Abstract: Water quality of the Xiaoqinghe Watershed in Shandong Province during rainy and dry seasons in the year of 2012-2013 were investigated. Based on the water quality monitoring data of the period, principal component analysis and cluster analysis were conducted to determine spatio-temporal distribution of various forms of nitrogen (N) and phosphorus (P) in the watershed. Meanwhile, spatial analysis and correlation analysis were carried out to analyze effects of types of land use on N and P export. Results show that nitrogen pollution of the watershed was very serious, with the over standard rate of total nitrogen (TN) reaching 100%. Ammonia nitrogen (NH₄⁺-N) and orthophosphate-phosphorus concentrations (PO₄^3--P) were significantly higher in the rainy season than in the dry season, whereas nitrate nitrogen concentration was in a reverse trend. Principal components (Z1) represented mainly by total phosphate (TP) and total dissolved phosphorus (TDP), explained about 50% of the total variability of water quality, while principal components (Z2), consisting mainly of TN, NH₄⁺-N and NO₃^--N, explained about 20% of the total variability. TN, TP, NH₄⁺-N, PO₄^3--P and TDP concentrations were significantly and positively related to proportion of the residential and industrial zones in the area of the watershed and NO₃^--N concentration was positively related to proportion of the cropland, but negatively to that of the forestland and grassland. In terms of spatial distribution of nitrogen and phosphorus, the watershed could be divided into three groups of sub-watersheds: Group One and Group Two were distributed mainly along the main stream and in the northern plain, and being receptors of point source pollution wherethe concentrationof N and P were generally quite high and varied sharply, and Group Three was mainly in the southern mountain area, where construction area was quite low in proportion and pollution was low, too, in degree.

Key words: Xiaoqinghe watershed, nitrogen and phosphorus, spatio-temporal variation, land use, multi-variable statistical analysis