This thesis aims to advance ecological modeling by modeling and analyses of the influence of configuration settings of ecological models. It aims to evaluate different models and configuration settings of such models in species distribution models (SDM) to provide improved SDM. The thesis provides an extensive review of SDM, modeling methods, and evaluation methods to understand the process of modeling species distribution and the importance of SDM in environment protection and strategy decisions. This thesis considers an important issue the setting of SDMs' configurations for ecological models. This thesis conducted an extensive literature search to find the typical default values and the range of values for configuration settings of several known algorithms for constructing species distribution models (SDM). This thesis sets out to comprehensively test the sensitivity of eight widely used models with a range of parameter values for them. This research study analyses the models' performance to analyze parameter settings and configurations of these models. This research study lists hypotheses for configuration sensitivity and evaluation performance in modeling based on a literature review of these models. Koala and Thorny Devil data is used in this research study to simulate their distribution models, which output visual maps and four evaluation methods' scores. Findings demonstrate that the simulated models have different sensitivities based on their parameter settings and configurations. This thesis covers the gap between configuration and model output analysis references and discusses the contribution of parameter setting and configuration in modeling applied to Koala and Thorny Devil distribution models. This thesis examines models that are employed in a cloud-based system ecological modeling, and it examines the modeling and accuracy of eight ecological models.
Modeling, configuration and evaluation of ecological systems
ZHAO, Y. (Author). 2024
Student thesis: Doctoral Thesis