Direct habitat descriptors improve the understanding of the organization of fish and macroinvertebrate communities across a large catchment.

In large-scale aquatic ecological studies, direct habitat descriptors (e.g. water temperature, hydraulics in river reaches) are often approximated by coarse-grain surrogates (e.g. air temperature, discharge respectively) since they are easier to measure or model. However, as biological variability can be very strong at the habitat scale, surrogate variables may have a limited ability to capture all of this variability, which may lead to a lesser understanding of the ecological processes or patterns of interest. In this study, we aimed to compare the capacity of direct habitat descriptors vs. surrogate environmental variables to explain the organization of fish and macroinvertebrate communities across the Loire catchment in France (105 km2). For this purpose, we relied on high-resolution environmental data, extensive biological monitoring data (>1000 sampling stations) and multivariate analyses. Fish and macroinvertebrate abundance datasets were considered both separately and combined to assess the value of a cross-taxa approach. We found that fish and macroinvertebrate communities exhibited weak concordance in their organization and responded differently to the main ecological gradients. Such variations are probably due to fundamental differences in their life-history traits and mobility. Regardless of the biological group considered, direct habitat descriptors (water temperature and local hydraulic variables) consistently explained the organization of fish and macroinvertebrate communities better than surrogate descriptors (air temperature and river discharge). Furthermore, the organization of fish and macroinvertebrate communities was slightly better explained by the combination of direct or surrogate environmental variables when the two biological groups were considered together than when considered separately. Tied together, these results emphasize the importance of using a cross-taxa approach in association with high-resolution direct habitat variables to more accurately explain the organization of aquatic communities.

Thermal signatures identify the influence of dams and ponds on stream temperature at the regional scale.

Anthropogenic impoundments (e.g. large dams, small reservoirs, and ponds) are expanding in number globally, influencing downstream temperature regimes in a diversity of ways that depend on their structure and position along the river continuum. Because of the manifold downstream thermal responses, there has been a paucity of studies characterizing cumulative effect sizes at the catchment scale. Here, we introduce five thermal indicators based on the stream-air temperature relationship that together can identify the altered thermal signatures of dams and ponds. We used this thermal signature approach to evaluate a regional dataset of 330 daily stream temperature time series from stations throughout the Loire River basin, France, from 2008 to 2018. This basin (105 km2) is one of the largest European catchments with contrasting natural and anthropogenic characteristics. The derived thermal signatures were cross-validated with several known catchment characteristics, which strongly supported separation into dam-like, pond-like and natural-like signatures. We characterize the thermal regime of each thermal signature and contextualize it using a set of ecologically relevant thermal metrics. Results indicate that large dams decreased summer stream temperature by 2 °C and delayed the annual stream temperature peak by 23 days relative to the natural regimes. In contrast, the cumulative effects of upstream ponds increased summer stream temperature by 2.3 °C and increased synchrony with air temperature regimes. These thermal signatures thus allow for identifying and quantifying downstream thermal and ecological influences of different types of anthropogenic infrastructures without prior information on the source of modification and upstream water temperature conditions.