The contributions of carbon reduction policies were evaluated and compared for six carbon trading pilot schemes in China, in four municipalities(Beijing, Shanghai, Tianjin, and Chongqing) and two provinces(Guangdong and Hubei). The carbon emissions accounting method of the Intergovernmental Panel on Climate Change was used to calculate the actual CO2 and the support vector machine model was used to predict CO2. Chinese carbon reduction policies abated CO2 in the six carbon trading pilot schemes after the comprehensive policies came into force. However, the contribution of policies to CO2 abatement varied among regions, and the effect of carbon reduction policy on municipality pilot schemes was greater than on provincial pilot schemes. The largest contribution of carbon reduction policy to CO2 abatement was 28.3%, for the pilot carbon trading scheme in Beijing, and the smallest contribution was 3.7%, for that in Hubei. It is crucial to consider "carbon leakage" and a carbon trading linking program in order to evaluate the effects of carbon reduction policies.
Using the Normalized Difference Vegetation Index （NDVI） as an indicator of vegetation growth, we explored the characteristics and differences in the response to drought of five vegetation biomes in Northeast China, including typical steppe, desert steppe, meadow steppe, deciduous coniferous forest and deciduous broad-leaved forest during the period 1982 2009. The results indicate that growing season precipitation may be the primary vegetation growth-limiting factor in grasslands. More than 70% of the temporal variations in NDVI can be explained by the amount of precipitation during the growing season in typical and desert steppes. During the same period, the mean temperature in the growing season could explain nearly 43% of the variations in the mean growing season NDVI and is therefore a dominant growth-limiting factor for forest ecosystems. Therefore, the NDVI trends differ largely due to differences in the vegetation growth-limiting factors of the different vegetation biomes. The NDVI responses to droughts vary in magnitude and direction and depend on the drought-affected areas of the five vegetation types. Specifically, the changes in NDVI are consistent with the variations in precipitation for grassland ecosystems. A lack of precipitation resulted in decreases in NDVI, thereby reducing vegetation growth in these regions. Conversely, increasing precipitation decreased the NDVI of forest ecosystems. The results also suggest that grasslands under arid and semi-arid environments may be more sensitive to drought than forests under humid environments. Among grassland ecosystems, desert steppe was most sensitive to drought, followed by typical steppe; meadow steppe was the least sensitive.
