Modeling the carbon dynamics of the dryland ecosystems in Xinjiang, China from 1981 to 2007—The spatiotemporal patterns and climate controls

Modeling the carbon dynamics of the dryland ecosystems in Xinjiang, China from 1981 to 2007—The spatiotemporal patterns and climate controls

Ecological Modelling

Abstract Dryland ecosystems are particularly sensitive to environmental stresses. Despite their importance to the global carbon (C) cycle, responses of the Central Asian dryland to the rapid climate change in recent decades are still unclear. Using AEM, a newly developed, spatially explicit process model for dryland ecosystems, a case study was conducted in Xinjiang, a 1.66 M km2 dryland in eastern Central Asia. The goal was to assess the impacts of environmental changes (climate change and elevated CO2) on the regional C dynamics from 1981 to 2007. The results indicated that over the last three decades, Xinjiang acted as a C sink of 138 Tg (1 T = 1012), 78.5% of which was contributed by increased vegetation C. The C dynamic overall was dominated by the CO2 fertilization effect, which resulted in 124 Tg C sequestration from 1981 to 2007. Temporal pattern of C dynamic was controlled by the climate change, which resulted in 10 Tg C sequestration. The model simulation also indicated that the ecosystem's response to the combined effect by CO2 and climate change together was nonlinear. Among climate factors, temperature change resulted in 12 Tg C loss, while precipitation change resulted in 24 Tg C sequestration. The rising temperature stimulated heterotrophic respiration, causing 17 Tg C loss from the soil. The climate change had complex effects on the regional C dynamics. It caused a 13 Tg C loss in southern Xinjiang, while resulting in a 23 Tg C sequestration in the north. The region acted as a C source in the 1980s, mainly due to the drought from 1983 to 1986, but has turned into a C sink since 1990. Unlike other dryland plant types, the irrigated crop and the phreatophytic shrub were barely affected by changes in precipitation. The ecosystem complexity in the dryland highlights the importance of addressing environmental heterogeneity with high resolution datasets and considering the characteristic ecophysiology of dryland plants with process-based modeling in climate change studies.