Carbon and water balance of an afforested shallow drained peatland in Iceland

Drainage of peatlands increases the depth of the oxic peat layer and can turn them into a carbon (C) source to the atmosphere. Afforestation of drained peatlands could help to reverse this process since the trees may enhance C sequestration. We followed the C and water dynamics of an afforested drained peatland in S-Iceland during a 2 year period, during which the Black Cottonwood (Populus balsamifera ssp. trichocarpa) plantation was 23–25 year old. Net ecosystem exchange (NEE) of carbon dioxide (CO2) was measured with the eddy covariance method and C pools of trees and ground vegetation were measured using the stock change method. Lateral losses of dissolved and particulated organic C (DOC, POC) were estimated from weekly water-runoff samples. Unexpectedly, the afforested drained peatland was a strong sink of carbon during the two years, with an average NEE value of 714 g C m−2 yr−1. Only 0.5% of the total NEE was lost through lateral DOC and POC transport, leaving 710 g C m−2 yr−1 as the total net ecosystem production (NEP). Ca. 91% of the observed NEP could be explained by the annual biomass increment of the Black Cottonwood trees and 1.3% by the ground vegetation. This means that the remaining 7.5% of the total NEP most likely accumulated in peat soil and litter, contributing to the soil C stocks. The dormant-season CO2 emissions were unexpectedly low, which was explained by a high groundwater level at this drained site outside the ca. 5 months of the active growing season. On average, 66% of the annual measured precipitation was estimated to have evaporated back to the atmosphere. This left 416 mm for potential runoff, which was somewhat lower value than the measured runoff (662 mm). These results indicate that during the age span of ca. 20–25 years, afforestation was a valid method to reverse the expected negative C-balance of this drained grassland pasture in Iceland. Although the site is currently a soil C sink, simulation studies with process models are needed to test whether such sites could remain C sinks when managed for forestry over several tree-stand rotations.