Abstract: Based on the ecosystem service value (ESV) assessment system, geo-detector and GWR model, the spatiotemporal evolution characteristics, influencing factors and dominant factors of ESV of 107 county-level administrative units in Shaanxi province from 2000 to 2015 were analyzed. The spatial heterogeneity of the intensity of the ESV was examined. The main conclusions are as follows. (1) From 2000 to 2015, the ESV in Shaanxi province showed the characteristics of rising and then decreasing. In 2005, the ESV reached the peak in the study period, while that of 2015 saw the bottom. The supply and regulation services were the core functions of the ESV in the province. The service contents in the food production and maintenance of nutrient recycling are the two types of services. (2) The distribution of ESV over the years showed a pattern of "low in the center, while high at the edge; high in the south, but low in the north". The areas with high values were distributed in the Qinling Mountains and the Ziwuling Mountains, and those with low values were concentrated in the Guanzhong Plain. The spatial changes in the ESV are more obvious. The whole province is basically covered by negative values. The high-value areas are less concentrated and concentrated in Northern Shaanxi. The second-highest values?areas are scattered in Guanzhong and Southern Shaanxi. (3) Economic factors have the greatest impact on the ESV, with the strongest interpretation, followed by social factors, and natural factors contribute the least to the interpretation of ecosystem services. (4) The spatial intensity of the effects of various dominant factors on the ESV is different. The land reclamation rate is characterized by a gradual decline from south to north. The per capita GDP gradually increases from south to north, and the population density increases from northeast to southwest. The land reclamation rate, per capita GDP, and population density are all negatively correlated with ecosystem services. The per capita income of urban and rural residents shows that the ESV is changing from negative to positive, from north to south and gradually increases.

Key words: ecosystem service value, GWR model, Shaanxi province, influencing factor, spatiotemporal characteristics