关键词: 呼伦湖, 地统计学, 地理信息系统, 空间变异, 溶解氧

Abstract: Lakes play an important role in the global water cycles. Its spatio-temporal dynamic monitoring is essential to reasonable utilization and protection of water resources. Hulun Lake， the largest and most important lake in the Inner Mongolia，China，holds the balance in the grassland eco-environmental protection. However， in the recent years, its water quality is still far from satisfactory under the influences of climate changes and human activities. Dissolved oxygen （DO） is the amount of oxygen that is present in the water. It is measured in milligrams per liter （mg/L）， or the number of milligrams of oxygen dissolved in a liter of water. Dissolved oxygen （DO） is probably the single most important water quality factor that lake managers need to understand. The development of geostatistics has provided a new tool in study of DO in lakes. This study aims at approaching the spatial distribution of DO in Hulun Lake by using the geostatistics method in order to look for one effective method to better investigate the spatial pattern of the water quality. For this purpose， the dissolved oxygen （DO） data were collected from 32 samples in Hulun Lake；　and the bordline derived from Landsat TM remote sensing image in 2011 based on MNDWI （Modified Normalized Difference Water Index） was used. Softwares of ENVI，ArcGIS，GS+ and SPSS were also applied. Based on the semivariogram function in geostatistics and Kriging interpolation method，the spatial variations and distribution patterns of DO were analyzed quantitatively；and the water quality of Hulun Lake was evaluated. The results indicate that，the optimal semivariogram model of DO is Spherical model among the four provided theoretical models of Exponential，Spherical，Linear and Gaussian model. The ratio of nugget value to still value is less than 25%，showing strongly spatial autocorrelation on DO. Spatial distribution map derived from Kriging interpolation reflects spatial variation pattern of DO，which can provide basic data for water quality monitoring and pollution control. Integrating with environmental quality standards for surface water，the water quality belongs to level IV，Fish that live or feed along the bottom require dissolved oxygen concentrations of 3 mg/L or more，so it is suitable for fish’s living under the current DO level （>3 mg/L） of Hulun Lake，and the most fitting area is located in the middle of the lake. Due to the restrictions of single water quality parameter and sampling condition，there may be some uncertainty existing in the modeled results. However，it can provide an effective approach for spatial water quality monitoring in Hulun Lake，and play an important role on the conservation and management of water resources.

Key words: Hulun Lake, geostatistics, geographic information system （GIS）, spatial variation, dissolved oxygen （DO）