7 Results

The following sections present the main results of the cumulative effects assessment. We first present the combined distributions of taxa and environmental drivers as an introduction of the assessment results. These results essentially concatenate the areas of potential cumulative effects, since effects are constrained by the co-occurrence of taxa and environmental drivers. We then discuss the spatial distribution and the change in spatial distribution between the periods of 2010-2015 and 2016-2021 for the species-scale and network-scale cumulative effects in the study area. Finally, we present the specific results of the network-scale cumulative effects assessment, i.e. the pathways of direct and indirect effects predicted by the cumulative effects assessment.

7.1 Species richness

This sectionpresents an evaluation of species richness, i.e. the number of taxa in each cell of the study grid, from the 205 taxa considered for this assessment. It is important to mention here that the binary distribution of each taxa (i.e. presence-absence) is based on observations for marine mammals and seabirds, while it is based on species distribution modeling predictions for all other marine species considered (see section 6.1). Figure 7.1 presents the assessment of species richness in the study area, and the distribution of all individual taxa may be viewed in the Atlas of the assessment in section 10.2.

We can see that the study area can be divided in two main zones of species richness, i.e. the continental shelf, including the Bay of Fundy, and the continental slope. Species richness is medium to high throughout the continental shelf and the Bay of Fundy, while it is lower on the continental slope. The transition between the shelf and the slope is however characterized by a latitudinal band of higher richness. Most of the shelf shows high species richness, especially the northeast and southwest tips of Nova-Scotia. The western part of the Bay of Fundy also has a particularly high species richness. It is however important to mention that the data used to characterize the distribution of species in the study area characterizes the continental shelf more extensively than the continental slope. Similarly, the survey data used to characterize the distribution of marine species comes from offshore trawl surveys that do not sample coastal waters; this results in lower species richness in coastal waters that should be interpreted in terms of sampling effort and design rather than as a general decrease in species richness.

Figure 7.1: Distribution of species richness, as evaluated from the 205 taxa considered in the cumulative effects assessment and the results of the species distribution modeling. See section 6.1 for more information. Click on figure to view in seperate tab.

7.2 Cumulative drivers

This section presents an evaluation of cumulative drivers, i.e. the sum of the normalized intensity of the 17drivers considered in the assessment, as a means to identify sites most exposed to the potential effects of environmental drivers in the study area. It does not provide an effects assessment, since it only considers the drivers. Still, it allows us to assess the environments that are most likely to be affected by the stressors in our study area (see Beauchesne et al., 2020 for more information on cumulative drivers). Figure 7.2 presents the assessment of cumulative drivers for both periods considered, and all individual drivers may be viewed in the Atlas of the assessment in section 10.1. Cumulative drivers can also be explored interactively through the eDrivers application, which is presented in section 8 of this assessment.

As with species richness, the study area has two main zones of cumulative exposure, i.e. the continental shelf, including the Bay of Fundy, and the continental slope. Drivers are ubiquitous on continental shelf, which as a whole is exposed to at least medium levels of environmental drivers; the continental slope, meanwhile, is exposed to very few drivers when compared to the continental shelf. It is however important to consider that of the 17 drivers considered, only positive and negative sea surface temperature anomalies and shipping are present on the continental slope (refer to section 10.1 of the assessment atlas). Meanwhile, all drivers are found on the continental shelf and in the Bay of Fundy; offshore areas of the southwest continental shelf and the eastern end of the Bay of Fundy are slightly more exposed to multiple drivers. The coasts of Nova-Scotia and of the whole Bay of Fundy are exposed to environmental drivers; this is true especially close to coastal towns, with Saint-John in the Bay of Fundy exhibiting the greatest values of cumulative drivers in the study area.

Figure 7.2: Distribution of cumulative drivers on the Scotian Shelf Bioregion in eastern Canada for the periods of 2010-2015 and 2015-2016. Cumulative drivers are evaluated as the sum of the normalized intensity (scaled between 0 and 1) of the 17 drivers considered for the assessment (\(E_{S_x} = \sum_{j=1}^m S_{j,x}\)); it identifies areas that are most exposed to the potential effects of the environmental drivers considered for the cumulative effects assessment. Refer to Beauchesne et al. (2020) for more details. Click on figure to view in seperate tab.

7.3 Cumulative effects

The methods used to perform the species-scale and network-scale cumulative effects assessments are represented in sections 4.1 and 4.2, respectively.

The main results of this assessment combine all data modules to assess the cumulative effects of environmental drivers on the ecological communities of the Scotian Shelf Bioregion. We present and interpret the results on the spatial distribution of cumulative effects from the species-scale (Figure 7.3) and network-scale (Figure 7.4) assessments jointly, as they identify similar spatial trends in cumulative effects. This is unsurprising, as the distribution of both assessments is driven mainly by the co-occurrence of environmental drivers and species. Considering species interactions amplifies or dampens the relative intensity of cumulative effects at the community scale, i.e. when all species are considered simultaneously. However, it refines the assessment for certain species such as marine mammals for which direct effects cannot properly capture the effects of drivers such as climate change that propagate through the ecological network. It should also be noted that values from the species-scale and network-scale assessments cannot be directly compared, as they are both relative to the component each considered. Hence, a cell with a greater value in the network-scale assessment cannot be interpreted as a greater impact; they can only be compared in relative form. See section 7.5 for more on the direct and indirect cumulative effects, and refer to Beauchesne (2020) for more details on the network-scale assessment.

The assessments of cumulative effects reveal a stark contrast between the continental shelf, including the Bay of Fundy, and the continental slope; this result is in accordance with the lower representation of species and environmental drivers that we previously discussed from the assessments of species richness and cumulative drivers. This result should therefore also take into account the absence of data characterizing environmental drivers in the study area. Still, it remains undeniable that the ecological communities of the continental shelf are more at risk from the cumulative effects of environmental drivers.

On the continental shelf, the whole Bay of Fundy as well as southwesrtern Nova-Scotia extending offshore towards the continental slope appears to be the region most at risk from cumulative effects. The regions northeast of Sable Island, as well as the coastal areas south of Cape Breton are also at greater risk from cumulative effects. Areas of greatest cumulative effects seem to be strongly associated with regions affected by a combination of climate-related drivers, commercial fisheries and shipping (see Atlas). Meanwhile, the middle of the continental shelf, besides a region associated with demersal destructive fisheries, as well as the northeastern corner of the study area close to the Laurentian Channel seem to be less at risk from cumulative effects.

Figure 7.3: Species-scale cumulative effects assessment of 17 drivers on 205 taxa of the Scotian Shelf Bioregion in eastern Canada, using the Halpern et al. (2008) method. See section 4.1 for more details on the approach. Click on figure to view in seperate tab.

Figure 7.4: Network-scale cumulative effects assessment of 17 drivers on 205 taxa of the Scotian Shelf Bioregion in eastern Canada, using the Beauchesne (2020) method explicitly considering the underlying structure of the ecological community. See section 4.2 for more details on the approach. Click on figure to view in seperate tab.

7.4 Change in cumulative effects

Note: driver standardization is currently done on a driver per period basis, e.g. demersal fisheries for 2010-2015 were transformed using data from that period only. This makes the chagne in cumulative effects results less reliable than if standardization was made from the joint distribution of demersal fisheries from 2010-2015 and 2016-2021. This should be changed accordingly so that we may be confident in the results for changes over time.

The cumulative effects assessments were performed on two periods for environmental drivers when data permitted, the first spanning 2010 to 2015 and 2016 to 2021 (see section 6.2 for more details). We thus evaluated the change in cumulative effects for the species-scale and the network-scale assessments (Figure 7.5). As with the cumulative effects themselves, the spatial distribution of relative change in cumulative effects is similar between both assessments. In general, there is a slight decrease in cumulative effects across the Scotian Shelf Bioregion (species-scale assessment: -0.54 \(\pm\) 2.71; network-scale assessment: -1.11 \(\pm\) 5.83). Areas identified at higher risk of cumulative effects previously (see Section 7.3) generally show a decrease in relative intensity, except for part of the Bay of Fundy, where a slight increase is predicted by both assessments.

Figure 7.5: Change in cumulative effects between the periods of 2010-2015 and 2016-2021 on the Scotian Shelf Bioregion in eastern Canada for the species-scale and network-scale assessments. Click on figure to view in seperate tab.

7.5 Pathways of direct and indirect effects

The species-scale and network-scale cumulative effects assessments allow for the exploration of the direct effects of all combinations of the 17 drivers and 205 species considered in the assessment; there are thus over 3485 possible pairs to explore. These pairs form the set of direct pathways of effect – i.e. the links between drivers and their potential impacts on various aspects of an ecosystem (Fisheries and ceans Canada, 2020a) – that environmental drivers can take to affect species. The network-scale assessment, meanwhile, also allows the exploration of the indirect effects of all drivers on all species that arise from species interactions. The metaweb of species interactions (see section 6.4) consists of 4492 predicted links structuring the 205 taxa considered for the assessment; this means that there are \(2^{4492}\) potential indirect pathways of effect through which the food web could be affected by environmental drivers.

Although we obviously cannot explore all of these direct and indirect pathways of effect, Figures 7.6, 7.7 7.8 and 7.9 provide a visual representation of all these combinations and of the contribution of individual drivers to the net, direct and indirect effects predicted on each taxa and groups of taxa considered. The regional effects of environmental drivers on each taxa was assessed as the mean effect over a taxa’s distribution. For example, a taxa with a distribution area of 2 \(km^2\) – i.e. spanning two 1 \(km^2\) grid cells – affected by a single driver with intensities of 0.25 and 0.75 would have a regional cumulative effects score of 0.5 \(km^{-2}\). The greater the spatial distribution of a driver, the greater its regional effect is likely to be; similarly, regional effects are likely to increase with a decrease of the distributional range of a taxa.

Here, we only present the results for the 2016-2021 period, as both periods have qualitatively similar results. In total, there are 0 direct pathways of effects with an average effect score of NaN \(\pm\) NA, and 0 indirect pathways of effects with an average effect score of NaN \(\pm\) NA predicted through the network-scale assessment. Figures 7.6 and 7.7 provide an overview of all the pathways of effects predicted between the drivers and taxa considered in the assessment. These figures should not be interpreted in detail; rather, they provide an at-a-glance view of the amount of information available and explorable through the network-scale cumulative effects assessment. Figures 7.8 and 7.9 complement these figures by presenting the relative contribution of each driver group to the overall direct and indirect effects predicted for taxonomic groups (Figure 7.8) and individual taxa (Figure 7.9).

Together, those figures offer interesting insights into the direct and indirect effects of environmental drivers on the species of the Scotian Shelf. The drivers with the most intense direct and indirect effects across all taxa are shipping, demersal destructive high-bycatch fisheries (e.g. trawls and dredges), and negative sea-bottom temperature anomalies; these are essentially some of the most widely distributed drivers that were considered in the assessment. In general, invertebrates appear more at risk from the cumulative effects of all drivers than vertebrates, especially from direct effects. This reflects broad taxa-specific sensitivities to the effects of environmental drivers. Invertebrates, predominantly composed of benthic species with limited mobility are generally more susceptible to climate drivers (Kroeker et al., 2010, 2013). A few invertebrate species such as hermit crabs (Pagurus; Figure 7.11) and Bolocera sea anemones (Figure 7.10) also show very high scores of cumulative effects (Figure 7.9). These predictions are mainly due to the predicted distribution of those species being very small and typically exposed to a combination of climate and fisheries drivers that affect benthic species; as such, these results should be explored further, particularly the quality of the predicted distribution for those species.

The predicted risks from direct effects to invertebrates are in contrast with those for vertebrates, who are generally mobile species capable of minimizing their exposure to drivers; this effectively makes them less sensitive to climate drivers (Kroeker et al., 2010, 2013; Sydeman et al., 2015). Vertebrates are, however, generally more sensitive to trophically-mediated indirect effects (Beauchesne, 2020; Beauchesne et al., 2021; Stier et al., 2016; Sydeman et al., 2015). They indeed appear to be at risk from the direct effects of few drivers, yet they are greatly at risk from all drivers when species interactions are taken into consideration to evaluate indirect effects. For instance, seabirds are generally unaffected by the direct effects of drivers; similarly, marine mammals and fish are almost exclusively affected by shipping and fisheries, respectively. However, almost all the seabird, marine mammal and fish species considered are greatly affected by the indirected effects of all drivers. Interestingly, the overall intensity of indirect effects on vertebrates tends to be greater than that of direct effects. These observations, available only through the network-scale assessment, could be crucial in the management of species for which the most important sources of stress may be those that are the least obvious, i.e. those arising from ecological interactions. For instance, the distribution of the predicted cumulative effects on Minke whales (Balaenoptera acutorostrata; Figure 7.12) essentially identifies shipping lanes across study areas as the areas where this species is most at risk. However, consideration of indirect effects drastically changes the distribution of cumulative effects and highlights areas that are also affected by fisheries likely targeting whale prey. Solely considering direct effects would thus miss such observations that could be crucial to the management of certain species.

Figure 7.6: Multiplex network of network-scale cumulative effects on the Scotian Shelf Bioregion presenting the metaweb of interactions between the 205 taxa considered (grey connections), the presence of an effect of individual drivers on each taxa (colored connections), the overall cumulative effect on each taxon (taxa point size) and the mean effect of drivers on all taxa (driver point size) Click on figure to view in seperate tab.

Figure 7.7: Multiplex network of network-scale cumulative effects on the Scotian Shelf Bioregion presenting the metaweb of interactions between the 205 taxa considered (grey connections), the presence of a direct (left) or indirect (right) effect of individual drivers on each taxa (colored connections), the overall cumulative effect on each taxon (taxa point size) and the mean effect of drivers on all taxa (driver point size) Click on figure to view in seperate tab.

Figure 7.8: Mean contribution of 17 drivers to the cumulative effects (\(C / km^2\)) of 205 taxa on the Scotian Shelf Bioregion in eastern Canada classified into 9 taxonomic groups. Taxonomic groups for invertebrates (grey box) and vertebrates are divided at the phyla and classes level, respectively. For each group, the mean contribution per species was used to evaluate each driver’s contribution to overall cumulative effects; this can be divided into their direct and indirect contributions to the cumulative effects on each taxon. Direct effects are those attributable to the effect of a driver on a focal species, while indirect effects are the mean effects of drivers spreading through all 3-species interactions a taxon is involved in. The total contribution to cumulative effect is the sum of the direct and indirect contributions of each driver. Click on figure to view in seperate tab.

Figure 7.9: Mean contribution of 17 drivers to the cumulative effects (\(C / km^2\)) of 205 taxa on the Scotian Shelf Bioregion in eastern Canada. Taxonomic groups for invertebrates (grey box) and vertebrates are classified at the phyla and classes level, respectively, for ease of representation and interpretation. The mean contribution for each taxon was used to evaluate each driver’s contribution to overall cumulative effects; this can be divided into their direct and indirect contributions to the cumulative effects on each taxon. Direct effects are those attributable to the effect of a driver on a focal species, while indirect effects are the mean effects of drivers spreading through all 3-species interactions a taxon is involved in. The total contribution of each driver to cumulative effect is the sum of their direct and indirect contributions to cumulative effects. Click on figure to view in seperate tab.

Figure 7.10: Cumulative effects assessment of global changes on bolocera sea anemones on the Scotian Shelf Bioregion showcasing the continuous (upper left) and binary (upper right) distributions, the results of the species-scale (middle) and network-scale (lower) cumulative effects assessment for both periods Click on figure to view in seperate tab.

Figure 7.11: Cumulative effects assessment of global changes on hermit crabs (Pagurus sp.) on the Scotian Shelf Bioregion showcasing the continuous (upper left) and binary (upper right) distributions, the results of the species-scale (middle) and network-scale (lower) cumulative effects assessment for both periods Click on figure to view in seperate tab.

Figure 7.12: Cumulative effects assessment of global changes on Minke whale (Balaenoptera acutorostrata) on the Scotian Shelf Bioregion showcasing the continuous (upper left) and binary (upper right) distributions, the results of the species-scale (middle) and network-scale (lower) cumulative effects assessment for both periods Click on figure to view in seperate tab.