Sensitivity of blanket peat vegetation and hydrochemistry to local disturbances

At the ecosystem scale, peatlands can be extremely resilient to perturbations. Yet, they are very sensitive to local disturbances, especially mechanical perturbations (e.g. trampling). The effects of these disturbances on vegetation, and potential effects on hydrochemical conditions along the peat surface, however, are largely unknown. We used three research tracks (paths researchers use to access their study sites) differing in time of abandonment to investigate the impact of local disturbance (trampling) on the vegetation and its short-term (< or = 2 year) recovery in a flagship research blanket peatland. Additionally, we examined the effects of local disturbance on fluvial runoff events and the concentrations of dissolved organic carbon (DOC) and particulate organic carbon (POC) in runoff water. Local disturbance heavily impacted peat vegetation, resulting in large areas of scarred and churned peat. Recovery of vascular plants along abandoned tracks was slow, but a functional Sphagnum layer re-established after just one year. The absence of vegetation elicited an increase in the number of runoff events along the tracks, by which POC runoff from the tracks increased. POC concentrations were highest in the surface water from the recently abandoned track, while they were low in the runoff water from the track abandoned longest and the undisturbed control track. We attribute this to the relatively fast recovery of the Sphagnum vegetation. DOC concentrations did not differ significantly either spatially or temporally in surface runoff or soil solution waters. While at an ecosystem scale local disturbances may be negligible in terms of carbon loss, our data points to the need for further research on the potential long-term effects of local disturbance on the vegetation, and significant effects on local scale carbon fluxes. Moreover, the effects of disturbances could be long-lasting and their role on ecosystem processes should not be underestimated.