Abstract: Based on the land use data extracted from remote sensing interpretation in Urumqi between 1995 and 2015, a prediction for the land use changes for the year 2025 and an ESV (ecosystem service value) assessment have been performed using the CA-Markov model. A further analysis of the distribution characteristics of spatial patterns in the ESV and the land use is made by means of the spatial statistical methods, such as the grid analysis, the spatial autocorrelation and Getis-Ord hot spot analysis. The relationship between the spatial autocorrelation in the ESV and that in the land use as well as the reason for the spatial distribution changes of the cold spots and the hot spots in the interannual variations are discussed. The change process of the high and low ESV values with respect to the center of gravity in the space is interpreted. The results show that:(1) Within the 30 years, certain differences had been presented among the ESV changes in different land use types in different periods, representing a fluctuation trend. However, a decreasing trend in the overall ESV was presented and the main reason for this is the decrease of the ESV provided by the cultivated lands, forests and grasslands. (2) Significant differences exist in the overall spatial distribution pattern of the ESV and, for many years, the areas with a high ESV value have been mainly distributed in the zones with a large woodland area in Urumqi County, while the areas with a low ESV value have been mainly concentrated in the zones that surround the four central districts of Urumqi (Tianshan District, Shaybak District, Shuimogou District and Toutunhe District). (3) The ESV in the study area has significant spatial positive autocorrelation and spatial clustering characteristics, and the high-high and low-low clustered distribution is highly coincident with the high-low ESV value areas. (4) The number of hot and cold spots in the ESV has been increasing year by year. The increase in the quantity of the hot spots is closely related to the ecological protection and restoration measures which Urumqi has vigorously implemented in the recent years. On the other hand, the increase in the quantity of the cold spots is of great concern, and this is closely related to the human activities such as the disorderly spread and expansion of urban built-up areas and land development.

Key words: ecosystem service value (ESV), CA-Markov, land use, spatial autocorrelation, grid analysis, Urumqi