Research interests

Handling non-homogeneous and preferential sampling

I developed an integrated spatio-temporal hierarchical model to map fish distribution. It feeds on massive catch declaration data and standardized fine-scale scientific survey data. It accounts for preferential sampling of the commercial declaration data.

You’ll find some description of this approach in the following references:

Handling change of support

I also developed a statistical approach handling change of support for complex environmental data (zero-inflated data with long tails). It allows to combine (massive) areal-level data and point-level data to map fish distribution.

Conceptual diagram of the hierarchical model

Modelling mechanisms in hierarchical spatio-temporal models

I have worked on the development of spatio-temporal population dynamics models that combines massive sources of data to tackle the effect of climate change on marine populations.

See:

  • Olmos et al. (2023) for a spatio-temporal model representing population dynamics of the Snow Crab in the Eastern Bering Sea.

  • Rovellini et al. (2024) for a mechanistic spatio-temporal model linking climate stressors to ecological processes.

Estimates of abundance of the spatio-temporal stock assessment method developped by Olmos et al. (2023)

Multivariate analysis of spatial/spatio-temporal data

Massive data sources give access to huge amount of spatial information over long time series. Methods to decompose the spatio-temporal variability of the signal into simple spatial and temporal patterns have been developed for the past 50 years based on the EOFs method (Empirical Orthogonal Functions). I am interested in the development of these methods, how these methods are related between each other and how they can be used to analyse environmental spatio-temporal processes in the context of climate change.

Here are some references to my work:

EOFs results for common sole in the Bay of Biscay between 2008 and 2018