Generalized additive mixed models for disentangling long‐term trends, local anomalies, and seasonality in fruit tree phenology

A significant increase in fruit presence was observed in Bwindi Impenetrable National Park (BINP), Uganda, during 2009-2010, driven by heightened productivity across most species. This finding highlights the park's dynamic fruiting patterns, characterized by both regular seasonality—peaking at the end of one dry season—and notable interannual variability. Researchers employed generalized additive mixed models (GAMMs) to disentangle these trends from empirical data, demonstrating the method's accuracy in identifying nonlinearity and seasonality through simulation experiments. The study provides a statistical framework for assessing the impacts of abiotic factors, including climate change, on fruiting phenology, with implications for understanding frugivore population biology in BINP.

Polansky, L., & Robbins, M. M. (2013). Generalized additive mixed models for disentangling long‐term trends, local anomalies, and seasonality in fruit tree phenology. Ecology and Evolution, 3(9), 3141–3151. Portico. https://doi.org/10.1002/ece3.707