Shaping an ecosystem for species distribution modeling

Species Distribution Modeling applications depend on large amounts of data containing important hidden insights and patterns that can be discovered using statistic and/or genetic algorithms to support predictive analyses for biodiversity conservation. Unfortunately the models themselves are typically lost after publishing the discovery and usually a one time projection of these hidden patterns on a geographic map in a publication. This is wasteful as those models can help other researchers to reproduce and analyze the publication while they read along with the text. Even more importantly, the scientific value of the generated models, that are evaluated by experts, is maximized by opening them up for new types of research, thus facilitating multi-disciplinary collaborations for today's complex biodiversity conservation challenges. The overall objective of this work is therefore to propose a self-managed cloud based architecture of a digital ecosystem for species distribution modeling to assist this process.