The use of functional traits to describe community structure is a promising approach to reveal generalities across organisms and ecosystems. Plant ecologists have demonstrated the importance of traits in explaining community structure, competitive interactions as well as ecosystem functioning. The application of trait‐based methods to more complex communities such as food webs is however more challenging owing to the diversity of animal characteristics and of interactions. The objective of this study was to determine how functional structure is related to food web structure. We consider that food web structure is the result of 1) the match between consumer and resource traits, which determine the occurence of a trophic interaction between them, and 2) the distribution of functional traits in the community. We implemented a statistical approach to assess whether or not 35 466 pairwise interactions between soil organisms are constrained by trait‐matching and then used a Procrustes analysis to investigate correlations between functional indices and network properties across 48 sites. We found that the occurrence of trophic interactions is well predicted by matching the traits of the resource with those of the consumer. Taxonomy and body mass of both species were the most important traits for the determination of an interaction. As a consequence, functional evenness and the variance of certain traits in the community were correlated to trophic complementarity between species, while trait identity, more than diversity, was related to network topology. The analysis was however limited by trait data availability, and a coarse resolution of certain taxonomic groups in our dataset. These limitations explain the importance of taxonomy, as well as the complexity of the statistical model needed. Our results outline the important implications of trait composition on ecological networks, opening promising avenues of research into the relationship between functional diversity and ecosystem functioning in multi‐trophic systems.

Species interactions are a key component of ecosystems but we generally have an incomplete picture of who-eats-who in a given community. Different techniques have been devised to predict species interactions using theoretical models or abundances. Here, we explore the K nearest neighbour approach, with a special emphasis on recommendation, along with a supervised machine learning technique. Recommenders are algorithms developed for companies like Netflix to predict whether a customer will like a product given the preferences of similar customers. These machine learning techniques are well-suited to study binary ecological interactions since they focus on positive-only data. By removing a prey from a predator, we find that recommenders can guess the missing prey around 50% of the times on the first try, with up to 881 possibilities. Traits do not improve significantly the results for the K nearest neighbour, although a simple test with a supervised learning approach (random forests) show we can predict interactions with high accuracy using only three traits per species. This result shows that binary interactions can be predicted without regard to the ecological community given only three variables: body mass and two variables for the species’ phylogeny. These techniques are complementary, as recommenders can predict interactions in the absence of traits, using only information about other species’ interactions, while supervised learning algorithms such as random forests base their predictions on traits only but do not exploit other species’ interactions. Further work should focus on developing custom similarity measures specialized for ecology to improve the KNN algorithms and using richer data to capture indirect relationships between species.

Species interactions are a key component of ecosystems but we generally have an incomplete picture of who-eats-who in a given community. Different techniques have been devised to predict species interactions using theoretical models or abundances. Here, we explore the K nearest neighbour approach, with a special emphasis on recommendation, along with other machine learning techniques. Recommenders are algorithms developed for companies like Netflix to predict if a customer would like a product given the preferences of similar customers. These machine learning techniques are well-suited to study binary ecological interactions since they focus on positive-only data. We also explore how the K nearest neighbour approach can be used with both positive and negative information, in which case the goal of the algorithm is to fill missing entries from a matrix (imputation). By removing a prey from a predator, we find that recommenders can guess the missing prey around 50% of the times on the first try, with up to 881 possibilities. Traits do not improve significantly the results for the K nearest neighbour, although a simple test with a supervised learning approach (random forests) show we can predict interactions with high accuracy using only three traits per species. This result shows that binary interactions can be predicted without regard to the ecological community given only three variables: body mass and two variables for the species' phylogeny. These techniques are complementary, as recommenders can predict interactions in the absence of traits, using only information about other species' interactions, while supervised learning algorithms such as random forests base their predictions on traits only but do not exploit other species' interactions. Further work should focus on developing custom similarity measures specialized to ecology to improve the KNN algorithms and using richer data to capture indirect relationships between species.

Temperatures in mountain areas are increasing at a higher rate than the Northern Hemisphere land average, but how fauna may respond, in particular in terms of phenology, remains poorly understood. The aim of this study was to assess how elevation could modify the relationships between climate variability (air temperature and snow melt‐out date), the timing of plant phenology and egg‐laying date of the coal tit (Periparus ater). We collected 9 years (2011–2019) of data on egg‐laying date, spring air temperature, snow melt‐out date, and larch budburst date at two elevations (~1,300 m and ~1,900 m asl) on a slope located in the Mont‐Blanc Massif in the French Alps. We found that at low elevation, larch budburst date had a direct influence on egg‐laying date, while at high‐altitude snow melt‐out date was the limiting factor. At both elevations, air temperature had a similar effect on egg‐laying date, but was a poorer predictor than larch budburst or snowmelt date. Our results shed light on proximate drivers of breeding phenology responses to interannual climate variability in mountain areas and suggest that factors directly influencing species phenology vary at different elevations. Predicting the future responses of species in a climate change context will require testing the transferability of models and accounting for nonstationary relationships between environmental predictors and the timing of phenological events.

The alarming decline of amphibians around the world calls for complementary studies to better understand their responses to climate change. In mountain environments, water resources linked to snowmelt play a major role in allowing amphibians to complete tadpole metamorphosis. As snow cover duration has significantly decreased since the 1970s, amphibian populations could be strongly impacted by climate warming, and even more in high elevation sites where air temperatures are increasing at a higher rate than at low elevation. In this context, we investigated common frog (Rana temporaria) breeding phenology at two different elevations and explored the threats that this species faces in a climate change context. Our objectives were to understand how environmental variables influence the timing of breeding phenology of the common frog, and explore the threats that amphibians face in the context of climate change in mountain areas. To address these questions, we collected 11 years (2009–2019) of data on egg-spawning date, tadpole development stages, snowmelt date, air temperature, rainfall and drying up of wetland pools at ∼1,300 and ∼1,900 m a.s.l. in the French Alps. We found an advancement of the egg-spawning date and snowmelt date at low elevation but a delay at high elevations for both variables. Our results demonstrated a strong positive relationship between egg-spawning date and snowmelt date at both elevations. We also observed that the risk of frost exposure increased faster at high elevation as egg-spawning date advanced than at low elevation, and that drying up of wetland pools led to tadpole mortality at the high elevation site. Within the context of climate change, egg-spawning date is expected to happen earlier in the future and eggs and tadpoles of common frogs may face higher risk of frost exposure, while wetland drying may lead to higher larval mortality. However, population dynamics studies are needed to test these hypotheses and to assess impacts at the population level. Our results highlight climate-related threats to common frog populations in mountain environments, but additional research should be conducted to forecast how climate change may benefit or harm amphibian populations, and inform conservation and land management plans in the future.