Abstract:Research on the spatial selection of the ecological compensation area is the basis of establishing ecological compensation mechanism, whereas there is scarce research regarding it in China. A reasonable method for selecting ecological compensation regional space could provide a reliable quantitative basis for the selection of priority areas of ecological compensation. Considering the ecological system service value and the economic development level of each district and county in Qinghai Province, this paper, based on the use of the ArcGIS tool and applying the method of "eco-efficiency targeting" and "priority of ecological compensation", researched the priority level of ecological compensation in all districts and counties of Qinghai Province. The research results showed that the method of "eco-efficiency targeting" is suitable for the spatial selection of the ecological compensation mechanism based on elements, such as forests and grasslands, whereas the method of "priority of ecological compensation" is more suitable for the spatial selection of regional ecological compensation. According to the method of "priority of ecological compensation", the priority compensation areas of transfer payment of national key ecological function areas of Qinghai Province are mainly concentrated in the south. The secondary compensation areas are mainly located in the middle part. The potential compensation areas are mainly in the northeast of Qinghai Province. Compared with the fund allocations to the key ecological function areas of Qinghai Province, many districts and counties in the Haixi state have not received the corresponding compensations, and the original method of fund allocation pays more attention to people's livelihood rather than the environment. In the spatial selection of ecological compensation areas, we should combine it with the ecological comprehensive compensation to put forward a reasonable ecological compensation performance appraisal system and improve the people's livelihood based on protecting the environment.