Tschumi, Elisabeth Andrea (2022). The effects of differing drought-heat signatures on terrestrial carbon dynamics and vegetation composition using dynamic vegetation modelling. (Thesis). Universität Bern, Bern

Preview

Text 22tschumi_ea.pdf - Thesis Available under License Creative Commons: Attribution-Noncommercial-No Derivative Works (CC-BY-NC-ND 4.0). Download (55MB) | Preview

Abstract

The global carbon cycle is essential for many aspects of the climate system and life in general. Within this carbon cycle, the natural terrestrial carbon sink plays an important role removing CO2 from the atmosphere. Many factors influence this terrestrial cycle, particularly climate variability, both natural and human-made. Especially in the face of global warming, the evolution of the natural carbon sink is quite uncertain. In particular, the effects of extreme weather and climate events, which are expected to become more frequent and more severe in the future, on the carbon cycle are hard to quantify, since the extremes themselves are also associated with uncertainties. To contribute to the ever-growing scientific field of carbon cycle and extreme event research, this thesis is concerned with analysing the effects of differing drought-heat signatures on terrestrial carbon dynamics using dynamic vegetation modelling. We built six hypothetical 100-year long climate scenarios which differ in their occurrence frequency of hot and dry extremes. They are based on a large ensemble simulation generated by the climate model EC-Earth. This data has several advantages. Firstly, it represents present-day climate without any trends. Secondly, it is available on a global grid, and thirdly, it offers a very long time series, which is needed to study extreme events in order to have a large enough sample size. The scenarios only differ in their extreme occurrence, but are similar in their global means. However, the data does present some regional biases which have a potentially large impact on modelled impacts. The scenarios are described and characterized in Chapter 2. The six hypothetical scenarios were used to run the dynamic global vegetation model LPX-Bern v1.4, as well as five additional dynamic global vegetation models. The models were run using constant CO2 concentrations and not allowing any land-atmosphere feedbacks which might change our initially sampled scenarios. They also only consider natural vegetation, meaning no crops or other land uses. The LPX-Bern results, described in detail in Chapter 3, show clear differences between scenarios as well as between climate zones. While trees thrive under climate scenarios with few extremes or only hot extremes, especially in higher latitudes, they show a clear reduction in coverage for dry extremes and especially compound hot and dry extremes. The relatively large increase in tree coverage in high latitudes under more hot extremes is associated with an increased growing season length in these regions which are generally energy-limited. Grasses tend to compensate the changes in tree coverage to some extent. Changes in tree coverage are also associated with changes in plant productivity and carbon stored in vegetation. Most of these results are shown as global means with regional differences being potentially large. Chapter 4 discusses the comparison of different vegetation models all run with the same input scenarios. The carbon variables are comparable between the models in the global mean, but we see quite large differences in vegetation coverage, most likely due to biases in the input data. Regionally, these differences may be even larger. There is still overall agreement between the models that a compound hot and dry climate leads to a reduction in tree coverage with an associated reduction in carbon stored in vegetation. The scenario with frequent dry extremes suggests similar results, with slightly less agreement between the models. The effects of a climate with hot extremes and those of a climate with no compound extremes are generally small. Large differences can be seen in a climate with no extreme events at all, where some models simulate an increase in vegetation and others a decrease. However, it is clear that compound hot and dry events are associated with a reduction of carbon stored on land. Our results suggest a possible reduction in the natural land carbon sink under future climate change. The coupling of temperature and precipitation can vary substantially between models and biases can exist in the input data. Therefore, results can differ when studying compound events. This thesis contributes to the understanding of feedbacks and processes concerning variable interaction, which is crucial to improve models. The field of compound event research is still emerging and ever-growing and there is still a lot to investigate when it comes to the effects of extremes on the terrestrial carbon cycle. Future work could focus on other types of compound events, such as temporally compounding or preconditioning, or different impact models could be used, for example, crop or fire models.

Item Type:	Thesis
Dissertation Type:	Cumulative
Date of Defense:	27 October 2022
Subjects:	500 Science > 530 Physics
Institute / Center:	08 Faculty of Science > Physics Institute > Climate and Environmental Physics
Depositing User:	Sarah Stalder
Date Deposited:	26 May 2023 15:17
Last Modified:	27 Oct 2023 22:25
URI:	https://boristheses.unibe.ch/id/eprint/4320

Actions (login required)

View Item