Biology subjects

Hull, P. M.

Publications and source records attributed to Hull, P. M..

3 recordsLinked to original sources

bioRxiv2024

Stabilization of fluctuating population dynamics via the evolution of dormancy

Dormancy is usually understood as a strategy for coping with extrinsically variable environments, but intrinsic population fluctuations also create conditions where dormancy is adaptive. By analyzing simple population models, we show that, very generally, population fluctuations favor the evolution of dormancy, but dormancy stabilizes population dynamics. This sets up a feedback loop that can enable the coexistence of alternative dormancy strategies. Over longer timescales, we show that evolution of dormancy to an evolutionary stable state can drive populations to the edge of stability, where dynamics are only weakly stabilized. We briefly consider how these conclusions are likely to apply in more complex community contexts. Our results suggest that chaos and high-amplitude population cycles are highly vulnerable to invasion and subsequent stabilization by dormancy, potentially explaining their rarity. At the same time, the propensity of ecological dynamics to fluctuate may be an underappreciated driver of the evolution of dormancy.

ecology↗

bioRxiv2024

A framework for reconstructing ancient food webs using functional trait data

O_LIFood webs provide quantitative insights into the structure and dynamics of ecological communities. Previous work has shown their utility in understanding community responses to modern and ancient perturbations, including anthropogenic change and mass extinctions. However, few ancient food webs have been reconstructed due to difficulties assessing trophic interactions amongst extinct species derived from an incomplete fossil record. C_LIO_LIWe present and assess the Paleo Food web Inference Model (PFIM). PFIM uses functional trait data--predictive of interactions in modern ecosystems and commonly available for fossil organisms--to reconstruct ancient food webs. We test the model by (i) applying it to four modern ecosystems with empirical constrained food webs to directly compare PFIM-constructed networks to their empirical counterparts, (ii) by carefully comparing discrepancies between PFIM-inferred and empirical webs in one of those systems, and (iii) by comparing networks describing feasible trophic interactions ("feasible webs") with networks to which we superimpose characteristic interaction distributions derived from modern theory ("realized webs"). As a proof of concept, we then apply the method to faunal data from two Cambrian fossil deposits to reconstruct ancient trophic systems. C_LIO_LIPFIM-inferred feasible food webs successfully predict [~]70% of trophic interactions across four modern systems. Furthermore, inferred food webs with enforced interaction distributions (i.e., realized webs) accurately predict [~]90% of interactions. Comparisons with a global database of trophic interactions and other food web models, suggest that under sampling of empirical webs accounts for up to 21% of the remaining differences between PFIM and empirical food webs. C_LIO_LIFood webs can be reasonably approximated by inferring trophic interactions based upon life habit traits. This study provides the foundation to use trait-based inference models across the fossil record to examine ancient food webs and community evolution. C_LI

paleontology↗

bioRxiv2022

Automated community ecology using deep learning: a case study of planktonic foraminifera

The development of deep learning methods using convolutional neural networks (CNNs) has revolutionised the field of computer vision in recent years. The automation of taxonomic identification using CNNs leads naturally to the use of such technology for rapidly generating large organismal datasets in order to study the evolutionary and ecological dynamics of biological communities across time and space. While CNNs have been used to train machine learning classifiers that can identify organisms to the species level for several groups, this vision of automated community ecology has yet to be thoroughly tested or fulfilled. Here, we present a case study of automated community ecology using a large dataset of Atlantic planktonic foraminifera for which the generation of species labels and morphometric measurements was completely automated. We compare standard community diversity metrics between the fully automated dataset and a "traditional" dataset with human-identified specimens. We show that there is high congruence between the results, and that machine classifications help avoid biases that can result in the inference of misleading biodiversity patterns. Our study demonstrates the viability and potential of fully automated community ecology and sets the stage for a new era of ecological and evolutionary inquiry driven by artificial intelligence.

ecology↗