Résumé: Aim Although species distribution models (SDMs) traditionally link species occurrences to free-air temperature data at coarse spatio-temporal resolution, the distribution of organisms might instead be driven by temperatures more proximal to their habitats. Several solutions are currently available, such as downscaled or interpolated coarse-grained free-air temperatures, satellite-measured land surface temperatures (LST) or in-situ-measured soil temperatures. A comprehensive comparison of temperature data sources and their performance in SDMs is, however, currently lacking. Location Northern Scandinavia. Time period 1970-2017. Major taxa studied Higher plants. Methods We evaluated different sources of temperature data (WorldClim, CHELSA, MODIS, E-OBS, topoclimate and soil temperature from miniature data loggers), differing in spatial resolution (from 1 '' to 0.1 degrees), measurement focus (free-air, ground-surface or soil temperature) and temporal extent (year-long versus long-term averages), and used them to fit SDMs for 50 plant species with different growth forms in a high-latitudinal mountain region. Results Differences between these temperature data sources originating from measurement focus and temporal extent overshadow the effects of temporal climatic differences and spatio-temporal resolution, with elevational lapse rates ranging from -0.6 degrees C per 100 m for long-term free-air temperature data to -0.2 degrees C per 100 m for in-situ soil temperatures. Most importantly, we found that the performance of the temperature data in SDMs depended on the growth forms of species. The use of in-situ soil temperatures improved the explanatory power of our SDMs (R-2 on average +16%), especially for forbs and graminoids (R-2 +24 and +21% on average, respectively) compared with the other data sources. Main conclusions We suggest that future studies using SDMs should use the temperature dataset that best reflects the ecology of the species, rather than automatically using coarse-grained data from WorldClim or CHELSA.

Résumé: The study is the first attempt to (i) model spring food webs in three SW Mediterranean ecosystems which are under different anthropogenic pressures and (ii) to project the consequence of this stress on their function. Linear inverse models were built using the Monte Carlo method coupled with Markov Chains to characterize the food-web status of the Lagoon, the Channel (inshore waters under high eutrophication and chemical contamination) and the Bay of Bizerte (offshore waters under less anthropogenic pressure). Ecological network analysis was used for the description of structural and functional properties of each food web and for inter-ecosystem comparisons. Our results showed that more carbon was produced by phytoplankton in the inshore waters (966–1234 mg C m-2 d-1) compared to the Bay (727 mg C m-2 d-1). The total ecosystem carbon inputs into the three food webs was supported by high primary production, which was mainly due to >10µm algae. However, the three carbon pathways were characterized by low detritivory and a high herbivory which was mainly assigned to protozooplankton. This latter was efficient in channelling biogenic carbon. In the Lagoon and the Channel, foods webs acted almost as a multivorous structure with a tendency towards herbivorous one, whereas in the Bay the herbivorous pathway was more dominant. Ecological indices revealed that the Lagoon and the Channel food webs/systems had high total system throughput and thus were more active than the Bay. The Bay food web, which had a high relative ascendency value, was more organized and specialized. This inter–ecosystem difference could be due to the varying levels of anthropogenic impact among sites. Indeed, the low value of Finn’s cycling index indicated that the three systems are disturbed, but the Lagoon and the Channel, with low average path lengths, appeared to be more stressed, as both sites have undergone higher chemical pollution and nutrient loading. This study shows that ecosystem models combined with ecological indices provide a powerful approach to detect change in environmental status and anthropogenic impacts.

Résumé: The strengths and limitations of a Bayesian state-space modelling framework are investigated for a De Lury depletion model that accommodates two recruitment pulses per year. The framework was applied to the Moroccan fishery for common octopus ( Octopus vulgaris) between 1982 and 2002. To allow identifiability, natural mortality ( M) and the recruitment rhythm were fixed, and the variance of both process and observation errors were assumed to be equal. A simulation-estimation ( SE) approach was derived to test the performance of the method. If the data showed responses to harvest, the estimates of the most important figures, i.e. the initial abundance and the second recruitment pulse, were accurate, with relatively small bias. Results confirm that greater depletion yields smaller bias and uncertainty and that inferences are sensitive to the mis-specification of M. The 21 depletion series in the Moroccan dataset were jointly treated in a hierarchical model including random walk to capture the systematic fluctuations in estimates of catchability and initial abundance. The model provides estimates of the annual recruitment and monthly octopus population size. The recruitment estimates could be used to investigate the link between recruitment variability and the coastal North African upwelling regime to improve understanding of the dynamics and management of octopus stocks.