Ontology Confounds Reproducibility in Ecology and Climate Science

Abstract

Abstract:The ability to reproduce ecological or climate change experiments or quantitative results is a hallmark of science, but not the only one. In addition, a scientist who subsequently examines original research must be able to reproduce the interpretation of the observations. Confidence in both ecology and climate science research is undermined when an a priori interpretation cannot be confirmed. This will happen when an a priori (before investigation) assumption was introduced by the original author(s), even inadvertently, regarding the nature of the universe or a subset of it. Such an assumption is here termed an ontological assumption when it then is treated as invariant; that is, it is not changed as unsupportive evidence, whether positive or negative, is accumulated. It thus conditions the interpretation of observed scientific phenomena. The results of the scientific investigation are therefore interpreted within the framework of this assumption, even though this framework is at variance with actual results obtained.Recent decades have seen increasing use of paleoclimatic studies as predictors of what will happen as current global climate change unfolds. Here we consider how the results from impact studies of paleoclimatic change on biotic communities have been used as a basis for neoclimatic and ecological research and prediction. We show that in some cases at least the observed results of these paleoclimatic studies were not in accord with the original interpretations, the paleoclimatic workers having held onto unwarranted ontological assumptions in the face of observations to the contrary. This practice is a threat to validity and undermines the applicability of these original paleonvironmental studies to later neoenvironmental work. For example, contrary to expectations, Pleistocene glacial-interglacial cycles did not always impact on community diversity, whether measured by species richness or the information function. Likewise, the quantitative assemblage turnover index (ATI) shows that in the greenhouse world of the Eocene, communities actually were little impacted by lithological changes associated with quantitatively-determined climatic perturbations. We demonstrate that the application of simple statistical measures can draw attention to unwarranted paleoecological ontological assumptions and avoid their being applied inappropriately to neoecological work.

Foraminifera on the Demerara Rise offshore Surinam: crustal subsidence or shallowing of an oxygen minimum zone?

“…Hayek & Wilson (2013) developed an assemblage turnover index (ATI), which they applied to the upper Quaternary section of ODP Hole 994C (Blake Ridge) (see also Wilson & Hayek 2014). For a set of samples from a stratigraphic section, the ATI between each pair of adjacent samples is calculated from in which ATI s is the between-sample assemblage turnover and p i 1 and p i 2 are the proportional abundances of the i th species.…”

Response to reviewers

“…That is, we showed that, despite ODP Site 926As being in an oligotrophic area, the community after the onset of glaciations was less stable at this site than on the nutrient-rich Demerara Rise (ODP Site 1261A). Application of the Conditioned-on-Boundary Index (CoBI) to the 2.54 Ma boundary, following the method presented by Wilson and Hayek (2014a), provided new insights, highlighting differences in the niches of Alabaminella weddellensis and Epistominella exigua,…”