Large Herbivores Affecting Permafrost – Impacts of Grazing on Permafrost Soil Carbon Storage in Northeastern Siberia
Abstract. The risk of carbon emissions from permafrost ground is linked to ground temperature and thus in particular to thermal insulation by vegetation and organic soil layers in summer and snow cover in winter. This ground insulation is strongly influenced by the presence of large herbivorous animals browsing for food. In this study, we examine the potential impact of large herbivore presence on the ground carbon storage in thermokarst landscapes of northeastern Siberia. Our aim is to understand how intensive animal grazing may affect permafrost thaw and hence organic matter decomposition, leading to different ground carbon storage, which is significant in the active layer. Therefore, we analysed sites with differing large herbivore grazing intensity in the Pleistocene Park near Chersky and measured maximum thaw depth, total organic carbon content and decomposition state by δ13C isotope analysis. In addition, we determined sediment grain size composition as well as ice and water content. We found the thaw depth to be shallower and carbon storage to be higher in intensively grazed areas compared to extensively and non-grazed sites in the same thermokarst basin. The intensive grazing presumably leads to a more stable thermal ground regime and thus to increased carbon storage in the thermokarst deposits and active layer. However, the high carbon content found within the upper 20 cm on intensively grazed sites could also indicate higher carbon input rather than reduced decomposition, which requires further studies. We connect our findings to more animal trampling in winter, which causes snow disturbance and cooler winter ground temperatures during the average annual 225 days below freezing. This winter cooling overcompensates ground warming due to the lower insulation associated with shorter heavily grazed vegetation during the average annual 140 thaw days. We conclude that intensive grazing influences the carbon storage capacities of permafrost areas and hence might be an actively manageable instrument to reduce net carbon emission from these sites.
Torben Windirsch et al.
Torben Windirsch et al.
Large herbivores affecting terrestrial permafrost in northeastern Siberia: biogeochemical and sediment characteristics under different grazing intensities https://doi.pangaea.de/10.1594/PANGAEA.933446
Torben Windirsch et al.
Viewed (geographical distribution)
This manuscript presents observations of soil cores from different landscape units and disturbance histories, and aims to answer an interesting and relevant question, “does grazing by large mammals impact permafrost carbon storage?” Unfortunately, the experimental design is fundamentally flawed, making any conclusions about the impact of herbivory on soil carbon storage impossible.
The main issue is lack of replication – the study relies on a single soil core for each combination of environment (drained lake basin or upland) and grazing (intensive or no grazing), which is insufficient given the variability of soil composition and the presence of confounding variables. We know that soil core properties are highly variable in permafrost environments due to cryoturbation, so any variation from one site to another could be due to natural spatial variability or the variable of interest, herbivory. Without replication within sites to account for spatial variability of permafrost soils there is no way to discern between those two possibilities. Additionally, soil moisture is a confounding variable that cannot be accounted for without additional samples in a wider range of environmental conditions. The authors showed that soil organic carbon varied with water/ice content and mentioned that the grazed sites in the drained lake basin flooded seasonally, while none of the other sites flood regularly. This means that patterns in soil organic carbon may be due primarily to variation in soil moisture rather than herbivory, because soil moisture and herbivory covary. Another potential confounding variable is the site history. The authors mentioned that the non-grazed drained lake site was cleared of forest a few years prior to the study while none of the other sites underwent the same treatment.
While the underlying soil core data are sound and could be used to describe some of the variability of the site, the flawed study design makes it impossible to disentangle the effects of spatial heterogeneity, soil moisture regime, site history, and herbivory. Therefore, I suggest that this manuscript be rejected and the authors reconsider the scope of question that can be answered with these data for a new submission.