Stochastic information in the assessment of climate change

• Main Author: Kleinen, Thomas Christopher
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: 2005
• Subjects: Anthropogene Klimaänderung; Stochastische Differentialgleichung; Überflutung; Thermohaline Zi; ddc:530; Institut für Physik und Astronomie; North Atlantic
• Online Access: https://publishup.uni-potsdam.de/opus4-ubp/frontdoor/index/index/docId/476; https://nbn-resolving.org/urn:nbn:de:kobv:517-opus-5382; https://publishup.uni-potsdam.de/opus4-ubp/files/476/kleinen.pdf

Stochastic information, to be understood as "information gained by the application of stochastic methods", is proposed as a tool in the assessment of changes in climate. This thesis aims at demonstrating that stochastic information can improve the consideration and reduction of uncertainty in the assessment of changes in climate. The thesis consists of three parts. In part one, an indicator is developed that allows the determination of the proximity to a critical threshold. In part two, the tolerable windows approach (TWA) is extended to a probabilistic TWA. In part three, an integrated assessment of changes in flooding probability due to climate change is conducted within the TWA. The thermohaline circulation (THC) is a circulation system in the North Atlantic, where the circulation may break down in a saddle-node bifurcation under the influence of climate change. Due to uncertainty in ocean models, it is currently very difficult to determine the distance of the THC to the bifurcation point. We propose a new indicator to determine the system's proximity to the bifurcation point by considering the THC as a stochastic system and using the information contained in the fluctuations of the circulation around the mean state. As the system is moved closer to the bifurcation point, the power spectrum of the overturning becomes "redder", i.e. more energy is contained in the low frequencies. Since the spectral changes are a generic property of the saddle-node bifurcation, the method is not limited to the THC, but it could also be applicable to other systems, e.g. transitions in ecosystems. In part two, a probabilistic extension to the tolerable windows approach (TWA) is developed. In the TWA, the aim is to determine the complete set of emission strategies that are compatible with so-called guardrails. Guardrails are limits to impacts of climate change or to climate change itself. Therefore, the TWA determines the "maneuvering space" humanity has, if certain impacts of climate change are to be avoided. Due to uncertainty ...