Quantitative Assessment of the Potential of Afforestation for Carbon Dioxide Removal: Evaluating carbon sequestration and biogeophysical impacts in a dynamic global vegetation model

Abstract:

This study presents a new method for representing permanent afforestation in Earth System models. Afforestation has attracted interest as an option to help to slow or reverse the growth of atmospheric carbon dioxide during the next century. However, its potential is poorly constrained with regard to land availability, rates of tree growth and carbon accumulation, and potential side effects.

This study aims to provide quantitative assessment of the carbon removal potential and side effects of 21^st century afforestation using a dynamic global vegetation model, in contrast to the majority of previous estimates which have used bookkeeping methods.

The land surface model JULES was used to simulate needleleaf afforestation on abandoned agricultural land during the 21st century under two future pathways (RCP4.5 and RCP8.5). These results are compared to a control scenario in which natural succession is allowed to act on the same area of land.

This study finds considerable spatial variation in the final carbon sequestration potential of afforestation sites. In addition to dieback and marginal growth in many regions, many sites showed minimal additionality of forest areas compared to natural succession. The most suitable sites were in Eastern Europe, central China and central North America. There was no major difference in the general spatial pattern of suitability between RCP4.5 and RCP8.5 by 2100.

Overall, this study produced a significantly smaller estimate of the CDR potential of permanent afforestation than previous studies have. The additional carbon stored in suitable sites by 2100 was only 19 Pg C (RCP4.5) and 2.1 Pg C (RCP8.5), a mean of 68 tC/ha.

This research also explored the biogeophysical impacts of afforestation on surface energy balance and hydrological cycles. The decrease in albedo caused by afforestation significantly offset the radiative forcing benefits of the carbon removal, although this effect was very sensitive to input assumptions. Flooding results suggested that at realistic levels of afforestation, river water supply and flooding regimes are unlikely to be affected at regional scales. The effect of afforestation on fire risk was mixed, with many suitable sites showing little effect while in others the risk was significantly lower or higher.