COVID, as devastating as it for humans, significantly reduced anthropogenic pressures in all ecological systems since they were confined to their homes.
My question is as the title suggests: when building any kind of a model that captures behaviour/movement analysis of animals on land or water, how do you account for the years 2020-2022? Any model trained on data till 2019 and tested on these years may be inaccurate and any that includes it may train on anomalous patterns that probably won't repeat anytime soon. So do you discard this data? Do you under-weight it? Do you train till 2019 and then test on data from 2023-2024? How exactly are you handling this?
25 July 2025 3:06pm
To clarify, are you talking about a model that carries out automated detection of vocalizations? or a model that detects specific patterns of behavior/movement? I would suspect that the former is not something that may be impacted while training as the fundamental vocalizations/input is not going to change drastically (although see Derryberry et al., where they show variation in spectral characteristics of sparrows at short distance pre and post-covid lockdowns).
Vijay Ramesh