Evaluating the carbon sequestration potential of volcanic soils in southern Iceland after birch afforestation

Afforestation is a strategy to sequester atmospheric carbon in the terrestrial system and to enhance ecosystem services. Iceland's large areas of formerly vegetated and now degraded ecosystems therefore have a high potential to act as carbon sinks. Consequently, the ecological restoration of these landscape systems is part of climate mitigation programmes supported by the Icelandic government. The aim of this study was to explore the change in the soil organic carbon (SOC) pools and to estimate the SOC sequestration potential during the re-establishment of birch forest on severely degraded land. Differently aged afforested mountain birch sites (15, 20, 25 and 50 years) were compared to sites of severely degraded land, naturally growing remnants of mountain birch woodland and grasslands which were re-vegetated using fertilizer and grass seeds 50 years ago. The soil was sampled to estimate the SOC stocks and for physical fractionation to characterize the quality of the SOC. The results of our study show that the severely degraded soils can potentially sequester an additional 20 t C ha^-1 (0-30 cm) to reach the SOC stock of naturally growing birch woodlands. After 50 years of birch growth, the SOC stock is significantly lower than that of a naturally growing birch woodland, suggesting that afforested stands could sequester additional SOC beyond 50 years of growth. The SOC fractionation revealed that at all the tested sites most of the carbon was stored in the <63 µm fraction. However, after 50 years of birch growth on severely degraded soils the particulate organic matter (POM) fraction was significantly enriched most (+12 t POM-C ha^-1) in the top 30 cm. The study also found a doubling of the dissolved organic carbon (DOC) concentration after 50 years of birch growth. Therefore and due to the absence of any increase in the tested mineral-associated SOC fractions, we assume that the afforestation process evokes a carbon deposition in the labile SOC pools. Consequently, parts of this plant-derived, labile SOC may be partly released into the atmosphere during the process of stabilization with the mineral soil phases in the future. Our results are limited in their scope since the selected sites do not fully reflect the heterogeneity of landscape evolution and the range of soil degradation conditions. As an alternative, we suggest using repeated plot measurements instead of space-for-time substitution approaches for testing C changes in severely degraded volcanic soils. Our findings clearly show that detailed measurements on the SOC quality are needed to estimate the SOC sequestration potential of restoration activities on severely degraded volcanic soils, rather than only measuring SOC concentration and SOC stocks.