Diatom-based water-table reconstruction in Sphagnum peatlands of northeastern China.

Peatlands are important ecosystems for biodiversity conservation, global carbon cycling and water storage. Hydrological changes due to climate variability have accelerated the degradation of global and regional ecosystem services of peatlands. Diatoms are important producers and bioindicators in wetlands, but comprehensive diatom-based inference models for palaeoenvironmental reconstruction in peatlands are scarce. To explore the use of diatoms for investigating peatland hydrological change, this study established a training set consisting of diatom composition and twelve environmental factors from 105 surface samples collected from five Sphagnum peatlands in northeastern China. Diatom communities were dominated by Eunotia species. Ordination analyses showed that depth to the water table (DWT) was the most important factor influencing diatom distribution, independently accounting for 4.99% of total variance in diatom data. Accordingly, a diatom-based DWT transfer function was developed and thoroughly tested. The results revealed that the best-performing model was based on weighted averaging with inverse deshrinking (R2 = 0.66, RMSEP = 8.8 cm with leave-one-out cross validation). Quantitative reconstruction of DWT on a short peat core collected from the Aershan Peatland (Inner Mongolia) recorded climate-mediated hydrological changes over the last two centuries. This study presents the first diatom-water table transfer function in Sphagnum peatlands, and highlights the potential of diatoms as a powerful tool to assess the magnitude of past hydrological changes in peatlands of northeastern China, as well as similar peaty environments worldwide.