Trawl impacts on the relative status of biotic communities of seabed sedimentary habitats in 24 regions worldwide

Abstract

Bottom trawling is widespread globally and impacts seabed habitats. However, risks from trawling remain unquantified at large scales in most regions. We address these issues by synthesizing evidence on the impacts of different trawl-gear types, seabed recovery rates, and spatial distributions of trawling intensity in a quantitative indicator of biotic status (relative amount of pretrawling biota) for sedimentary habitats, where most bottom-trawling occurs, in 24 regions worldwide. Regional average status relative to an untrawled state (=1) was high (>0.9) in 15 regions, but <0.7 in three (European) regions and only 0.25 in the Adriatic Sea. Across all regions, 66% of seabed area was not trawled (status = 1), 1.5% was depleted (status = 0), and 93% had status > 0.8. These assessments are first order, based on parameters estimated with uncertainty from meta-analyses; we recommend regional analyses to refine parameters for local specificity. Nevertheless, our results are sufficiently robust to highlight regions needing more effective management to reduce exploitation and improve stock sustainability and seabed environmental status-while also showing seabed status was high (>0.95) in regions where catches of trawled fish stocks meet accepted benchmarks for sustainable exploitation, demonstrating that environmental benefits accrue from effective fisheries management. Furthermore, regional seabed status was related to the proportional area swept by trawling, enabling preliminary predictions of regional status when only the total amount of trawling is known. This research advances seascape-scale understanding of trawl impacts in regions around the world, enables quantitative assessment of sustainability risks, and facilitates implementation of an ecosystem approach to trawl fisheries management globally.

Keywords: habitat sensitivity; recovery; spatial upscaling; trawl footprints; trawl impacts.

Fig. 1.

Maps of regional average RBS for continental shelves (0- to 200-m depth) and slopes (200- to 1,000-m) in 24 regions. Pie charts show proportional area by region in six RBS category intervals; the pie legend (Center) also indicates the average of category proportions across all regions. Black boundaries indicate study regions (i.e., exclusive economic zones or fishery management jurisdictions or large marine ecosystems). Region numbers and names follow Fig. 2.

Fig. 2.

Distributions of grid cell RBS values (ordered 1 through 0) versus cumulative percentage of regional area. Where RBS = 1 at top/left indicates untrawled seabed, and RBS = 0 at bottom/right indicates depleted seabed. The lower uncertainty interval is indicated by the band between cell mean RBS and the lower 95% CL of cell RBS. Dotted horizontal and vertical lines at RBS = 0.8 and 80% of region area indicate example thresholds. The regions legend is ordered by regional average RBS.

Fig. 3.

Bar plots of (A) average RBS for gravel, sand, and mud (G, S, and M; colored bars) habitats within regions, and reduction of RBS (=1 − RBS) because of cumulative impacts of different trawl gear types (stacked gray bars); (B) percentage area of each regional habitat in six RBS category intervals. Vertical dotted line indicates RBS = 0.8 in A and 80% of regional habitat area in B.

Fig. 4.

Relationships for regional average RBS versus (A) regional SAR for all 24 regions, fitted relationship and prediction interval, with fitted relationships for sedimentary habitats and continental shelves and slopes; vertical dotted line indicates SAR = 0.25 (see text); and (B) stock exploitation as the ratio of f over f_MSY reference point for individual trawl fishery stocks in 12 regions for years 2010, 2011, and 2012 (9) and regional average f/f_MSY; green vertical dotted line at f/f_MSY = 1 indicates an accepted sustainable upper limit on fishing rate; light-green shading emphasizes data for regions where most stocks are managed sustainably (f/f_MSY < 1) and average RBS ≥ 0.95; linear fit to all 87 stocks in 12 regions: slope = −0.101, R² = 0.71, P < 0.001; linear fit to 12 regional means: slope = −0.131, R²= 0.91, P < 0.001.