Agenda 21 emphasizes the role of local authorities as an important resource in pursuit of sustainability. Key aspects in this process are the collection of information about the state of the environment and the integration of this information at the various levels of local governance to highlight useful patterns for policy planning. As it is very difficult to evaluate the environmental performance on the grounds because of too many environmental indicators, i.e. too much information, we should reduce the number of indicators by aggregating them to one or several composite environmental indexes to make the environmental information more accessible.
Based on “naturally occurring state; human pressures; environmental effects; and social responses” four environmental sustainability themes, this paper developed an indicator system including ten contributing factors and twenty-eight indicators for assessing the environmental sustainability of each province in China. In this categorical system, indicator weight was determined through analytic hierarchy processes and expert survey methods. Four sub-indices——including natural state index (NSI), human pressure index (HPI), environmental effect index (EEI), and social response index (SRI)——were aggregated through additive weighting method. Then, the composite environmental sustainability index (CEI) of each province was calculated by averaging the grey relation performance correlations between the four sub-indices and their hypothetical ideal point. Higher values of CEI demonstrated that the environment of a local province is more sustainable. At the same time, indicators in each theme were analyzed by correlation and principal component analysis (PCA) in search of meaningful underlying variables which derive regional differentiation of each sub-index. Finally, the paper conducted a cluster analysis, which identifies statistically related groups of provinces based on the similarity of sub-indices, a step that would likely prove very helpful in identifying appropriate provinces against which local environmental sustainability and policies can be benchmarked and compared.
The results demonstrated that in each province there were obvious regional differentiations of natural conditions, human pressures, environmental effects, and social responses. The NSI in each province of East and South China was greater than West and North China, a result that was mainly determined by rainfall, air temperature, water-resource index and rate of forest cover in each province. The HPI in each developed province of Eastern China was greater than in poor provinces of Central and Western China, which demonstrated significant positive correlations with economic level, namely the GDP per capita in each province. The EEI in Northwest China was greater than for other regions, and it is negatively correlated with NSI; in regions with a poor natural state, little human pressure would lead to significant ecological effects. The SRI did not show any obvious special differentiation rules——that is, economic growth would promote social response, but wasn’t decisive factor; and social response also influenced by national policy and local development strategy. Eco-environmental sustainability in each province was determined by NSI, HPI, EEI and SRI. Through grey correlation analysis of the four sub-indices, this paper developed the CEI of each province. According to 2003 data, Hainan Province, Yunnan Province, Guangxi Autonomous Region, and Jiangxi Province rate at the top ranks of CEI. These four regions are endowed with good natural states, less human pressure, unapparent eco-environmental effects, and high social responses. At the bottom ranks of CEI are four Autonomous Regions——Neimengu, Ningxia, Tibet and Sinkiang——with poor natural states and serious eco-environmental effects, along with Shanghai Municipality which exhibits high human pressure and serious eco-environmental effects. The cluster analysis classed thirty-one provinces into six groupings which could be seen as having observable similarities and thus representing a useful point of departure for policy comparisons.