Responses of Terrestrial Ecosystem to Climate Change: Results from Approach of Conditional Nonlinear Optimal Perturbation of Parameters

Abstract

Climate change is an important factor influencing the structure and carbon cycle of a terrestrial ecosystem. The estimation of a terrestrial ecosystem may be uncertain due to the unknown change about the climate in the future. In this chapter, we review the recent progress of the authors on the variations in a terrestrial ecosystem due to climate change using conditional nonlinear optimal perturbation of parameters (CNOP-P) approach. The stability of a grassland ecosystem to climate change is discussed first. A five-variable theoretical model of a grassland ecosystem is employed. The type of climate change described by the CNOP-P approach is called the CNOP-P or nonlinear type of climate change. Two linear types of climate change are used to compare stability of a grassland ecosystem. The results show that when it is affected by the CNOP-P-type climate change, a grassland ecosystem abruptly becomes a desert ecosystem. However, the two linear types of climate change do not lead to this abrupt change within a specific amplitude range. Similar results are found when a desert ecosystem is used as the reference state. Second, the maximum impact of climate change on the soil carbon in China is explored using the CNOP-P approach with the Lund-Potsdam-Jena (LPJ) model. The variations in the amount of soil carbon due to CNOP-P-type temperature or precipitation perturbation are compared with those caused by a linear perturbation of temperature or precipitation. The CNOP-P-type temperature or precipitation perturbation could lead to the variation of variability of temperature or precipitation. However, the linear type of temperature or precipitation perturbation fails. The numerical results demonstrate that, in southern China, the amount of soil carbon augments as a result of the CNOP-P-type temperature perturbation, and the variation in the amount of soil carbon due to the linear temperature perturbation is minor. The pool of fast decomposing soil carbon is an important factor to lead to the difference. The above numerical results imply that the CNOP-P approach is an available tool to explore the response of terrestrial ecosystem to climate change.