Integrated Evaluation Of Low Carbon Development Level And Combined Prediction Of Carbon Emission In Gansu-Qinghai-Shaanxi Provinces

Greenhouse gases produced by human activities are the main cause of global warming,and it has been a worldwide consensus to take low carbon development path to cope with climate change challenges.The economic development,technology level,industrial structure and resource endowment of each region in China vary greatly,therefore,it is more relevant and feasible to explore low carbon development according to the characteristics of each province.Gansu,Qinghai and Shaanxi provinces are located in the northwest of China,in the upper and middle reaches of the Yangtze and Yellow Rivers headwater region and the upper tuyere of the northwest monsoon,so their environmental conditions have a great influence on the environmental quality of the middle and lower reaches.Therefore,a full understanding of the current status and trends for low carbon development in Gansu,Qinghai and Shaanxi is important for the low carbon transition of the three provinces and the achievement of carbon peak targets,as well as for the sustainable development of the three provinces and their middle and lower reaches.According to the literature research,there is a lack of research results on the evaluation of low carbon development level and carbon emission prediction in Gansu,Qinghai and Shaanxi provinces.Therefore,this paper focuses on the following studies:Firstly,this paper constructs a TOPSIS-entropy-CRITIC integrated model to conduct a comprehensive evaluation of the low carbon development level in three provinces.The integrated evaluation model introduces the idea of combined weighting,and determines the index weights by entropy and CRITIC methods,which overcomes the problem that the traditional TOPSIS method cannot reflect the relative importance of the indicators.The low carbon development level of Gansu,Qinghai and Shaanxi is evaluated by examining the distance of all indicators to the ideal solution.The results indicate that the low carbon development levels of Gansu,Qinghai and Shaanxi all show an increasing trend in the last decade,and the comprehensive evaluation value increase from 0.348,0.426 and 0.389 in 2011 to 0.633,0.605 and 0.546 in 2020,respectively.The contributions of economic,social and environmental on low carbon development levels improvement all show positive effects.Secondly,this paper constructs GM-GRA-DPC-PSOSVR nonlinear combination model for carbon emission prediction in three provinces.This paper constructs a single prediction model based on GM(1,1)grey prediction model,support vector machine model and error correction theory,and then constructs a GM-GRA-DPC-PSOSVR nonlinear combined prediction model using DPC algorithm and PSOSVR model,which has the effectiveness of grey model in small sample prediction and the robustness of combined model.Compared with the GM(1,1)model,the carbon emission errors MAPE of the GM-GRA-DPC-PSOSVR model are reduced by 1.42%,7.62%,and 6.18%for three provinces,respectively.The study results show that the carbon emissions of Gansu and Qinghai provinces show a decreasing trend,while the carbon emissions of Shaanxi province show an increasing trend from 2021 to 2025.