空间分异视角下中国碳排放强度的Markov稳态预测

Markov Steady State Prediction of Carbon Emission Intensity in China Based on the Perspective of Spatial Differentiation

摘要:

由于传统的外生分组将无法准确地刻画区域碳排放的分异状况,从而不能有效地反映碳排放的演进趋势。本研究在考虑空间地理因素驱动下,通过PS收敛方法确定中国30个省区碳排放强度的时空分异类型,进一步利用ArcGIS10.0构建Markov转移矩阵分析时空分异的区域碳排放强度的流动性,揭示碳排放强度的时空格局和动态演进的特征,为研究中国碳排放强度的收敛路径提供更加直观的情境描述。采用Matlab7.0行矩阵转换确定转移概率的估计结果,研究表明：中国碳排放强度时空格局演进中存在五种分异类型,不同类型区域碳排放强度跨区域状态转移发生的概率较小,低碳和中低碳排放强度的区域具有长期的稳定性,中国碳排放强度整体上将向低与中低碳排放强度水平的长期空间均衡状态演进。在此结论基础上,提出建立分区域有重点的碳排放强度空间预警决策系统,评测各省区碳排放强度的时空分异情况,将各省区碳排放强度的变动控制在可接受的范围内,进一步根据各省区自身特点及演进趋势制定更加有针对性的差异化碳减排措施。

关键词:碳排放强度;时空分异;PS收敛; Markov链; 稳态预测;碳减排措施

Abstract:

Due to the traditional external grouping can’t accurately describe the different conditions of regional carbon emissions, thus it is difficult to reflect the evolution trend of regional carbon emissions. On the considering of the spatial geographic driving factors, this paper determined the space differentiation type of 30 provinces’ carbon emissions intensity in China by the PS convergence method, further to build up the Markov transfer matrix by ArcGIS10.0 to analyze the internal liquidity of regional carbon emission intensity, and reveal the spatial-temporal pattern and dynamic evolution characteristics of carbon emission intensity, so as to provide the intuitive description of the convergence path of carbon emission intensity in China. This paper estimated the transition probability by Matlab7.0 row matrix transformation, and the results showed that there are five types of the spatial-temporal pattern evolution difference for carbon emission intensity in China, The low probability across the region state transition occurs among different types of regional carbon emission intensity, The region of the low carbon emission intensity has long term stability, The carbon emission intensity in China on the whole will be up to the low carbon emission intensity level of the equilibrium evolution state in the long-term space. Based on the conclusions, space warning decision system of sub regional carbon emissions intensity should be early established according to the distribution of carbon emission intensity evaluation among different provinces, so as to control the change of carbon emissions intensity within the acceptable range. The government should put forward to more differential measures to reduce carbon emission.

Key words:carbon emission intensity; temporal and spatial differentiation; PS convergence; Markov chain; steady state prediction; carbon emission reduction measures