| Carbon source/sink, namely, "carbon source" and "carbon sink", were two relative concepts, which were the result of the Kyoto Protocol. The carbon sequestration quantity of ecosystem was lower than the carbon emission, which was considered as carbon source(CSE). Similarly, the carbon sequestration quantity of ecosystem was higher than carbon emission, which was considered as carbon sink(CSK). We usually regarded ecosystem carbon source/sink as net ecosystem productivity(NEP). From a global perspective, the temporal-spatial distribution and intensity of provincial ecosystem carbon source/sink were uncertain. Meanwhile, they were the key to predict scientific problems, such as the future concentration change of CO2 and other greenhouse gases, and interactions between atmosphere and biosphere. They were also the basis of understanding water cycling, nutrient cycling and biodiversity change of earth system.Here we took Gansu province as the target, using different models, like an improved CASA model, soil microbial respiration rate estimation model, ecosystem carbon source/sink estimate model, a simple linear regression analysis model, coefficient of partial correlation, human activity intensity index and GWR model, etc. The temporal-spatial difference of ecosystem carbon source/sink in Gansu province and the influence of natural and human factors were discussed from different perspectives. Finally, we put forward corresponding suggestions and policies for improvement of ecological environment, ecological civilization construction in Gansu, and construction of ecological security barrier of the west of China and even the whole country. The main research contents and conclusions of this paper were as follows:(1) Spatial pattern and change analysis of carbon source/sink of ecosystem among the different biomes in Gansu province at the inter–monthly time scale.The spatial pattern of mean monthly NEP was consistent with the year and seasonal scales in Gansu province, and the mean monthly NEP value was the maximum(298.80 g Cm-2month-1) in July and minimum(33.13 g Cm-2 month-1) in December during 2000 – 2010. The Gansu province showed the highest(-1.19 g Cm-2month-1) and lowest(0.00 g Cm-2month-1) mean monthly NEP values changed in May and January, respectively. As for individual biomes, themaximum change of mean monthly NEP value was different, while the minimum change of mean monthly NEP value was January among months.(2) Spatial pattern and change analysis of carbon source/sink of ecosystem among the different biomes in Gansu province at the inter–seasonal time scale.There were significantly differences among spatial differentiations and changes of four seasons, reflecting that variations were decreasing form the southeast with high values to northwest with low values for the spatial distribution of the seasonal NEP from 2000 to 2010 in Gansu province. The mean seasonal NEP values for the spring, summer, autumn and winter were 137.77 g Cm-2season-1, 279.77 g Cm-2season-1, 94.59 g Cm-2season-1 and 34.72 g Cm-2season-1, respectively. As for individual biomes, the change of the maximum amplitude of the mean seasonal NEP in the forest biome was-8.98 g Cm-2season-1, while the change of the minimum amplitude of the mean seasonal NEP in the desert biome was-0.02 g Cm-2season-1 in spring. The mean seasonal NPP for the summer was characterized with decrease, however, the mean seasonal NEP in the agricultural and desert biomes slightly increased. To a certain extent, the mean seasonal NEP increased, but the growth rate was relatively lower in autumn. The mean seasonal NEP was in a stable state or lower amplitude of variation and the vegetations were in a slow "winter dormancy" during the winter.(3) Spatial pattern and change analysis of carbon source/sink of ecosystem among the different biomes in Gansu province at the inter–annual time scale.The spatial distribution of the annual NEP from 2000 to 2010 varied significantly over the region, with NPP high in the southeast, but low in the northwest, its mean value was 126.22 g Cm-2yr-1, of Gansu. The mean annual NEP values for forest, grassland, agricultural and desert biomes were 330.38 g Cm-2yr-1, 213.18 g Cm-2yr-1, 149.22 g Cm-2yr-1 and 30.79 g Cm-2yr-1, respectively. Overall, the NEP values displayed a decreasing trend. The annual NEP values with an extremely significant increase(ESI) and significant increase(SI) from 2000 to 2010 accounted for 6.02% and 16.84% of the total area, respectively. The annual NEP values with no significant change(NSC) from 2000 to 2010 accounted for 38.09% of the total area. The annual NEP values with an extremely significant decrease(ESD) and significant decrease(SD) from 2000 to 2010 accounted for 11.99% and 27.06% of the total area, respectively. As for individual biomes, the fastest change rate of NEP was in the forest biome, followed by the agricultural, grassland and desert biomes.(4) Relationship between climate factors and carbon source/sink of ecosystem in Gansu province at several different scales.In the section, we discussed the relationship between climate factors(precipitation andtemperature) and carbon source/sink of ecosystem in the different biomes(forest, grassland, agricultural and desert biomes) of Gansu province during 2000 to 2010, at different temporal(month, season and year) scales. ① In the perspective of monthly scale, the correlation of NEP and climate factors was lower on an inter-monthly level than those of NEP and climate factors on an inter-annual level. The desert biomes were mainly influenced by the factor of precipitation and the temperature was the main control factor in forest and grassland biome, while control function of precipitation and temperature were quite in the agricultural biome. ② In the perspective of seasonal scale, the summer was mainly influenced by the factor of precipitation, and then the temperature was the main control factor in spring, autumn and winter of Gansu province. The grassland, agricultural and desert biomes were mainly influenced by the factor of precipitation in summer. The forest biome was mainly influenced by the factor of temperature in winter and temperature was the main control factor in other circumstances. ③ In the perspective of annual scale, the correlation with NEP was stronger for precipitation than for temperature in Gansu province. As for individual biomes, the forest and grassland were influenced mainly by temperature, while the precipitation was the main control factor for the desert and agricultural biomes.(5) Spatial distribution of human activity intensity and its influence on carbon source/sink of ecosystem in Gansu province.The average value of human activity intensity was 0.31 in Gansu province, which was characterized by obvious regional distribution difference, reflecting that the central area was stronger than the marginal, and the southeastern regions was stronger than northwestern. As for individual biomes, the intensity of human activity was higher in the agricultural biome, and was lower in the desert biome. ① In the perspective of County administration center, human activities intensity played a positive correlation on the carbon source/sink of ecosystem in Guazhou County, Jinta County, Yugur Autonomous County of Sunan and its adjacent regions. However, it played a negative correlation in Yongchang County, Jingyuan County, Tongwei County and Wenxian County. ② In the perspective of the different biomes, the relationship between human activity intensity and carbon source/sink of ecosystem showed the most significant effect in the forest biome, relative stability in the grassland biome, the increase or decrease of staggered distribution and low-value zone in the middle of the area in the agricultural biome. The regional change of the carbon source/sink of ecosystem which was influenced by human activity intensity was uneven, and its high value point in this region was only found in a small area of the existence of oasis in the desert biome.(6) GWR analysis of influencing factors of ecosystem carbon source/sink at the county leveladministrative unit in Gansu province.The article selected 12 major indicators of reflecting society, economy and environment from 2000 to 2010, and chose main factors of the ecosystem carbon source/sink by using SPSS13.0. Among them, the influence of the total population at year-end on ecosystem carbon source/sink reflected that the variation was decreasing form the northwest with high to southeast with low, which was opposite to spatial pattern of the total population at year-end. The gross regional production affected ecosystem carbon source/sink and they were positive and negative correlation, however, the regression coefficient was just the opposite of the total population at year-end. The impact of the first industry on ecosystem carbon source/sink showed obvious negative correlation, and regression coefficient of the first industry was high in the middle, while it was low in the southeast and northwest. The second industry had positive and negative correlation with ecosystem carbon source/sink, with high in the northwest, but low in the southeast. There were lots of factors influencing ecosystem carbon source/sink at the county level administrative unit in Gansu province, which were related to the site environment, the resources endowment, economic development and relevant policies, etc.At last, the article put forward corresponding suggestions and policies from various aspects, such as the policy and system, legislation and law enforcement, transportation and management administration, etc. |
PDF Full Text Request