Remote Sensing And Predicting Shifts In Biome Distribution And Resilience Using NDVI Data


The dependence of humans on stable, intact ecosystems, whose future characteristics and responses to perturbation are understood, has lead to an increasing appreciation of the need for resilience goals to be incorporated into environmental policy catalogues. The concept of resilience is not new to ecological sciences. However, the great variety of understandings of its meaning and ensuing confusion of the base-line terminology often impede accurate, quantitative assessments of resilience properties. Nevertheless, such knowledge of resilience of natural ecosystems is paramount for the formulation, implementation and evaluation of policy plans which are at the core of environmental Decision Support Systems (DSSs). This thesis focussed therefore on presenting a novel framework for (1) identification of alternative stable states, whose existence is a premise to multiple definitions of resilience, (2) determining what drives their distribution patterns and how, as well as (3) assessing the resilient behaviour exerted by those alternative stable states and the greater systems they belong to. The results of the statistical analyses employed within this study proved valuable in adhering to the three corner stones of the proposed framework and resulted in the delineation of ecosystem types which were easily linked to real-world formations of vegetation compositions. Although one of the approaches used to assess resilience within this study had to be dismissed, two other approaches employed for the same task yielded useful information on resilience properties. Additionally, one of these methods was used to identify regions of ecological uncertainty on which to direct future research and policy-making efforts. Consequently, this study may be a helpful stepping stone for refining and combining already existing methodology which could, in turn, generate important knowledge of ecosystem functioning and resilience.

Feb 24, 2017 12:30 — 13:00
Thesis Defense
Bergen, Norway
Erik Kusch
Erik Kusch
PhD Student

In my research, I focus on statistical approaches to understanding complex processes and patterns in biology using a variety of data banks.