Impact of grassland afforestation on soil carbon in New Zealand: a review of paired-site studies

Abstract

Afforestation of grassland provides an opportunity for partial mitigation of increasing carbon dioxide (CO2) levels in the atmosphere through carbon (C) fixation in biomass, but little is known of the impact of afforestation on soil C. To determine the impact of afforestation on soil C levels, data from published papers, theses, and unpublished studies of paired adjoining grassland and afforested sites in New Zealand were assembled and compared. The forest sites within each pair were planted into grassland rather than some other land use, and were a minimum of 10 years old. A total of 28 paired sites had information on both mineral soil C concentration and bulk density, 17 with the forest part of the pair aged 10-20 years, and 11 with the forest aged more than 20 years. Forest floor C information was available for 9 sites. Only 3 of the forest stands had been harvested. Results indicated that afforestation of grassland soils reduces upper mineral soil (mainly 0-10 cm layer) C levels by about 4.5 t/ha or 9.5% in the short-term; however, beyond forest age 20 years there was no difference mineral in soil C between the two systems. Soil bulk density in the 0-10 cm layer was unaffected by afforestation during the first rotation. This allowed comparison of a larger number of sites (27 with forest aged 10-20 years, 18 with forest aged >20 years) that had C concentration data only. Analysis of this larger data set confirmed results obtained from the C mass data alone. Effects of afforestation on mineral soil C were most pronounced in the upper soil and declined rapidly with depth to the extent that at most sites there was no influence of afforestation on soil C below the 0-10 cm layer. At some sites, however, the impact of afforestation proceeded to greater depths, and further study is required to determine reasons for differences between sites in this regard. The impacts of afforestation on soil C observed from the paired-site studies agree well with those of recent analyses for the upper soil layer using New Zealand national soils databases. At greater depths, however, analyses using the databases appear to greatly overestimate the influence of afforestation on soil C. The available data indicate that C accumulating in the forest floor is likely to exceed any short-term reduction in mineral soil C arising from grassland afforestation.