ATTRIBUTING EXTINCTIONS TO ENVIRONMENTAL FORCINGS: AN INVERSE APPROACH

Patterns of selectivity are critical in understanding and constraining past mass extinctions. However, insight into the links between climate perturbations and resulting environmental shifts is usually confounded by incomplete occurrence records and sometimes limited proxy data. This can lead to a broad set of potential extinction scenarios, as climate perturbations/transitions are coupled to a specie’s distribution. We address this issue by taking a systematic inverse method approach to evaluate which climate transitions match with known species occurrence information. This approach utilizes a large range of hypothetical climatic shifts (both cooling and warming scenarios) created by cGENIE, an Earth System models capable of creating ocean conditions based on alternating climate forcings. These climate models can then be used in probabilistic ecological niche models (ENMs) which are based on observed occurrences pre-climatic transition and tested with model output and post-climatic transition occurrences. Because ENM’s are presence-only models, we adapt a threshold-based evaluation system dependent on presence/absence records post-climate transition. This information can thus be used to evaluate the probability of extinction vs surviving during a climate transition, which can be directly compared to species distributions, resulting in insight into which hypothetical sequence of climatic/environmental changes are most consistent with the known fossil record. We will present preliminary results of this study as a blueprint for how to handle a larger number of model parameters and empirical information (such proxy information). We focus on the Late Ordovician Mass Extinction and its relationship to graptolites. Graptolites are particularly useful for this methodology, as they are widespread, diverse, and undergo large macroevolutionary changes temporally linked to climatic change. In addition, there are known patterns of extinction, diversity blooms, and dispersal during this time, which can be directly evaluated using ENMs.