Interspecific interactions and spatial heterogeneity: using key species to up-scale and map ecosystem functioning

Abstract

Soft sediments cover most of the ocean seafloor and dominate estuarine and coastal habitats. Most of our current knowledge on the functioning of these ecosystems is derived from experimental studies conducted at scales that are much smaller than those most relevant to society. The results of these experiment can be hard to extrapolate beyond their scope and the outcomes can be affected by the way we scale them. Moreover, heterogeneity and processes interacting across scales of space and time can further hinder our ability to extend the generality of experimental studies. Besides, due to the challenge of extensive sampling marine ecosystems, large scale ecosystem models mainly rely on physical attributes and often overlook the role of the underlying biodiversity. In this dissertation, I investigate at the existence of species–ecosystem functions relationships in heterogeneous marine landscapes and their persistence across different spatial scales. In particular, I look into the role of key, functionally important infaunal species and of their interactions for sediment biogeochemistry and then up-scale this information to create ecologically nuanced maps of ecosystem functions at the landscape level. After providing a general introduction (Chapter 1), I begin by investigating the interaction between two functionally important but different species (Macomona liliana and Macroclymenella stewartensis) on sediment biogeochemistry in a laboratory experiment (Chapter 2). I then explore the importance of transitional areas, where the distribution of these two species overlap, for the overall ecosystem functioning and I weight the role of the two key species compared to that of the rest of the community (Chapter 3). Finally, I use this information to create models that relate ecosystem functions rates to key species and extrapolate the models through a high-resolution drone survey of the distribution of those species (Chapter 4). As we then demonstrate, scaling these ecological relationships without adequately taking into account the role of biodiversity and heterogeneity would lead to inaccurate results that are more sensitive to scaling methods chosen than to the ecological characteristics of the system (Chapter 5). I demonstrated that interspecific interactions and the heterogeneity of processes generate differences of orders of magnitudes in the delivery of functions. Most of these interactions happen in transition areas, where patches of different species overlap, creating ecological boundaries. The contribution of these areas to the overall functioning of heterogeneous systems is significant and needs to be taken into account to accurately estimate functioning at coarse scales. Our findings show how rise in the last decades of new remote sensing technologies and artificial intelligence allows the extrapolation of this complex information to larger extents and the creation of ecologically meaningful maps of ecosystems.