Uncertainties In The Occurrence Time Of Warming Control Target Arising From Model Internal Variability

To address the risks caused by global warming,international governments have established global warming control targets of 1.5? and 2?.Projection,based on Earth system models,is an important method to research the occurrence time of the warming control targets(or occurrence time for short).However,there are great uncertainties in simulating the occurrence time.Model internal variability has been recognized as one of the most important sources of uncertainties in climate simulation results.However,its impact on simulating the occurrence time has not yet been well studied.To study the uncertainties of occurrence time simulation caused by model internal variability,this thesis employed the corresponding outputs of several models selected from the Coupled Model Intercomparison Project Phase 5(CMIP5),conducted the research work and made the following observations or conclusions:(1)This thesis first evaluated the spatial distribution of the occurrence time simulated by the selected models,to demonstrate that the projections of these models for the warming control targets are consistent with other research results.The average results of the multi-model ensemble(MME)show that the global average surface temperature(or global temperature for short)will reach 1.5? warming in 2030,2026,2035 and 2023 respectively under the four different emission scenarios(RCP2.6,RCP4.5,RCP6.0 and RCP8.5).Under RCP2.6,the global temperature rise will not reach 2? until the end of the 21st century,while it is predicted to exceed 2? in 2048,2063 and 2037 under RCP4.5,RCP6.0 and RCP8.5 respectively.The occurrence time shows a clear sea-land distribution,i.e.,the warming control targets on land are reached much earlier than those on the oceans at the same latitude.(2)Regarding each selected model,this thesis analyzed the impact(including its spatial distribution)of the uncertainties arising from model internal variability on the occurrence time simulation(the following occurrence time only refers to the time when the warming control targets are stably reached),as well as sensitivity of the uncertainties to different emission scenarios.The results show that,regarding the occurrence time simulation,the uncertainties arising from model internal variability can introduce an equivalent impact compared to the external forcing of different emission scenarios,and the impact of these uncertainties is more significant at regional scale than for the global average and has obvious global spatial distribution characteristics.Maximum uncertainties can be observed at the ocean north of Eurasia,the area around the Bering Straits,northeastern North America and the ocean between America and Greenland,and the coast of the Antarctic.The uncertainties arising from model internal variability are closely related to the intensity of emissions,generally there are more significant uncertainties at lower intensity of emissions.(3)This thesis utilized MME to further study the impact(including its spatial distribution)of the uncertainties arising from model internal variability.Similarly,the impact of these uncertainties is more significant at regional scale than at global scale,and generally more significant under lower intensity of emissions.Compared with the uncertainties among the samples of a single model caused by internal variability,the uncertainties of MME results are reduced,which indicates that MME can reduce the uncertainties caused by internal variability.