Interpretation of interannual variability in long-term aquatic ecological surveys

Abstract

Long-term ecological surveys (LTES) often exhibit strong variability among sampling dates. The use and interpretation of such interannual variability is challenging due to the combination of multiple processes involved and sampling uncertainty. Here, we analysed the interannual variability in ∼30 years of 150 species density (fish and invertebrate) and environmental observation time series in four aquatic systems (stream, river, estuary, and marine continental shelf) with different sampling efforts to identify the information provided by this variability. We tested, using two empirical methods, whether we could observe simultaneous fluctuation between detrended time series corresponding to widely acknowledged assumptions about aquatic population dynamics: spatial effects, cohort effects, and environmental effects. We found a low number of significant results (36%, 9%, and 0% for spatial, cohort, and environmental effects, respectively), suggesting that sampling uncertainty overrode the effects of biological processes. Our study does not question the relevance of LTES for detecting important trends, but clearly indicates that the statistical power to interpret interannual variations in aquatic species densities is low, especially in large systems where the degree of sampling effort is always limited.

Quantifying the relative importance of stock level, river temperature and discharge on the abundance of juvenile Atlantic salmon (Salmo salar)

“…Many previous studies have attempted to interpret interannual variability in juvenile salmonid abundance solely in relation to environmental drivers (Blum, Kanno, & Letcher, 2018; Cauvy‐Fraunié et al, 2019; Jensen & Johnsen, 1999; Lobón‐Cerviá, Rasmussen, & Mortensen, 2017). This potentially results in erroneous attribution of processes, or for environmental effects to go undetected due to apparently high levels of noise in recruitment data that is associated with temporally varying stock levels.…”

Abundance trends for river macroinvertebrates vary across taxa, trophic group and river typology

“…Finally, the dynamics of invertebrate trends are difficult to capture and model over the long term due to high interannual variation that is inherent across these taxa (Baranov et al, 2020 ; Cauvy‐Fraunié et al, 2020 ); this appears to also be the case with our data, shown in figures 2 and 3 . We have chosen to model long‐term abundance change of macroinvertebrates using hierarchical linear modelling, and while this approach allows us to provide our best estimate of how abundance has changed on average since 2002, the models presented do not capture changes from 1 year to the next, nor explain occasional short‐term non‐linear patterns in geometric means.…”

“…Although we discuss family-level trends in the context of water quality changes (due to different families varying in response to water quality improvement and pollution), within families there is also variation among species in their sensitivity to water quality metrics, or their 'saprobic index', which we were not able to capture in this analysis (Metcalfe, 1989). Finally, the dynamics of invertebrate trends are difficult to capture and model over the long term due to high interannual variation that is inherent across these taxa (Baranov et al, 2020;Cauvy-Fraunié et al, 2020); this appears to also be the case with our data, shown in figures 2 and 3. We have chosen to model long-term abundance change of macroinvertebrates using hierarchical linear modelling, and while this approach allows us to provide our best estimate of how abundance has changed on average since 2002, the models presented do not capture changes from 1 year to the next, nor explain occasional short-term non-linear patterns in geometric means.…”

“…Understanding where abundance of important invertebrates has declined, including freshwater macroinvertebrates, has been hampered by a lack of long‐term data from standardized monitoring schemes (Isaac & Pocock, 2015 ; Powney et al, 2015 ; Thomas et al, 2019 ). Long‐term trends in large systems are also difficult to characterize with statistical confidence as sampling effort is often limited compared to the system scale, causing high fluctuation in interannual variation (Cauvy‐Fraunié et al, 2020 ). An exception is abundance data for riverine freshwater macroinvertebrates collected over multiple decades by the Environment Agency (EA); the government authority responsible for monitoring the health and water quality of freshwaters in England.…”

Exploring temporal patterns in fish feeding ecology: Are ontogenetic dietary shifts stable over time?

“…Temporal variation (i.e., year-to-year variation) in ecological conditions (e.g., prey availability and species interactions) may impose an important limitation in recognising meaningful conclusions about biological processes like feeding when studies are projected on short-time scale designs. More specifically, interannual variability might drive ecological research focused on short-term studies towards less reliable conclusions compared to studies conducted on a longer term (Purdom et al 2015;Cauvy-Fraunié et al 2020). In addition, a practical issue with ontogenetic niche shift studies is that the diversion of a population sample into ontogenetic subgroups often possesses a problem with respect to low sample sizes for some of the subgroups.…”