63 Spatial and temporal variations in carbon balances of forest ecosystems in Hokkaido, Japan

Monday, 20 June 2016
Ryuichi Hirata, National Institute for Environmental Studies, Tsukuba, Japan; and A. Ito and N. Saigusa

We evaluated spatial and temporal variations in carbon balance of forest ecosystems using a process-based ecosystem model taking a forest age map into account. Carbon balances in forest ecosystems change by disturbances and recovery-processes, because forest ecosystem structure and functions are abruptly changed. Forests whose age are less than 60 years old occupy more than 70% in Japan because forest stands were directly replaced by disturbances such as thinning, harvest, plantation, fire, typhoon and insects. However, few studies have attempted to quantify how much disturbances affect spatial and temporal variations in carbon balances. In this study, we focus on how disturbance and following regrowth affect spatial and temporal variations in carbon balance. We used Vegetation Integrative SImulator for Trace Gases (VISIT) model in order to simulate carbon balance in Hokkaido, which is northern island of Japan. Simulations of the carbon balance were conducted for the period 1948–2010 following 1000 years of spinning-up at a spatial resolution of 1km × 1km. We conducted two cases of simulations concerning the carbon balance: one taking account of spatial distribution of disturbance-induced forest age derived from forest inventory data and another ignoring the disturbance impact (i.e., no disturbance, even-aged case). In the case of non-disturbance, NEP was gradually and spatially changed ranging from 0 to 1 t C/ha/y depending on meteorological conditions such as temperature or solar radiation. On the other hand, in the case of disturbance, large NEP ranging from 3 to 5 t C/ha/y were distributed patchwise like hotspots, because forest age of these spots ranging from 20 to 100 years old and then younger than those of the non-disturbance case. In the 1970s, wood harvest and tree planting were intensively conducted in Hokkaido. In the disturbance case during this period, there were many hotspots which show negative NEP. Next, we evaluated the impact of disturbance on NEP by the difference between the disturbance and non-disturbance scenarios. Until 1970, impact of disturbance was less than 0.01 t C/ha/y. After 1970, impact of disturbance became large and reached about 0.5 t C/ha/y. It means that regional NEP of the disturbance case increased 2-5 times of the non-disturbance case. Our results showed the importance of considering forest age when simulating carbon balance. Carbon balance maps taking forest age into account are useful for carbon managements and prediction of ecosystem feedbacks to climate change.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner