Various modelling techniques are available to understand the temporal and spatial variations of the phenology of species. Scientists often rely on correlative models, which establish a statistical relationship between a response variable (such as species abundance or presence-absence) and a set of predominantly abiotic covariates. The choice of the modelling approach, i.e., the algorithm, is a crucial factor in addressing the multiple sources of variability that can lead to disparate outcomes when different models are applied to the same dataset. This inter-model variability has led to the adoption of ensemble modelling techniques, among which stacked generalisation, which has recently demonstrated its capacity to produce robust results. Stacked ensemble modelling incorporates predictions from multiple base learners or models as inputs for a meta-learner. The meta-learner, in turn, assimilates these predictions and generates a final prediction by combining the information from all the base learners. Our study utilized a recently published dataset documenting egg abundance observations of Aedes albopictus collected using ovitraps. This dataset spans various locations in southern Europe, covering four countries -Albania, France, Italy, and Switzerland- and encompasses multiple seasons from 2010 to 2022. Utilising these ovitrap observations and a set of environmental predictors, we employed a stacked machine learning model to forecast the weekly average number of mosquito eggs. This approach enabled us to i) unearth the seasonal dynamics of Ae. albopictus for 12 years; ii) generate spatio-temporal explicit forecasts of mosquito egg abundance in regions not covered by conventional monitoring initiatives. Beyond its immediate application for public health management, our work presents a versatile modelling framework adaptable to infer the spatio-temporal abundance of various species, extending its relevance beyond the specific case of Ae. albopictus.
