Identification Of Habitat Change,core Habitats And Corridors For The Amur Tiger

The Amur tiger(Panthera tigris)is the largest tiger subspecies that distributing most northern in the world.Once widely distributed in northeastern China,they were on the very edge of extinction as affected by climate changing,anthropogenic factors and other factors.As a flagship species of biodiversity,the Amur tiger's conservation has received extensive attention at home and abroad.Since the end of the 20th century,China has successively implemented a series of ecological protection projects and measures,whereas the degree to which these efforts had an impact on the restoration of the Amur tiger habitat is still unknown.In addition,it is very challenging to effectively protect big cats in human-dominated landscapes.It is crucial to accurately identify and protect their core habitats and the corridors connecting them,these works should be based on empirical study in order to provide solid basis for conservation decisionmaking.Roads are widely distributed in the habitat of the Amur tiger,and the tiger may have to cross these roads because of its large home range.Therefore,accurate prediction of the location of road crossings is of great significance for maintaining safety,population connectivity,and facilitating the dispersal of Amur tiger to the inland of China.Based on the results of large-scale and long-term monitoring of Amur tigers and their prey,this study comprehensively used wildlife survey technology,geographic information system technology,species distribution models,ecological corridor simulation,and spatial conservation priority identify methods to carry out a region-scale research in northeast China.The results obtained in this research are as following.1.Habitat change of Amur tigerThe author and his team used several approaches including information network monitoring,genetic sample detection technology,camera-trapping methods,and snow transect survey to obtain Amur tiger occurrence data from 2000 to 2019.These data were divided into two temporal groups:2000-2005 and 2014-2019.After correcting spatial autocorrelation and sampling bias,I used an optimized maximum entropy model-MaxEnt to model the distribution of Amur tiger with occurrence data and environmental factors from 2000-2005,then projected this model to data from 2014-2019 and evaluated the effectiveness of conservation.Optimized model selected according to the AICc criterion had the combination of feature types LQ and regularization multiplier 1,three evaluation metrics AUC,TSS,and CBI all showed good model performance of these models.By comparing predictions of the two phases,it is found that in the nearly past 20 years,the area of suitable habitat for the Amur tiger has increased by 8115 km2;the area where the habitat suitability showed significant improvement is 13731 km2,of which 12398 km2 and 1333 km2 is out of and in the protected areas,respectively.The area where the habitat suitability showed significant debasement is 1438 km2.The research results showed that conservation actions in China have protected the Amur tiger habitat not only at the level of the protected areas,but also restores the habitat outside the protected areas.In addition,the results find that the fragmentation of Amur tiger in China has not been improved.Increasing of forest area and net primary productivity,decreasing of human disturbance are the driving factors of the increasing Amur tiger habitat suitability.2.Spatial conservation priority for Amur tigerBased on the monitoring results of the Amur tiger and its prey from 2012-2019,I used stack species distribution model(SSDM)to model the relative abundance of ungulate prey,the I used ensemble models to model habitat suitability for the Amur tiger.Then I used spatially synoptic connectivity simulation methods to identify core habitat for the Amur tiger and corridors among them,and ranked the importance of these core habitat patches.Finally,I identified spatial conservation priorities for the expansion of protected area network by trade-off biological and socioeconomic dimensions.The results showed that(1)ensemble model significantly improved performance of the Amur tiger habitat suitability model with AUC and TSS reached 0.935 and 0.737,respectively.The distribution of ungulate prey relative abundance had the largest influence on shaping the Amur tiger distribution with variable importance at 68.2%.Ensemble model prediction indicated that areas with high Amur tiger habitat suitability are mainly occurred in the Hunchun-Suiyang area near the Sino-Russian border,the Tianqiaoling-Muling reserve,the southern part of the Zhangguangcailing mountain,the eastern part of Wanda mountain,and the northern part of Laoyeling mountain.The resistant kernel method was used to identify the core habitat patches,among which 4 patches with an area exceeded 2000 km2,their importance ranking based on patch area and cumulative resistant kernel value were:Hunchun-Suiyang patch,eastern Wanda patch,Tianqiaoling-Muling patch,and Huangnihe-Xiaobeihu patch.Current protected area network only cover 23.4%(2387 km2)of the core habitat patches.Combining the soon-to-be established Northeast China Tiger and Leopard National Park,44.5%(4542 km2)area of the core habitats will be covered.The analysis of landscape connectivity found that the patches within the mountain system are connected to each other,while the patches in the Wandashan area lack connection with other patches in other mountain system.(3)Priority areas for expansion of the protected area after balancing biological and socio-economic factors occurred in the eastern part of Wanda Mountain,the southern part of Zhangguangcailing mountain,the LaoyelingDalongling area,and a small proportion in the northern part of Zhangguangcailing mountain and the western part of Wanda mountain.3.Identifying road crossings for Amur tiger in the Northeast China Tiger and LeopardNational Park(NCTLNP)Based on the monitoring results of the Amur tiger from 2012 to 2019,I sperately model habitat suitability for the male Amur tiger and female Amur tiger using random forest algorithm under multi-scale habitat modeling framework.Then,I used connectivity modeling approaches to identify core habitat and movement corridors networks in the NCTLNP.Finally,I idenfitied the intersection of Amur tiger movement corridor netwrok with the main road network and ranked the importance of thes intersections.The main results obtained in this chapter are:(1)the Amur tiger respond to different variables at varying,mostly at relatively large scales.(2)Multiscale random forest habitat models showed excellent performance on predicting habitat suitablity for Amur tiger,with AUC higher than 0.9 and TSS higher than 0.75,respectively.(3)The core habitats for Amur tigers are distributed in the east and southeast of the park,34.29%of tiger core habitats are covered by the established protected area network.(4)The intersection between Amur tiger movement corridor network and main road network occurrs mostly and strongly on the G331,occupying seven of the top ten intersections.(5)Based on the point where the Amur tiger actually crosses the road,the verification using the spatial randomlization method shows that using connectivity model can well predict the potential road-crossings for the Amur tiger.The results can provide a scientific identification method for the NCTLNP management authority to implement road mitigation measures,reduce landscape resistance,improve landscape connectivity,and reduce the probability of human-tiger conflict.