Prediction Of Suitable Distribution Areas Of Tylototriton Shanjing In Yunnan,Based On Climate Change

Ectotherms are the most vulnerable group to climate change,and one of the commonly reported responses of them to climate change is shift the geographic range.The global average surface temperature will increase by 0.3-4.8?by the end of this century according to the Fifth Assessment Report(AR5)of Intergovernmental Panel on Climate Change(IPCC)if the greenhouse gas emission was not controlled.Tylototriton shanjing(Amphibia,Caudata,Salamandridae)is a second-grade animal under state protection,which is mainly distributed in Yunnan Province,China.At present,there is still not a comprehensive understanding about the distribution of them due to the frequent discontinuities or gap in the intraspecific distribution.additionally,climate change has caused significantly impact on biodiversity,and amphibians are more sensitive to that.Will climate change affect the geographical distribution of Tylototriton shanjing in the future and how serious it will be,this still needs further study.Finding a shelter for T.shanjing form climate change by analyzing and assessing the species distribution under ongoing climate change scenarios thus reduce the effects of climate on T.shanjing,and this will be valuable for protecting them.In this study,the Maximum entropy model(MaxEnt)was used to simulate the potential distribution of T.shanjing by analyzing the benchmark climate data in Yunnan Province.There were three ways we used to assess the accuracy of the prediction results:1)The value of area under the ROC curve(AUC),2)the distribution range defined by IUCN,3)the relevant references and the data of professional database.To identify the environmental variables which might have potential effects on T.shanjing distribution by Jackknife Method.We chose three representative CO₂ concentration pathways under CCSM4 climate model to predict the suitable distribution,distribution change trend and the climate change shelter of T.shanjing in the future(2050),The results were as follows:(1)The distribution range of T.shanjing was wildly with the current climatic background(1950-2000),Mainly concentrated in the west,northwest and southwest of Yunnan,and there are also certain suitable distribution areas in central Yunnan.The distribution area is about56931.89 km²,accounting for 14.85%of Yunnan Province.A total of fourteen cities:such as Baoshan City,Lincang City and Dali Bai Autonomous Prefecture were mainly involved,As the prediction results shown,Xiangeri-La,Fugong,Weixi,Huaping,Yongren and Luquan have suitable conditions for the distribution.However,whether there is distribution of T.shanjing still needs further field investigation and materials consult to confirm(2)After running the model for 10 times,the average training AUC is 0.937and the average test value of AUC is 0.873,this result shows that the model has a good simulation effect.The prediction was analyzed again by superimposing the distribution range defined by IUCN after comparing the references with the professional database.The results showed that the prediction was highly matched with the data records and the distribution range gived by IUCN,indicating that the prediction results were close to the real geographical distribution of T.shanjing.(3)The Jackknife method showed that the distribution of T.shanjing was greatly affected by altitude,annual mean precipitation(Bio12),monthly mean temperature difference between day and night(Bio2),isothermal(Bio3),warmest month and highest temperature(Bio5)and annual temperature range(Bio7).The response curve showed a steep single peak and the most suitable area was narrow.It showed that T.shanjing was highly sensitive to change of climate temperature.(4)in the Last Glacial Maximum,T.shanjing was widely distributed,The distribution area is about 27 000 km²,accounting for 59.35%of Yunnan Province.the change tendency results show that from the LGM to the current,this species suitable area and habitat suitability were reduced,the loss of area of about 182 000 km²,reduce the rate about 80%,the central region of the appropriate area to reduce seriously.The value of₍(6)was 0.25,indicated that the distribution area was not stable.Although the whole area contracted,it also showed a trend of small expansion to northwest Dali and Lijiang city.(5)the current suitable habitat(SH₍₈₎of T.shanjing will be lost in large numbers with the climate changing.Until 2050,the SHc will lost about 88.89%,95.26%,96.62%,respectively under the three emission scenarios(RCP2.6,RCP 4.5,RCP 8.5).The loss proportion is proportional to the intensity of greenhouse gas emission scenario,and with the increase of greenhouse gas emission intensity,the loss of suitable distribution area increased and the suitable distribution area decreased.Based on the analysis of the changes of altitude,longitude and latitude of the suitable distribution area in 2050,it was found that the suitable distribution area will move significantly to high altitude under three emission scenarios in 2050,but the changes in longitude and latitude were not consistent,that is,different RCP scenarios(RCP 2.6,RCP 4.5 and RCP 8.5)had different longitude and latitude trends.And the change trend is significantly different from the current suitable distribution area.(6)The area of climate refugia is about 863.97 km²,Although the area is small,it covers many cities where T.shanjing are distributed,among which Baoshan City,Lincang City and Pu'er City are the most widely distributed.As nature reserves have vital function in species conservation,we superimpose nature conservation and climate refugia.It is considered that 10nature reserves can provide climate shelter,such as NanGunhe National Nature Reserve,Gaoligong National Nature Reserve and Yulong Snow Mountain Provincial Reserve.These areas will be the key area for the protection of T.shanjing.