| Permafrost region in the northeast China is a typical of the mid-high latitude permafrost regions,and it is rich in biological resources,wetland resources and tourism resources,etc.In recent years,due to the interaction of human interference and natural factors,the local natural environment has been undergoing dramatic changes,especially the climatic environment and soil environment.Changes in climate and soil environment are closely related to changes in permafrost.Rising temperature could cause permafrost degradation.Therefore,it is of great scientific significance to carry out research on the permafrost warming and its effects on greenhouse gas emissions in Northeast China and identify the main driving factors causing the permafrost degradation in the study area.Aimed at the spatial complexity and heterogeneity of the permafrost system in Northeast China,this thesis applied the Bayesian belief network to assess and predict the interacting factors contributing to permafrost thaw and the greenhouse gas production effects.On the basis of summarizing the previous studies about the permafrost thaw,the indicators representing the impact of permafrost degradation and greenhouse gas production were screened,and a Bayesian belief network model was constructed based on the index system.On this basis,the present status of permafrost degradation in Northeast China was evaluated.The main causes of permafrost degradation and the impact on greenhouse gas production were systematically analyzed.Using the spatial analysis function of ArcGIS software analyzed the spatial distribution of permafrost thaw in northeast China.The relationship between the index and the permafrost thaw and greenhouse gas production was also explored from multiple perspectives.Sensitivity analysis of the model showed that:（1）permafrost degradation had high sensitivity to ground temperature and methane,respectively:27.3%,28.8%;（2）in terms of greenhouse gas production,the sensitive node of carbon dioxide was soil moisture,and methane was most sensitive to permafrost thaw with a sensitivity index of 30.6%,followed by ground temperature,the sensitivity index was10.9%;（3）Climate feedback was saw as a sensitive node for methane,which in turn was the most sensitive node for climate feedback.Under the Bayesian belief network,different scenarios about air temperature,ground temperature and soil moisture were set and predict the state of permafrost degradation and greenhouse gas emissions in the whole permafrost region and three kinds of permafrost region in northeast China,which can provide a certain scientific basis for permafrost and ecological environment protection in the northeast China.Permafrost in the northeast China can be divided into three types:large continuous permafrost,large-island permafrost,and island permafrost.Mohe,Genhe and Jiagedaqi were selected as the three types of permafrost representative points.The analysis shows that the sum of the medium and high probability of permafrost thaw of Mohe,Genhe and Jiagedaqi frozen soil are:19.24%,95.1%,57.74%,respectively.In terms of the probability of greenhouse gas emissions,the high probability of CO2emissions from Mohe is 40.8%,that of Genhe City is 23.9%,that of Jiagedaqi is 34.6%;the sum of medium and high probability of Jiagedaqi,Genhe and Mohe CH4 emissions is 70.51%,65.82%and 44.25%,respectively.In terms of spatial distribution,the probability of permafrost degradation increases from south to north,which is consistent with the law of latitude differentiation.The spatial distribution of methane production probability is very similar to the spatial variation of permafrost thaw probability.The probability of carbon dioxide production has a longitude difference law.From east to west,the probability of carbon dioxide production is gradually increasing,and the probability of carbon dioxide production is particularly high in Inner Mongolia.The results of the scenario analysis show that:（1）From past to present and to future,as the temperature rises,the sum of medium-high probability of permafrost thaw increases from 55.40%to 78.80%in permafrost region in northeast China;With ground temperature rising,the medium-high probability of permafrost thaw increased from 39.21%to 88.8%to 96.3%,an increase of 57.09%;the change of soil moisture has little effect on the permafrost,the high probability of permafrost thaw is less than1%.（2）From the past to the future,under the different scenarios about air temperature,ground temperature,and soil moisture,the high probability of methane production increased by 8.2%,36.72%and 8.2%,respectively;the ground temperature has the greatest impact on methane production;the high probability of carbon dioxide production increased by 1%.3%,52.11%,respectively,which shows the impact of soil moisture on carbon dioxide can’t be ignored.（3）From the perspective of different types of permafrost,the influence of ground temperature on permafrost in Mohe and Jiagedaqi is much greater than that of Genhe.The soil moisture has little difference in greenhouse gas emissions from the three regions,but it has a greater impact on Mohe methane emissions.A series of scenario analysis shows that it is an effective method to assess and predict the permafrost thaw and the impact on greenhouse gases based on the Bayesian belief network.In summary,the Bayesian belief network model can synthesize all kinds of natural environment information and can be effectively integrated to infer the causal relationship between various factors,providing a new idea for studying permafrost degradation prediction. |
