Endurance of larch forest ecosystems in eastern Siberia under warming trends

摘要

The larch (Larix spp.) forest in eastern Siberia is the world's largest coniferous forest. Its persistence is considered to depend on near‐surface permafrost, and thus, forecast warming over the 21st century and consequent degradation of near‐surface permafrost is expected to affect the larch forest in Siberia. However, predictions of these effects vary greatly, and many uncertainties remain about land – atmosphere interactions within the ecosystem. We developed an integrated land surface model to analyze how the Siberian larch forest will react to current warming trends. This model analyzed interactions between vegetation dynamics and thermo‐hydrology, although it does not consider many processes those are considered to affect productivity response to a changing climate (e.g., nitrogen limitation, waterlogged soil, heat stress, and change in species composition). The model showed that, under climatic conditions predicted under gradual and rapid warming, the annual net primary production of larch increased about 2 and 3 times, respectively, by the end of the 21st century compared with that in the previous century. Soil water content during the larch‐growing season showed no obvious trend, even when surface permafrost was allowed to decay and result in subsurface runoff. A sensitivity test showed that the forecast temperature and precipitation trends extended larch leafing days and reduced water shortages during the growing season, thereby increasing productivity. The integrated model also satisfactorily reconstructed latitudinal gradients in permafrost presence, soil moisture, tree leaf area index, and biomass over the entire larch‐dominated area in eastern Siberia. Projected changes to ecosystem hydrology and larch productivity at this geographical scale were consistent with those from site‐level simulation. This study reduces the uncertainty surrounding the impact of current climate trends on this globally important carbon reservoir, and it demonstrates the need to consider complex ecological processes to make accurate predictions.