Abstract:The Baiyangdian Lake is the largest freshwater lake and freshwater wetland in Hebei Province, North China. It is 162 km south of Beijing and covering an area of 366.0 km2 during high water periods. It consists of about 143 separated small water bodies connected by 3700 web-like waterways. The lake not only provides water for local use and different habitats for aquatic organisms, but also plays an important role on both regional ecological modulations and biodiversity conservation. Since the middle 1960s, industrial waste from the upper reaches and domestic sewage from nearby villages have been the main sources of water pollution. The lake thus became eutrophic due to an increase in nutrients. Diatoms, widely distributed in water environments, are excellent ecological indicators for their sensitivity to a wide range of environmental variables, and their community structure can quickly respond to the changing physical, chemical and biological conditions in the water. In order to evaluate the water quality, we estimated the diatom populations from different micro-habitats, analyzed the relationship between diatom assemblages and water environments, and attempted to reconstruct the total phosphorus (TP) and pH by using diatom-environment transfer functions provided by the European Diatom Database (EDDI). Water samples were collected from 9 sites of the lake including different micro-habitats in July and August 2008. The water pH, water temperature and water depth were measured during sampling. After centrifugal settling, water samples were cleaned by H2O2 and HCl digestion. The diatoms were identified and counted under and scanning electron microscope (SEM) the light microscope. The relationships between diatoms and environmental variables were examined using canonical correspondence analysis (CCA). The environmental reconstructions of water pH and TP were performed using an online program provided by EDDI. The diatom assemblages recorded from the Baiyangdian Lake consists of 85 species, belonging to 31 genera of 22 families. Cyclotella, Nitzschia, Caloneis and Navicula are the predominant taxa in the lake. The different diatom assemblages found in the different water environments vary in their composition. The eutrophic scenic area was dominated by Cyclotella meneghiniana, contaminated inlet channel characterized by Sellaphora seminulum and Navicula veneta, village area dominated by Nitzschia-Navicula taxa, waterway area characterized by Aulacoseira granulata, in area lotus pond Rhopalodia gibba is dominate, and fish pond characterized by Nitzschia taxa. The result of CCA indicated that the TP and pH are the main environmental factors controlling the distribution of diatoms. The estimated value of TP is 0.13-0.25 mg/L and pH is 6.92-7.23, respectively, which are very close to the actual measurement data from the Baiyangdian Lake. The result indicate that the diatom compositions are extremely sensitive to changes of water quality, and can be used to reflect different water habitats and monitor water quality very well. The distribution of diatom species is closely related to several environmental variables, primarily TP and water pH. The estimated value from EDDI also proved the feasibility of environmental reconstructions based on diatom-environment transfer functions.