Developing Testate Amoeba-Environmental Vairable Transfer Function In Sphagnum-Dominated Peatlands In Xiaoxing’an Moutains, Northeast China

Testate amoebae are valuable indicators of reconstruction of palaeohydrologicalconditions, which have been mainly developed in Europe and North America, but less used inother area of the world until now. Testate amoebae are unicellular organisms, that substratewetness is the primary control on species abundance and community composition. Theyinhabit in water films of leaves, stems of vascular plants and sphagnums, and fossil tests havebeen recovered in sediments from all these environments. They are abundant in peatlands,have short life cycle, and response to environmental changes sensitively. For those reasons,testate amoebae are suitable for the study of hydrology environment changes. This paper usedmodern testate amoebae species data and environmental factor data to analyze the mainenvironmental factors controlling to the testate amoebae species, and constructs the testateamoebae-environment factor transfer function model.Northeast is major peat distribution area in China, where the peat continuously depositsin situ, and saved well. In order to cover different microhabitats as more as possible, we chosethe sampling points including the top, the middle and low part of the hummocks inTanghongling peatland and Tangbei peatland in Xiaoxing’an Mountains, Northeast China.The study involved with environmental factors including depth to water table, peat surfacemoisture, pH,conductivity, bulk density and ash content. The91samples,55species taxa,and38936tests are analyzed and identified.In the redundancy analysis (RDA), the first and second axes reflect the80.3%ofspecies-environment change information and16.7%of species change information. Themarginal effect of six environment variable shows DWT and moisture of peat are mainenvironmental factors controlling to the testate amoebae species composition in this study.The results of correlation analysis show that the numbers of testate amoebae significantlyassociate with pH and peat surface moisture. And there is significant relationship between thetestate amoebae species and depth to water table, peat surface moisture and pH.WA model and WA-PLS model were used to constructed the transformation functions，and the Leave-one-out validation were used to assess the relative performance of thesemodels.With20%data filtered, the performance of models was improved. WAPLS Comp01model is the best one for DWT, of which the RMSE is7.0670cm, and the R2is0.7366. ButWAPLS Comp03model is the best one for moisture of peat, the RMSE is1.57%, and the R2is0.6382. The DWT and moisture of peat can be reconstructed with7.0670cm and1.57%average error if the profiles samples have the similar species with this research.