The shallow Antarctic continental shelf experiences strong physical disturbances in the form of ice-scour which play a key role in maintaining biodiversity. Due to climate change, both the extent and duration of sea ice cover is expected to rapidly decline, leading to complex shifts in disturbance regimes with unknown impacts on successional dynamics and biodiversity in benthic communities. We introduce an individual-based model to study the assemblage dynamics of Bryozoa, sessile, suspension feeding animals which are key pioneers in the shallows. The model captures colonisation events, colony growth as well as intra- and interspecific overgrowth competition for space between colonies. Mortality due to predation is modelled as the removal of zooids within small areas of the model world. Using the model, we simulate various disturbance regimes, e.g. by varying the timing of ice-scour events, the length of the growing season and the spatial distribution of predation events. We find that the timing of ice-scour events throughout the growing season has only minor short-term effects on successional dynamics in bryozoan assemblages, while an extended growing season substantially accelerates succession in the long term. We furthermore find that relatively rare but large predation events lead to a slower recovery, whereas relatively frequent but small events result in a faster succession with higher overall abundances. These results highlight that in order to understand how benthic biodiversity will be impacted by climate change, it is necessary to consider the interplay between biotic interactions and complex changes in physical disturbance regimes.
The Hidden Power of Normalization: Exponential Capacity Control in Deep Neural Networks
arXiv:2511.00958v1 Announce Type: cross Abstract: Normalization methods are fundamental components of modern deep neural networks (DNNs). Empirically, they are known to stabilize optimization dynamics and