Simulations of the last Millennium using a comprehensive climate model: The impact of the external forcing on the climate

Hofer, Dominik (2010). Simulations of the last Millennium using a comprehensive climate model: The impact of the external forcing on the climate (Unpublished). (Dissertation, Universität Bern, Philosophisch–naturwissenschaftliche Fakultät, Physikalisches Institut, Abteilung für Klima– und Umweltphysik)

Text hofer10phd.pdf - Other Restricted to registered users only Available under License BORIS Standard License. Download (27MB)

The Earth’s climate continuously varies under the influence of internal processes and external factors, which are generally referred to as external forcing. To identify the main drivers of the variations is a fundamental challenge in climate sciences. This topic is of great interest, especially in the context of the growing influence of human activities on different parts of the climate system. However, due to the multitude of involved processes and the complex interactions in the climate system it is often difficult to attribute an observed change to one forcing factor.
In this perspective, climate models are a valuable tool that offer new possibilities to analyse the forcing impacts and the involved mechanisms. In contrast to the real world, it is possible to keep the external forcing level constant or to include only a part of the transient external forcing, so that the internal variability and the impact of different forcing factors can be investigated separately. A second approach is to use ensemble simulations, which are forced by the same boundary conditions (external forcing) but started from slightly different initial conditions, also allowing to separate the forced climate signal from the internal variability. Another major application area for climate models are projections of the future climate change based on the expected evolution of the external forcing factors. In all these cases, computer models can contribute to a better understanding of the climate system. However, a model does never represent the real world and each model has its shortcomings (e.g., missing processes and parameterizations). Thus, it is essential that the simulated climate is evaluated against observations, i.e., the past climate is used as a test bed for climate models.
The main purpose of this thesis is to analyse the response of the climate system to external forcing in a comprehensive climate model. Using control simulations, that represent the mean climate state at different times in the last millennium, and transient simulations, which represent an evolving climate for the entire or only parts of the last millennium and the 21st century, different aspects of the forcing impacts are investigated. The focus is set on the pre-industrial period, so that the thesis contributes to a better understanding of the natural, externally forced climate variations versus internal variability.
The first chapter gives an overview of the climate system – focusing on the role of the ocean – and the major processes that perturb the climate. In a second part the use of computer models is motivated, the uncertainties one has to deal with when modelling the past climate are discussed, and some results of other modelling studies are presented.
Chapter 2 introduces the climate model that is used, i.e., the Community Climate System Model version 3 (Collins et al., 2006). In addition, the experimental design (the setup of the simulations and the applied forcing) and the detrending procedure are presented. In the last part of the chapter the climate in the transient simulations and the Atlantic meridional overturning circulation (AMOC) variability in the control simulations are evaluated.
In chapter 3 the impact of the volcanic forcing is investigated. On timescales of months to a few years, volcanic eruptions are a dominant factor for climate variability, and thus provide an ideal test bed to investigate the climate’s response to a perturbation. For the analysis five transient simulations covering the last 500 to 1000 years and 16 short sensitivity experiment are used. The results show that the direct radiative effects – a warming of the lower stratosphere in the layers of the volcanic aerosols and a cooling of the troposphere and at the surface – are correctly represented in extent, but that the warming in the stratosphere is clearly overestimated. Considering the hydrological cycle, a significant reduction of the tropical precipitation is found, in agreement with the results of other models (Robock and Liu, 1994). In contrast to the direct effects, the dynamical response of the atmosphere in the mid and high latitudes is not well resolved in our simulations. Possibly due to the low vertical resolution in the high atmosphere, the stratospheric temperature anomaly is not transfered into a significant change of the North Atlantic Oscillation, and thus no winter warming pattern is found. The importance of the stratospheric resolution for the response to the external forcing – especially on regional scales – is also highlighted in the study by Spangehl et al. (2010) where simulations with different vertical resolutions (including our simulations) are compared (see appendix).
Chapter 4 presents a study examining the variability of the AMOC strength in control and transient simulations (Hofer et al., 2010). For the analysis, three control simulations with the external forcing level appropriate for 1990 AD, 1500 AD, and 1000 AD and the same five transient simulations as in chapter 3 are used. It is shown that even though the transient forcing has no significant direct linear impact on the AMOC strength, the low-frequency variability of the AMOC is increased in the transient simulations. The main cause for this increase is a more frequent occurrence of transitions between two semi-stable circulation states that differ by approximately 10% of the total AMOC strength. These events are triggered by strong salinity anomalies in the deep water formation regions in the North Atlantic and go along with a multitude of changes in the ocean, e.g., regional changes of the sea surface temperature (SST) up to 3 K. The SST forcing then impacts the surface air temperature in Scandinavia leading to a change of approximately 1 K. Thus, it is concluded that changes in the ocean circulation significantly contribute to the Scandinavian surface air temperature variability in the last millennium. This conclusion is consistent with a new tree-ring reconstruction for Scandinavian summer temperatures which illustrates the importance of internal processes for the temperature variability in this region (Büntgen et al., 2010, see appendix).
The thesis is completed with an outlook regarding future research tasks. In the appendix, the two studies by Spangehl et al. (2010) and Büntgen et al. (2010) are presented and some technical aspects of running the model on the current system at the supercomputing center are discussed.

Interest & Impact

Downloads

1 since deposited on 07 Mar 2024
1 in the past 12 months

Citations

5 Citations in Web of Science ®
6 Citations in Scopus

Search

in Google Scholar™

Services

Actions (login required)

Edit item

Actions (login required)

Edit item