2019 & 2020
Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, 2019
Veröffentlicht auf dem Publikationsserver der RWTH Aachen University 2020
Genehmigende Fakultät
Fak03
Hauptberichter/Gutachter
;
Tag der mündlichen Prüfung/Habilitation
2019-12-19
Online
DOI: 10.18154/RWTH-2020-02131
URL: https://publications.rwth-aachen.de/record/782867/files/782867.pdf
Einrichtungen
Inhaltliche Beschreibung (Schlagwörter)
agent based modeling (frei) ; coasts (frei) ; flood frequency (frei) ; flood memory (frei) ; flood risk management (frei) ; human-flood interaction (frei) ; risk perception (frei)
Thematische Einordnung (Klassifikation)
DDC: 624
Kurzfassung
One of the major challenges in current flood management studies is to include human-flood interaction in their modeling approach in order to investigate how individuals respond to flooding and how their involvement results in a more effective flood risk management (FRM). Furthermore, humans are heterogeneous in their socio-economic attributes as well as their risk awareness which result in feedbacks between humans and the environment. Therefore, individual adaptation responses, knowledge exchange, flood memory, and flood risk perception shape a new mode of interaction and temporal changes in exposure and vulnerability. All these factors cause nonlinear behaviors in the subsystems exposing the whole system to major changes beyond the scope of traditional FRMs. Moreover, there are limitations to the availability of information as well as to the processing capacities of decision makers in reality resulting in non-optimizing behaviors and bounded-rationality. Therefore, formalizing the individual adaptive behavior on the basis of rational behavior and economic optimizing as well as perfect information has its limitations. In addition, FRM studies assume static conditions in which humans and their surrounding environment are inactive and their vulnerability is constant. Under such assumptions, time dependent features such as interactions, adaptations, and technology innovation cannot be incorporated in current models and there is lack of modeling approaches to include social aspects of human behavior in FRM. To fill these knowledge gaps, interdisciplinary approaches, which allow formulating adaptive individual decision-making under uncertainty, are in demand. More specifically, there is a need to a technique that allows us to model social processes and complexities of human behaviors from the bottom-up approach and in combination with engineering practices. Agent Based Modeling is such an approach that relies on a more realistic set of assumptions. This study employs Agent Based Modeling within the framework of FRM, particularly for the agricultural sector, and presents an experimental platform to simulate farmers’ adaptive strategies in coastal regions. An Agent Based Model (ABM) of farmers’ behaviors is developed including three parts: farmers’ decision-making module, flood risk analysis module as well as risk perception module. It is then linked to the hydrological module and hydrodynamic module designed in the study for this purpose. The coupled model, which is called the “Agent Based Model for farmer-flood interaction (ABMFaFo)”, introduces the interactions among farmers and includes individual risk judgment in their decision-making. Additionally, farmers’ decisions are formulated in the model through bounded-rationality theory to consider limited information availability as well as limited information processing capacities of people. Pellworm Island in north of Germany is chosen as the virtual study area and the established ABMFaFo is applied to 37 semi-hypothetical farmers living on the Island. The model is run using a series of in silico experiments to investigate farmers’ decision-making in flood-prone areas in response to coastal flooding. More specifically, the effect of flood frequency, risk perception, social interaction, past experience, and flood memory are examined and discussed. In addition, the interdependencies between vulnerability of the agricultural sector at farm-level and regional-level are explored using several macro-metrics. Every experiment is run for the time horizon 2005-2016, including one year of warm up period for the model.
OpenAccess:
PDF
(additional files)
Dokumenttyp
Dissertation / PhD Thesis
Format
online
Sprache
English
Externe Identnummern
HBZ: HT020378268
Interne Identnummern
RWTH-2020-02131
Datensatz-ID: 782867
Beteiligte Länder
Germany
The record appears in these collections: |