Quantifying carbon accumulation and loss in Russian Arctic peatlands

点击翻译按钮获取中文摘要

Peatlands in Russia (and North America) are a major terrestrial carbon store and a natural sink for atmospheric CO2 during the Holocene (Charman et al., 2013, holding around 612 gigatonnes of carbon, which represents a larger stock than the worlds live vegetation. Persistence of this carbon store depends upon waterlogging, which may be threatened by rising atmospheric temperatures, shifting rainfall patterns, anthropogenic disturbance, increases in the cover and height of shrub dominated communities and more frequent fires in high-latitude ecosystems (Young et al., 2017). 'Natural Experiments' recorded in the peat layers make it possible to measure and date carbon accumulation rates during former periods of wetter/drier climatic conditions. Burning can be readily identified by counting the number of macrofossil charcoal fragments preserved in peat profile samples. The burning intensity of former fires can also be estimated using Raman spectroscopy (Muirhead et al., 2017). Changes in the balance between rainfall and evaporation drive changes in peatland surface wetness, which can be reconstructed on decadal, centennial and millennial timescales using the preserved remains of the former vegetation (plant macrofossils). Changes in the preserved microorganism assemblages (testate amoebae) can also be used to provide quantitative reconstructions of former water table depths, which in these ecosystems is dependent upon the atmospheric moisture balance (rainfall minus evaporation). In order to understand the long-term interaction between carbon accumulation rates, fire intensity during burning, mire surface wetness and the types of plants growing in Russian Arctic peatlands, a multiproxy approach is essential. This level of research detail is still relatively rare, so this studentship will offer an exciting opportunity to sample and investigate the long-term carbon sequestration of Russian Arctic peatlands during previous periods of burning disturbance and climate change (on decadal, centennial and millennial timescales). Without robust chronologies the value of these environmental reconstructions will be limited, so a key part of this research will be directed towards Bayesian age/modelling techniques.