Estimates of entanglement risk for lobster fishing off the coast of Maine Chris Brehme, Keene State College Hauke Kite-Powell, Woods Hole Oceanographic

Embed Size (px)

Citation preview

  • Slide 1
  • Estimates of entanglement risk for lobster fishing off the coast of Maine Chris Brehme, Keene State College Hauke Kite-Powell, Woods Hole Oceanographic Institution Scott Kraus, Kerry Lagueux, and Brooke Wikgren, New England Aquarium Patrice McCarron and Heather Tetreault, Maine Lobstermens Association 2013 Northeast Regional Sea Grant Meeting New Bedford, 19 November 2013
  • Slide 2
  • Approach: Estimate the expected number of whale/fishing gear encounters per year this will depend on: fishing effort vertical line density trap string configuration whale activity density behavior (transiting, feeding, etc.) topography water depth bottom characteristics Consider adjustments to fishing effort (time, location, gear configuration) to reduce risk.
  • Slide 3
  • Vertical Line Risk Probability of whale- line encounter = f (lines/km 2 whale track/km 2 [whale size] )
  • Slide 4
  • Ground Line Risk Probability of whale-line encounter = f (h line length/km 2 whale track/km 2 [whale size] water depth whale diving )
  • Slide 5
  • Model Polygons and Fishing Zones
  • Slide 6
  • Fishing Activity Data 5.9 million VL-months Exempt state: 70.6% Non-ex. state:22.7% 3-6 miles: 3.2% 6-12 miles: 1.4% 12-40 miles: 2.1%
  • Slide 7
  • Fishing Gear in the Water
  • Slide 8
  • Whale Activity Data
  • Slide 9
  • SPUE from on-effort survey only
  • Slide 10
  • Modeled Whale Activity Interpolation approach based on: E. Bellier and Monestiez P., 2008. Model based block kriging to estimate spatial abundance of wildlife populations from count data. In Proceedings of the Eighth International Geostatistics Congress. Julian Ortiz & Xavier Emery (Eds.). GECAMIN. pp 119-128. ISBN:978-956-8504-18-2.
  • Slide 11
  • Whale Activity
  • Slide 12
  • Whale Activity by Zone FTWEs: Dec, Jan:52.5 Feb, Mar: 3.6 Apr, May: 5.7 Jun, Jul: 9.0 Aug, Sep:16.2 Oct, Nov:28.6 511,500 km whale track
  • Slide 13
  • Whale Activity Inshore Whale track distribution: State exempt: 0.3% State non-ex.: 0.8% 3-6 miles: 1.7% 6-12 miles: 5.7% 12-40 miles: 91.5%
  • Slide 14
  • Model Input Summary
  • Slide 15
  • Baseline: Expected Encounters 2011 expected encounters/year: VL 2,163 GL 342 total 2,505
  • Slide 16
  • Baseline: Expected Encounters
  • Slide 17
  • Baseline Risk by Zone
  • Slide 18
  • Sensitivity Analysis Both fishing effort and whale activity data sets incorporate assumptions Fishing effort: total number of active traps, and allocation of traps to areas designated by fishermen as rarely/sometimes/always fished Whale activity: whale activity level in nearshore waters where there is little systematic survey work Question: do risk hot spots predicted by the model depend on these assumptions?
  • Slide 19
  • Sensitivity Scenarios: Fishing Effort Five vertical line models: 5/35/60 allocation of traps to rarely/sometimes/often fished areas (base case) 0.5/9.5/90 shifts more traps to often fished 25/35/40 shifts traps to rarely/sometimes fished 5/ +1 conf is 5/35/60 with more traps active 5/ -1 conf is 5/35/60 with fewer traps active Moderate uncertainty; +/- 35% (median)
  • Slide 20
  • Sensitivity Scenarios: Whale Activity Three whale activity models: Predicted WA: assumes some degree of whale activity (0.5 x standard error) in all polygons with at least one sighting (base case) Low WA: Predicted WA minus 2 x standard error (truncated at zero); represents lower than predicted nearshore whale activity High WA: Predicted WA plus 2 x standard error; represents higher than predicted nearshore whale activity High uncertainty, esp. inshore; +/- 90% (median)
  • Slide 21
  • Sensitivity Analysis: Base case hot spots Base case risk hot spots (63%): A-3 (Aug, Sep) 32% A-40 (Jun Nov) 8% C-40 (Oct Jan) 9% D-3 (Jun Sep) 5% F-40 (Nov Jan) 4% G-40 (Dec, Jan) 5% Share of risk in these hot spots is insensitive to fishing effort assumptions and Predicted/Low WA But decreases to 30% with High WA
  • Slide 22
  • Sensitivity Analysis continued Base case has 50-60% of risk in state waters (more than half of this in A3 Aug, Sep), and 30% in 12-40 nm. With half of base case near shore whale activity, state waters risk drops to less than 20%, and 12-40 nm risk rises to 60%. (Plausible.) With 10% of whale activity in state waters, state waters risk rises to 90%. (Conceivable but not likely.)
  • Slide 23
  • Model Application Examples (Note: Area 1 Maine only) Effect of Sinking Ground Line (rule in place) NMFS NE Proposal (2011) Offshore Trawl-up (12-40nm)
  • Slide 24
  • Sinking Ground Line (rule in place) 347 avoided encounters/year; 12% reduction in total pre-rule encounter risk Risk would rise 14% if all zones went back to floating ground line, mainly from: A non-exempt (Aug, Sep) A40 (Aug, Sep) C non-exempt (Jun, Jul) C40 (Dec, Jan) D40 (Dec, Jan) E40 (Nov Jan) F40 (Nov Jan) G non-exempt (Oct, Nov) G40 (Nov-Jan)
  • Slide 25
  • NMFS NE Proposal (2011)
  • Slide 26
  • Results: NMFS NE Proposal (2011) expected encounters/year: VL 1,987 (-167) GL 342 total 2,329 7% reduction from baseline ~80% of this from: C40 (Dec, Jan) D non-ex (Jun Sep) F40 (Nov Jan) G40 (Dec, Jan)
  • Slide 27
  • Offshore Trawl-ups Most significant months for 12-40 trawl up risk reduction: Zone A: Aug, Sep, Oct, Nov Zone B: Oct, Nov Zone C: Oct, Nov, Dec, Jan Zone D: Dec, Jan Zone E: Dec, Jan Zone F: Nov, Dec, Jan Zone G: Dec, Jan Risk reduction: from 12% (min 20) to 25% (min 30) VL-month reduction: from 46,000 (min 20) to 72,000 (min 30) DMR2012: 70,000 NMFS NE: 456,000 (8%)
  • Slide 28
  • Summary Highest resolution fishing effort & more complete representation of Right Whale activity to date Model is operational (spreadsheet implementation), can easily reflect alternate management measures and assumptions about fishing & whale activity Spatial distribution of risk is sensitive to assumptions about inshore whale activity Assuming inshore activity = 0 is a sure way to NOT address significant percentage of total risk Risk is concentrated in temporal/spatial hot spots Danger: blanket measures that miss hot spots are unlikely to reduce risk significantly (though they may reduce VLs) Opportunity: targeted management measures can reduce risk with minimal disruption of fishing (but may be harder to enforce)
  • Slide 29
  • Acknowledgements Collaborators: Chris Brehme, Keene State College Tara Hetz, WHOI Summer Student Fellow Scott Kraus, New England Aquarium Kerry Lagueux, New England Aquarium Cris Lutazzi, WHOI Summer Student Fellow Patrice McCarron, Maine Lobstermens Association Heather Tetreault, Maine Lobstermens Association Sophia Weinman, WHOI Guest Student Brooke Wikgren, New England Aquarium Funding provided by the WHOI and Maine Sea Grant Programs and: