The biology of populations and ecosystems has recently undergone both technological and conceptual advances, leading to an increasing recognition of the complexity of the studied systems. This awareness for several years has been increasingly motivated by the interest of modern society in understanding the position of Humans in the ecosystem and their role in preserving the diversity of life on earth and maintaining a sustainable development. Thus, new questions have emerged on the role of biodiversity in the functioning of ecosystems and on the interpretation of the different sources of its spatio-temporal variability. On the pedagogical level, this leads to redefine, in addition to the basic knowledge necessary for training in ecology, the complementary knowledge resulting from the research of ecosystem studies. In this context, the boundaries between disciplines concerned with the individual (behaviour and physiology), the population (dynamics, demography and growth) and the community (interaction, regulation, evolution) tend to blur. The functioning of the ecosystem is then considered as the result of biotic interactions between species, according to their own physiological, demographic or behavioural logics and their environment. Thus, research focused on the individual contributes significantly to the understanding of how the ecosystem works. We are therefore moving towards a functional ecology that is interested in the dynamics and functioning of ecological systems at different spatial and temporal scales. It is in this context to understand and analyze the consequences of spatial-temporal heterogeneity on the dynamics and functioning of ecosystems that we place this teaching module.

In addition to the basic knowledge at different levels of ecosystem organization acquired in other disciplines, the purpose of this course is to address the notion of ecosystem complexity as a whole in two aspects. The first aspect is related to sampling strategies in ecology. Ecological studies are largely based on the interpretation of in situ data. However, the relevance of the results obtained depends to a large extent on the adequacy of the sampling strategy adopted and the data presentation and processing techniques. It is therefore this preliminary step before the data is obtained that will be discussed and detailed. In other words, the fundamental questions addressed will be: when, where and how to do the sampling to carry out such a study in ecology? What is the sampling theory and what are the difficulties of sampling in ecology?

A sampling strategy depends on the ecological and/or operational question asked and requires the implementation of a scientific approach beforehand (pre-model), identification of sampling constraints and development of a reasoned sampling strategy. It is these crucial steps that we will address and develop in this first part of the teachings. SAM3E is based on several representative examples of the diversity of treated ecosystems (terrestrial, aquatic, wetlands, etc).

The second part of this course is dedicated to introduce ecosystem modelling. The modern biologist is often led to use or develop formal tools derived from mathematics or computer science to model the studied system. It is therefore essential to familiarize students with these modelling techniques that are increasingly used in most disciplines. The aim here is not to give a detailed modelling course but to illustrate the modelling approach with a few specific examples and to use suitable modelling tools. Like the first part of the course, we will focus more on the preliminary but indispensable phase of the modelling that leads to develop a conceptual diagram of the functioning of an ecosystem (system studied). This forces us to develop a spirit of synthesis and simplification indispensable to the study of the ecosystem. Ecosystem modelling tools based on new computer techniques (multi-agent systems, artificial intelligence, biological simulators) for biologists will be used during the practical classes sessions to illustrate the course. In particular, we will use the MOBIDYC modelling platform for biologists and non-computer encoding users. The practical work in the computer room will allow students to create (or use) different individual-based models (easily accessible using the MOBIDYC tool) and to test certain fundamental concepts in ecology (density-dependence, relationship between individual and population, spatial heterogeneity, life cycle, inheritance of attributes (traits), genetic drift, etc.).