Groundfish biodiversity change in northeastern Pacific waters under projected warming and deoxygenation


credit: Northeast Seamounts Expedition Partners

Supplementary Material


Patrick L. Thompson, Jessica Nephin, Sarah C. Davies, Ashley Park, Devin A. Lyons, Chris Rooper, M. Angelica Peña, James Christian, Karen Hunter, Emily Rubidge, Amber M. Holdsworth


This Shiny app is an interactive version of the single species supplemental figures. It allows users to visualize both environmental response curves and distribution changes for groundfish species in northeastern Pacific waters under projected warming and deoxygenation.

Abstract

Projections of how climate change will impact marine species and communities are urgently needed to inform management measures aimed at stemming biodiversity loss. In the coming decades, warming and deoxygenation of marine waters are anticipated to result in shifts in the distribution and abundance of fishes, with consequences for the diversity and composition of fish communities. Most projections to date have focused on temperature, but have not accounted for the confounding influence of oxygen and depth and are limited by the spatial resolution of global climate models. Here, we combine fisheries independent trawl survey data spanning the west coast of the USA and Canada with high resolution regional ocean models to make projections of how 40 groundfish species will be impacted by changes in temperature and oxygen in British Columbia (B.C.) and Washington. By leveraging coast-wide survey data, we quantify how temperature, oxygen, and depth jointly constrain the ranges of species. Then, using two high-resolution regional ocean-biogeochemical models, we make projections of biodiversity change at a high spatial resolution. Our projections suggest that, in B.C. and Washington, the number of species that are projected to decrease in occurrence is roughly balanced by the number that are projected to increase, resulting in considerable compositional turnover. Many, but not all, species are projected to shift to deeper depths as conditions warm, but low oxygen will limit how deep they can go. Thus biodiversity will likely decrease in the shallowest waters (< 100 m) where warming will be greatest, increase at mid depths (100—600 m) as shallow species shift deeper, and remain stable or decrease at depths where oxygen is limited (> 600 m). These results highlight the critical importance of accounting for the joint role of temperature, oxygen, and depth when projecting the impacts of climate change on marine biodiversity.

References

Holdsworth, A.M., Zhai, L., Lu, Y. & Christian, J.R. (2021). Future Changes in Oceanography and Biogeochemistry Along the Canadian Pacific Continental Margin. Frontiers in Marine Science, 8, 602991.Retrieved August 27, 2021, from https://www.frontiersin.org/articles/10.3389/fmars.2021.602991/full
Peña, M.A., Fine, I. & Callendar, W. (2019). Interannual variability in primary production and shelf-offshore transport of nutrients along the northeast Pacific Ocean margin. Deep Sea Research Part II: TopicalStudies in Oceanography, 169–170, 104637. Retrieved July 14, 2020, from http://www.sciencedirect.com/science/article/pii/S0967064519300220

Citation

Thompson, P.L., Nephin, J., Davies, S.C., Park, A.E., Lyons, D.A., Rooper, C.N., Peña, M.A., Christian, J.R., Hunter, K.L., Rubidge, E. and Holdsworth, A.M. 2022. Groundfish biodiversity change in northeastern Pacific waters under projected warming and deoxygenation. In review.

Data Availability

Data from the trawl surveys were obtained from gfdata package

Data are openly available:
DFO Groundfish Synoptic Bottom Trawl Survey data
NOAA U.S. West Coast Groundfish Bottom Trawl Survey data
NOAA Alaska Groundfish Bottom Trawl Surveys
Regional Ocean Model Data Monthly averages for the NEP36 model
Historical climatologies from the BCCM

Observed and Projected Response Curves



Panels a–d show modelled (contour lines) and observed (coloured points) occurrences based on seafloor depth and temperature (a,c) or temperature and dissolved oxygen (b, d). Panels a and b show observed data from the entire coast, panels c and d show the subset of this data from the focal region where future projections were made. The coloured circles represent gridded values across this parameter space, with the size of each circle showing the number of corresponding trawls, and the colour showing the proportion of trawls where the species was present. The SDM modelled responses are shown as contour lines over this environmental space, with each line corresponding to a given probability of occurrence as indicated by its colour. These SDM modelled responses were fit on the entire coastwide dataset (a–b), but we include these contour lines in panels c and d to show how these SDM modelled fits correspond to the environmental space and observations in the focal region. The model response curves for seafloor depth and temperature are produced by making predictions using the fitted SDM model across a regular grid (30 x 30) of environmental conditions from the minimum to the maximum depth (log) and temperature values in the coastwide dataset, excluding combinations that fell outside of the observed environmental space. Dissolved oxygen for each depth, temperature combination was obtained by binning trawls within this grid and calculating the median value. The model response curves for seafloor temperature and oxygen (b, d) were produced in the same way, but with depth instead of oxygen estimated from the binned data.

Spatial Comparison of Different Models



Panels e–h shows projected changes in occurrence between 1986–2005 and 2046–2065. These are based on applying the fitted SDM model to seafloor depth, temperature, and dissolved oxygen at a 3 km resolution from a hindcast (1986–2005) and future projection (2046–2065) of the BCCM (e, g; Peña et al. 2019) and the NEP36 model (f, h; Holdsworth et al. 2021). Panels e and f show projections based on the RCP 4.5 emissions scenario, panels g and h show projections based on the RCP 8.5 emissions scenario. The map shows the difference between the hindcast and future projected mean occurrences, with blue shades indicating increased occurrence and red and orange shades indicating decreases in occurrence. We assume that areas with an occurrence probability of 0.1 or greater are suitable habitat. Grid cells where both the hindcast and the future projection fall below this threshold are coloured grey. The inset histogram serves as a colour legend and shows the distribution of projected occurrence change values across all 3 km grid cells in the focal region.