Do tree species influence soil carbon stocks in temperate and boreal forests?

Information on tree species effects on soil organic carbon (SOC) stocks is scattered and there have been few attempts to synthesize results for forest floor and mineral soil C pools. We reviewed and synthesized current knowledge of tree species effects on SOC stocks in temperate and boreal forests based on common garden, retrospective paired stand and retrospective single-tree studies. There was evidence of consistent tree species effects on SOC stocks. Effects were clearest for forest floor C stocks (23 of 24 studies) with consistent differences for tree genera common to European and North American temperate and boreal forests. Support for generalization of tree species effects on mineral soil C stocks was more limited, but significant effects were found in 13 of 22 studies that measured mineral soil C. Proportional differences in forest floor and mineral soil C stocks among tree species suggested that C stocks can be increased by 200-500% in forest floors and by 40-50% in top mineral soil by tree species change. However, these proportional differences within forest floors and mineral soils are not always additive: the C distribution between forest floor and mineral soil rather than total C stock tends to differ among tree species within temperate forests. This suggests that some species may be better engineers for sequestration of C in stable form in the mineral soil, but it is unclear whether the key mechanism is root litter input or macrofauna activity. Tree species effects on SOC in targeted experiments were most consistent with results from large-scale inventories for forest floor C stocks whereas mineral soil C stocks appeared to be stronger influenced by soil type or climate than by tree species at regional or national scales. Although little studied, there are indications that higher tree species diversity could lead to higher SOC stocks but the role of tree species diversity per se vs. species identity effects needs to be disentangled in rigorous experimental designs. For targeted use of tree species to sequester soil C we must identify the processes related to C input and output, particularly belowground, that control SOC stock differences. We should also study forms and stability of C along with bulk C stocks to assess whether certain broadleaves store C in more stable form. Joint cooperation is needed to support syntheses and process-oriented work on tree species and SOC, e.g. through an international network of common garden experiments. (C) 2013 Elsevier B.V. All rights reserved.