Stressors act at the individual level, and responses at the community level emerge from individual and population processes. Project A18 aims to obtain a better understanding of the mechanisms that regulate responses at the individual and subsequently at the population levels using empirical and mechanistic modelling. In Phase I, we focussed on the role of physiological tolerance during the phase of degradation. We developed a model for mechanistically simulating the dynamics of dissolved oxygen, as one of the main stressors affecting individuals, considering the influence of temperature and salinity. Building on these results, we evaluated the physiological tolerance of freshwater diatoms to temperature and salinity. We conducted a global meta-analysis to assess the influence of salinity on the thermal tolerance of aquatic organisms. Usually, only physiological tolerance is included in the stressor-response relationship at the individual level. Our understanding of the responses to stressor release is also limited. In Phase II, we will include the other processes of the Asymmetric Response Concept (ARC), i.e. dispersal and biotic interactions, especially with regard to the recovery phase and apply respective models to microorganisms (diatoms) and macroorganisms (invertebrates and fish). We will test three specific hypotheses that are related to RESIST’s Main Hypotheses 1 and 2. First, responses at the individual and population level will be empirically modelled. A population growth model will be developed to simulate the effects of drought (one of the focal stressors in Phase II) on diatoms. In addition, we will conduct a meta-analysis to assess the potential of aquatic organisms to recover from temperature increases using published data on their tolerance to acute and chronic thermal stress. Second, we will qualitatively evaluate the relative importance of physiological tolerance compared with the other ARC processes in responses at the individual or population level under conditions of degradation and recovery using data from experimental projects. Third, the dynamics and structure of populations will be simulated with mechanistic population models. A stage-structured population model will be developed for Gammarus pulex. The individuals within a population will be divided into five groups, corresponding to two juvenile stages and three adult stages. In these models, the environmental state (including temperature, salinity, or water level) will be considered in deriving the processes at the individual level (e.g. reproduction and survival). Fourth, a mechanistic metapopulation model will be developed for G. pulex considering mortality, movement, growth, and reproduction in three types of ecosystems: ditches, streams, and fragmented landscapes. In all these mechanistic models, the relative importance of ARC processes will be evaluated in sensitivity analyses. Overall, A18 will provide a qualitative and quantitative assessment on the contribution of all ARC processes to responses at the individual and population levels.

Luan Farias (University of Duisburg-Essen)

Delineating multiple stressor-response relationships at the individual level: A mechanistic modelling approach

The goal of this thesis is to model the causal chain of effects from multiple stressors to environmental variables to organisms in stream ecosystems. The model will be mechanistic and operates on the individual level. The model will simulate binary combinations of the stressors temperature increase, salinisation, and hydromorphological modifications. These stressors influence environmental variables like dissolved oxygen (DO), dissolved organic carbon (DOC) and electrical conductivity (EC) as well as organisms (microorganisms, microphytobenthos, parasites, invertebrates, and fish). The environmental variables also interact with each other and affect the organisms. The stressors can affect the organisms directly or indirectly through the environmental variables.

The organisms are classified by body size and environmental effects on them are calculated in relation to these size classes. Depending on the size we expect direct or indirect effects of the stressors to be more prevalent. The model will increase the understanding of stressor-response mechanisms and help predict responses of organisms in streams to multiple stressors. In further steps biotic interactions can be added to elevate the model from the individual level to the community level.

Contact: luan.farias@uni-due.de

First Supervisor: Dr. T. T. Yen Le (University of Duisburg-Essen, Aquatic Ecology)
Second Supervisor: Prof. Dr. Ralf Schäfer (University of Koblenz-Landau, Quantitative Landscape Ecology)
Mentor: Dr.-Ing. Daniel Teschlade (Ruhrverband)