, and some reserves could be set beforehand, stocks of these species probably could not be so easily going down to extremely low, and status of these species could not be so quickly going from common to rare or endangered.
Xinjiang is a region with many minorities.There is more possible plant utilizing ways in different cultures.Besides traditional Chinese materia medica (CMM), Uygur medicine (UM) and Kazak medicine (KM) were also very popular in Xinjiang region.As the unplanned exploitation of these medicine source plants, vegetation in the region was highly destroyed, for example, the high mountain ecosystems including Saussurea involucrate, desert ecosystems including Cynomorium songaricum, Cistanche deserticola, and piedmont regions with Ferula fukanensis, Ferula sinkiangensis were all damaged due to the specific medicinal plant gathering.
The distribution patterns of vascular and medicinal plants were mapped and basic conclusions were given, for example, high plant richness in mountainous areas (Li et al., 2011(Li et al., , 2015)).While for conservation, some important issues are still not known, for example, what plants are protected in nature reserves, what plants are not, and where are the unprotected plants.Here in this research, with data of plant distributions and current nature reserves, we comprehensively analysed the vascular plant and medicinal plant diversity hotspots, protection gap species and protection gap areas in Xinjiang to give suggestions for plant conservation planning of this region.With this case study, we also would get a systematic method to obtain plant conservation strategies in arid regions.

Study area
Xinjiang was anciently the main region of "Xiyu" (Western Regions) where there were various minorities.It was an important place in ancient Silk Road, especially the regions around north and south of Tianshan mountains.In 2014, jointly submitted by China, Kazakhstan and Kyrgyzstan, "Silk Roads: the Routes Network of Chang'an-Tianshan Corridor" was inscribed on the World Heritage List.The ecosystems in Xinjiang are also very important.Tianshan mountains, including areas of Bogda, Tomur, Kalajun-Kuerdening, and Bayinbuluk, were announced as world natural heritage in 2013.
With the great development chance under the "Belt & Road" economic policy of Chinese government, more infrastructure projects are going which makes the environment protection planning urgently needed in this region.There are currently more than 30 nature reserves (national and provincial level) in Xinjiang region (Figure 1).They are distributed very unevenly with some nature reserves targeted for wild animals occupying very large areas, for example, Lop Nur Wild Camel National Nature Reserve (4.5×10 4 km 2 ) and Altun Mountain National Nature Reserve (6.7×10 4 km 2 ).
From an intensive literature survey, we collected scientific reports (or project planning) of 29 nature reserves (Table 1), including most of the national and provincial level reserves and part of the reserves in planning.They were mostly managed by State Forestry Administration.The boundaries and plant lists of each reserve were acquired from these literatures.

Data analyses
Based on the plant distributions (Li et al., 2011(Li et al., , 2013(Li et al., , 2015)), with some 400 more species altitude ranges, species distributions were re-mapped and 5% of the highest plant richness regions were selected as the diversity hotspots.These were conducted for vascular plants, CMM, UM and KM, respectively.
In this research, we would deem species not appear in nature reserves the protection gap species, i.e. the species not being protected.While for a species appearing in a nature reserve, whether it is practically well protected or not, we deem it protected species.
The ranges of some nature reserves were digitalized and species lists of the nature reserves were carefully structured and checked.Then two methods were used to get the protection gap species.Method 1 uses mainly the nature reserve species lists as basic data and method 2 uses mainly the species richness data clipped by the nature reserve ranges.
Method 1: Merge the plant lists from these reserves and remove the duplicates, then species appearing in nature reserves of Xinjiang were obtained.Compared with the species of Xinjiang (Commissione Redactorum Florae Xinjiangensis, 1992-2011), plants not appearing in nature reserves were considered protection gap species.
Method 2: Firstly, digitalize the nature reserve ranges according to the range maps in scientific reports (Figure 1a).For the nature reserves without range maps in the scientific reports, their ranges were simulated with buffer analysis in GIS.In particular, according to the area of the specific nature reserve, the range was drawn as an equal area circle with the management station as circle center.Secondly, by overlaying the ranges of each plant, we got the species that fell into the ranges as species in nature reserves.Then we overlaid the data of all nature reserves, removed the duplicates and the species lists of nature reserves could be obtained.Lastly, compared the lists with the species of Xinjiang (Commissione Redactorum Florae Xinjiangensis, 1992-2011), plants not appearing in nature reserves were considered protection gap species.
Combined the two results above, and after an expert review, we got the final list of protection gap species.Then the distributions of these gap species were mapped.Regions with high richness of these gap species were defined conservation priority regions for the future nature reserve construction.

Vascular and medicinal plant biodiversity hotspots
High medicinal plant diversity regions were mainly in Altay mountains, Yili valley, part of mountains in west Jungar, and part of northern slopes of Tianshan mountains (Figure 2a,b,c).High vascular plant diversity regions were also concentrated mainly in mountains (Figure 2d).The hotspot regions for vascular plant, CMM, UM, and KM were concordant (Figure 2e).

Vascular and medicinal plant protection gaps
After comparing the species list of nature reserves and Florae Xinjiangensis (Commissione Redactorum Florae Xinjiangensis, 1992-2011), we found that about 24% of vascular species were not under the protection of nature reserves.From this method, we further found that about 105 species of CMM (12% of whole CMM) not appearing in nature reserves.Luckily, most species of UM and KM were in the ranges of nature reserves (Table 2).By sending the results to experts, we modified the list of species not protected.The experts went over the lists and removed the ones that were ever seen in a nature reserve with their field experiences.The experts eventually removed 694 vascular species from the not protected plant lists.From this method, there were finally 175 of vascular plant species, including 73 of medicinal plant species, identified not appear in reserves.With the second method, we differentiated the species from the plant distribution maps by using the digitalized nature reserve ranges.We found 419 vascular plants, including 40 CMM (e.g., Physochlaima physaloides, Lycopodium japonicum, Matricaria recutita), 1 UM (Hyssopus cuspidatus var.cuspidatus), and 9 KM (e.g., Echinops sphaerocephalus, Sonchus palustris, Daphne altaica) species were not conserved in nature reserves.These results were also sent for an expert review.
Combining the results of the above two methods, and after the expert review, we preliminarily concluded that there were totally about 262 species of vascular plants, including 124 species of medicinal plants (CMM, UM, and KM) not appearing in nature reserves.From this result, we could conclude that most of the native vascular plants (93%) and medicinal plants (90% of the whole medicinal plants) were distributed in nature reserves.
By mapping the distributions of these plants, we found that northern Xinjiang had more gap species than southern Xinjiang.After the careful consideration, we defined six regions waiting for more attentions, including northwest of Tacheng basin (belonging to Tacheng city), southwest of Yili valley (belonging to Zhaosu county), northwest of Altay mountains (belonging to Aletai city), middle Tianshan Mountains in the northern slope (mainly in Manasi county and Qitai county) and part of Kunlun mountains (especially Hetian area) (Figure 3).The diversity of gap species in Hetian area was lower than the regions in northern Xinjiang, but it was relatively high in Pamir Plateau and Kunlun Mountains, so we listed this area also as high gap species region.

Plant protection status evaluation
Vascular and medicinal plant biodiversity was formerly mapped and the distribution patterns were clear in Xinjiang region (Li et al., 2011(Li et al., , 2015)).Some conclusions were obtained with concrete data, for example, the high biodiversity regions were concentrated in mountainous areas.This could guide the regional conservation policy.But how the species was protected in this region was not evaluated, and what species was not protected was still not clear.Based on the former research, by further collecting the nature reserve species lists and boundaries, this study comprehensively analysed the presence of species within and out nature reserves, and mapped the locations of species not in reserves.Analysis was conducted for plants and medicinal plants, respectively.This is beneficial for the future conservation plan of vascular and medicinal plants.
Some important plants were evaluated in Chinese national level nature reserves, for example, Yuan et al. (2009) found that 20% of the Chinese state key protected plant species was not included in national nature reserves.Protection of orchid species in China were also evaluated (Qin et al., 2012).Although being a highly concerned family, orchids were very poorly protected, with nearly 50% of the orchid species being not well protected in nature reserves (Qin et al., 2012).The large scale methods were not very precise, for example, data from scientific reports was not of good quality, not having enough new references and materials included in the analysis, and without specimen identification (Jin 2012).New protection status evaluation of orchid species in China was conducted recently (Zhang et al., 2015b) and found that 154 of 1449 orchid species were not included in nature reserves.This means about 11% of orchid species was till now still without any protection in China.Similar research found that about 12% of threatened species in China was not covered by reserves (Zhang et al., 2015a).
Comparing species lists of nature reserves with targeted protection species is the most frequently used method for evaluating species protection status.With this method, most studies got high percentages of species not protected, for example, the above mentioned studies, all higher than 10% even for threatened and mostly concerned species, e.g., orchids.We also used this method as method 1 in this study.After the expert review, we found that this method under-estimated the species protection status.Caused possibly by data problems, including not enough reserve information and the species lists of reserves not containing the entire species in the reserves, some more species would be listed as unprotected species.We suggested that an expert review was needed after comparing the species list.
Not as imagined, with low nature reserve numbers in Xinjiang, there are actually only low percentages of species never appear in reserves in Xinjiang.It includes the species of some specific variations in taxonomy, e.g., Hyssopus cuspidatus var.albiflorus and Artemisia sacrorum var.messerschmidtiana; and some important medicinal plant species, e.g., Ferula spp.We concluded that most species in Xinjiang are currently included in reserves, but whether they were well protected still lacks evaluation.

Vascular and medicinal plant conservation
In the last six decades, Chinese government built more than 3000 nature reserves (according to the database of NSII, http://www.papc.cn/).These reserves played very important role in species conservation.But some of the reserves were not well managed, for example, there were not any investigation data, there were not workers and they even did not have fixed boundaries.Because of data deficiency, there has not been comprehensive research on all these nature reserves till now.The very few comprehensive research included only national nature reserves (Zhao et al., 2013).
Xinjiang government has built more than 30 nature reserves (national and provincial level) since 1980.The reserves cover about 15% of the total region area.The areas of the reserves are quite uneven with five reserves (> 10,000 km 2 , Lop Nur Wild Camel, Altun Mountains, Middle Kunlun Mountains, Karamayle Mountain Ungulate, and Taxkorgan) cover 75% of the total reserve area.From the Statistical Yearbook (Statistic Bureau of Xinjiang Uygur Autonomous Region, 2006, 2013), there were few reserves established after 2005.There is still big development space for reserve construction.Because of the climate change and also human activities, the landscapes and ecosystems are changing in Xinjiang, for example, the ecological functions of wetlands are reducing in Bayin bulah (Shi et al., 2015).
Using as medicines is an important reason for the species loss.So some specific measures for protecting medicinal plants are necessary.The same as China, India is a country with traditional herbal medicines broadly used.Although without systematic analysis, medicinal-plant conservation areas were tentatively put forward 10-15 years ago in India (Dhar et al., 2000;Kala, 2000Kala, , 2005)).In Pakistan, Important Plant Areas (IPAs) for medicinal and economic plants were identified (Sher et al., 2012).Here in this research, we found that some medicinal plants were not well protected, including the broadly used species Ferula fukanensis and Ferula sinkiangensis.Some more attentions should be given to these species.

Data quality and method problem
The gap analysis is based on data of regional species lists / distributions and nature reserve species lists / boundaries.Regional species data were formerly published (Li et al., 2011(Li et al., , 2015) ) while the nature reserve data were not easy to be acquired.In this study, we included part of the nature reserve information of Xinjiang.Our current nature reserve dataset included the all national ones and most of the provincial ones, while the county level nature reserves were mostly not included.This may increase the number of species not in reserves but we believe the impact from this is few.According to our investigation, the county level nature reserves were not very effectively managed.Species in these county level nature reserves were actually very possibly not very different with species out of the nature reserves.So our dataset highly represents the situations of nature reserves in Xinjiang.
During the analysis process, we found that the data quality of species lists in Xinjiang reserves was very poor.Species in the checklists of nature reserves were very frequently not the accept names.This makes mistakes on the judgement of species appears in a reserve or not.Through the expert review, nearly 80% of the gap species obtained from the species list comparison (method 1) was found actually in nature reserves.Further, the data quality problems lead to some conflict results for gap species analysis with different methods (for example, method 1 and method 2 in this research).Thus, more investigations of these nature reserves are needed.
This research defined plants in the nature reserves well protected but actually it might not.Species in the nature reserves could also be severely damaged.So, further methods are still needed in the evaluation processes, for example, detail investigations of the plant protection status in the nature reserves.

Conclusion
We preliminarily found that most plants in Xinjiang ( > 90%) appeared in nature reserves.So we concluded that besides building new reserves, it should be very important improving the management status of current nature reserves and more resource investigation should be conducted in nature reserves.And further, more attentions should be given to the 262 unprotected species.The gap species were mostly distributed in Aletai, Tacheng, Zhaosu, Manasi, Qitai and Hetian.These areas could be the possible target region for future new reserve construction.

Figure 1 .
Figure 1.Current nature reserve distribution patterns (a) and the changing patterns in 30 years (b) in Xinjiang, China.In (a), dots show the location of the nature reserve management administrations, green area is the nature reserve extent digitalized from literatures, and blue area is the nature reserve extent downloaded from the IUCN World Database on Protected Areas (WDPA, http://www.protectedplanet.net/); the data of (b) was mainly from Statistic Bureau of Xinjiang Uygur Autonomous Region (2006, 2013) and Du & Zhang (2006); the division of Xinjiang in (a) is according to Li et al. (2011).

Figure 2 .
Figure 2. Biodiversity hotspots of vascular and medicinal plants in Xinjiang, China (a, CMM; b, UM; c, KM; d, vascular plants; e, the combination of the above four).Vascular plant layer was at the bottom in (e).With more species altitudinal distribution range data in this study, there were slight differences between (d) and hotspots in Li et al. (2011).

Figure 3 .
Figure 3. Vascular and medicinal plant protection gap species distribution patterns in Xinjiang region.

Table 1 .
Nature reserves with plant species lists or ranges collected

3
International Symposium on Earth Observation for One Belt and One Road (EOBAR) IOP Publishing IOP Conf.Series: Earth and Environmental Science 57 (2017) 012034 doi:10.1088/1755-1315/57/1/012034Year: the initial founding year of the reserve; In the Management column, MEP: Ministry of Environmental Protection of the People's Republic of China, SFA: State Forestry Administration of the People's Republic of China.
Note: Latitude and Longitude is the geographic location of the management administration; Area: km 2 ;