The Gear Seabed Impact Model A tool for assessing the potential adverse effects of fishing on Essential Fish Habitat in the Northeast region

Photo Credit: SMAST Video Survey ‐ University of Massachusetts – Dartmouth, Dept of Oceanography

Drafted by the New England Management Council’s Habitat Plan Development Team

Updated February 27, 2009

The New England Fishery Management Council’s Habitat Plan Development Team:

Ms. Jennifer Anderson, NOAA NMFS NERO Dr. Peter Auster, Univ. of CT Ms. Michelle Bachman, NEFMC Ms. Patricia Clay, NOAA NMFS NEFSC Mr. Chad Demarest, NEFMC (Chair) Dr. David Dow, NOAA NMFS NEFSC* Mr. Steve Eayrs, Gulf of Maine Research Institute Dr. Steve Edwards, NOAA NMFS NEFSC* Dr. Jonathan Grabowski, Gulf of Maine Research Institute Mr. Brad Harris, Univ. of MA Dartmouth, SMAST Mr. Chad Keith, NOAA NMFS NEFSC Dr. Mark Lazzari, ME Dept. of Marine Resources Mr. Vincent Malkoski, MA Division of Marine Fisheries Mr. David Packer, NOAA NMFS NEFSC Mr. Chris Powell, RI Dept. of Fish and Wildlife* Dr. David Stevenson, NOAA NMFS NERO Dr. Page Valentine, USGS

*no longer assigned to the PDT

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 2 DRAFT April 17, 2009 1.0 Table of contents

1.0 Table of contents...... 3 2.0 Introduction...... 7 2.1.1 Summary of the Fishing Gear Seabed Impact model (FiGSI)...... 7 2.1.2 Document content...... 9 3.0 Background and purpose ...... 10 3.1.1 Regulatory context...... 10 3.1.2 Management history ...... 11 3.1.3 Purpose of this document...... 18 4.0 Evaluating the effects of fishing gears on habitats (Vulnerability Assessment)..... 19 4.1 Overview...... 19 4.2 Fishing gear descriptions ...... 19 4.2.1 Demersal otter trawls ...... 27 4.2.1.1 Groundfish trawls...... 29 4.2.1.2 trawls...... 29 4.2.1.3 Shrimp trawls ...... 29 4.2.1.4 trawls ...... 30 4.2.1.5 Raised footrope trawls...... 30 4.2.2 New Bedford‐style scallop dredges ...... 30 4.2.3 Hydraulic dredges...... 31 4.2.4 Fish and shellfish traps ...... 32 4.2.4.1 Lobster traps ...... 32 4.2.4.2 Deep‐sea red crab traps...... 32 4.2.5 Demersal longlines ...... 33 4.2.6 Sink gill nets ...... 33 4.3 Habitat component descriptions...... 34 4.3.1 Geological ...... 34 4.3.2 Biological...... 36 4.3.3 Prey ...... 38 4.4 Gear impacts literature review...... 40 4.4.1 Demersal otter trawls ...... 45 4.4.2 New Bedford‐style scallop dredges ...... 58 4.4.3 Hydraulic clam dredges...... 62 4.4.4 Lobster and deep‐sea red crab traps ...... 64 4.5 Estimating susceptibility and recovery...... 64 4.5.1 Geological ...... 66 4.5.2 Biological...... 74 4.5.3 Prey ...... 74 5.0 Estimating effective fishing effort (Swept Area Seabed Impact model)...... 76 5.1.1 Methods...... 76 5.1.1.1 Demersal otter trawl...... 77

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 3 DRAFT April 17, 2009 5.1.1.2 New Bedford‐style scallop dredge ...... 78 5.1.1.3 Hydraulic clam dredge...... 78 5.1.1.4 Lobster traps and deep‐sea red crab traps...... 79 5.1.1.5 Demersal longline and gill net ...... 79 5.1.2 Data and parameterization...... 80 5.1.3 Results ...... 85 5.1.4 Discussion ...... 85 6.0 Estimating spatially‐defined habitat and energy (Spatial model)...... 86 6.1.1 Base grid...... 86 6.1.2 Data and parameterization...... 87 6.1.3 Substrate classification and data sources ...... 87 6.1.3.1 SMAST video survey ...... 89 6.1.3.2 usSEABED database ...... 90 6.1.3.3 NMFS Trawl Survey Gear Hangs ...... 91 6.1.4 Classifying natural disturbance – Critical Shear Stress Model ...... 92 7.0 The FiGSI model: combining the Vulnerability Assessment, the SASI model and the Spatial model ...... 96 7.1 Methods...... 96 7.1.1 Estimating sensitivity (Se) coefficients ...... 96 7.2 Results...... 97 7.2.1 Geological habitat components...... 97 7.2.2 Biological habitat components...... 104 7.2.3 Prey habitat components ...... 104 7.3 Future directions ...... 104 8.0 References ...... 105 9.0 Appendices...... 113 9.1.1 Definitions...... 113

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 4 DRAFT April 17, 2009 Tables

Table 1 – Landed pounds by gear type (1,000 lbs, source: NMFS vessel trip reports) ..... 20 Table 2 – Percent of total landed pounds by gear type (source: NMFS vessel trip reports) ...... 21 Table 3 – Revenue by gear type (1,000 dollars, all values converted to 2007 dollars; source: NMFS vessel trip reports) ...... 22 Table 4 – Percent of total revenues by gear type (source: NMFS vessel trip reports)...... 23 Table 5 – Days absent by gear type (source: NMFS vessel trip reports)...... 24 Table 6 – Percent of days absent by gear type (source: NMFS vessel trip reports)...... 25 Table 7 ‐ Fishing Gears Used in Estuaries and Bays, Coastal Waters, and Offshore Waters of the EEZ, from Maine to North Carolina ...... 26 Table 8 ‐ Geological habitat component classes, subclasses, and features. Asterisk (*) indicates level used for mapping purposes ...... 35 Table 9 ‐ Bedform classification (after Twichell 1983)...... 36 Table 10 ‐ Biological habitat components...... 37 Table 11 ‐ Prey habitat components...... 39 Table 12 ‐ Literature review database fields ...... 40 Table 13 ‐ Studies reviewed by gear type(s), substrate type(s), and habitat component(s) evaluated. Habitat components are denoted as G (geological), B (biological), and P (prey). Substrate classes are denoted as M (mud), S (sand), and G (gravel). Studies are listed alphabetically...... 42 Table 14 ‐ Impacts of otter trawls to geological, biological, and prey habitat components on mud, sand, and gravel substrates...... 47 Table 15 ‐ Impacts of New Bedford‐style scallop dredges on geological, biological, and prey habitat components...... 59 Table 16 ‐ Effects of hydraulic clam dredges on geological, biological, and prey habitat components. (‘S’ indicates statistical significance.) ...... 62 Table 17 ‐ Impacts of crustacean traps on biological habitat components (“S” indicates statistical significance) ...... 64 Table 18 ‐ Generic matrix construction...... 64 Table 19 ‐ Susceptibility values...... 65 Table 20 ‐ Recovery values ...... 65 Table 21 ‐ Generic geological matrix...... 67 Table 22 – Groundfish trawl matrix with summary S, Rhigh, and Rlow values...... 69 Table 23 ‐ Raised footrope trawl matrix with summary S, Rhigh, and Rlow values...... 70 Table 24 ‐ Shrimp trawl matrix with summary S, Rhigh, and Rlow values...... 71 Table 25 ‐ Squid trawl matrix with summary S, Rhigh, and Rlow values ...... 72 Table 26 ‐ New Bedford‐style scallop dredge matrix with summary S, Rhigh, and Rlow values ...... 73 Table 27 Generic biological matrix...... 74 Table 28 ‐ Generic prey matrix...... 75 Table 29 – Tow speeds by year and fishing gear type...... 81

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 5 DRAFT April 17, 2009 Table 30 –Independent group t‐test for observer‐reported trips made between 2003‐2008 with trawl gears, and VTR‐reported trips for the same years; paired records discarded from VTR group (Class 1 = VTR, Class 2 = OBS)...... 84 Table 31 ‐ Trawl contact indices ...... 85 Table 32 – Substrate subclasses by particle size range ...... 88 Table 33 ‐ Critical shear stress model components...... 93 Table 34 ‐ Sensitivity values...... 96 Table 35 ‐ Weighted sensitivity values, groundfish trawl gear ...... 98 Table 36 ‐ Weighted sensitivity values, raised footrope trawl gear ...... 98 Table 37 ‐ Weighted sensitivity values, shrimp trawl gear ...... 99 Table 38 ‐ Weighted sensitivity values, squid trawl gear ...... 99 Table 39 ‐ Weighted sensitivity values, New Bedford‐style scallop dredge gear ...... 100

Figures

Figure 1 – FiGSI model flowchart ...... 9 Figure 2 Existing management areas and the Habitat Closure Areas established under Amendments 10 and 13 to the Atlantic sea scallop and Northeast multispecies FMPs... 17 Figure 3 ‐ Literature review database form. Data field descriptions provided in Table 12...... 41 Figure 4 – Linear regression of otter board length on otter board weight ...... 81 Figure 5 – Linear regression of otter board weight on vessel gross tonnage and vessel horsepower, observer data 2003‐2008...... 82 Figure 6 – Linear regression of ground cable length on vessel length, observer data 2003‐ 2008...... 82 Figure 7 – Linear regression of sweep length on vessel gross tonnage and horsepower, observer data 2003‐2008...... 83 Figure 8 – Voronoi diagram of geological data points used as base grid in spatial model ...... 87 Figure 9 – Base grid cell coding of dominant substrate ...... 92 Figure 10 – FVCOM domain and nodes...... 94 Figure 11 – Base grid cell coding of energy (CSS model plus depth) ...... 95 Figure 12 – Spatially defined sensitivity (Se) values for groundfish and raised footrope trawl gears...... 101 Figure 13 – Spatially defined sensitivity (Se) values for shrimp and squid trawl gears. 102 Figure 14 ‐ Spatially‐defined sensitivity (Se) values for scallop dredge gear...... 103

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 6 DRAFT April 17, 2009 2.0 Introduction The purpose of this document is to present the Fishing Gear Seabed Impact (FiGSI) model, a tool for the New England Fishery Management Council (Council) to use in assessing the adverse effects of fishing on fish habitats. The model is being developed as part of a comprehensive review of fishing impacts undertaken for the Council’s Omnibus Essential Fish Habitat Amendment 2. In the future, the model is intended for use in evaluating the impacts of fishery management actions on fish habitat.

2.1.1 Summary of the Fishing Gear Seabed Impact model (FiGSI) The FiGSI model is an adaptive tool used to evaluate the impacts of fishing gears on fish habitats in a spatial context. FiGSI has three underlying components: assessment of the vulnerability of habitats to gears, a conversion of fishing effort data to contact‐adjusted swept area, and a substrate and energy based spatial grid. These are combined to generate spatially specific estimates of contact‐ and habitat‐sensitivity‐adjusted area swept.

Fish habitats are divided into three major components: geological, biological, and prey, which are in turn further disaggregated into habitat features. The Vulnerability Assessment uses a matrix‐based evaluation to estimate quantitative susceptibility and recovery values for each feature by fishing gear type. Susceptibility is a measure of the number or amount of a habitat’s features that are reduced in functional value due to the impact of a particular fishing gear. Recovery is a measure of the amount of time it would take for the functional value of those diminished habitat features to be restored following the cessation of impact. Recovery is evaluated separately for high and low energy environments. Susceptibility and recovery values are assigned based on knowledge of the fishing gears and the habitat features, and informed by the scientific literature to the extent possible.

Fishing effort data is divided into major gear types, which include groundfish trawls, shrimp trawls, squid trawls, raised footrope trawls, New Bedford‐style scallop dredges, lobster and deep‐sea red crab traps, bottom gill nets, and bottom longlines. Using the Swept Area Seabed Impact (SASI) model, fishing effort data is represented universally as area swept, and scaled based on a particular gear’s contact with the seabed to obtain contact‐adjusted area swept. Effort data is compiled from various fishery‐dependent sources including observer, vessel trip report, and vessel monitoring system.

The Spatial Model uses three types of substrate data to generate a dominant‐seafloor‐ substrate base grid on which to combine fishing effort data with habitat vulnerability (susceptibility and recovery). The grid is composed of cells of varying sizes and shapes based on the density of substrate data available throughout the region. Grid cells are classified as high or low energy using a Critical Shear Stress Model.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 7 DRAFT April 17, 2009 Finally, the FiGSI Model combines contact‐adjusted area swept and habitat vulnerability with the spatial grid. Susceptibility and recovery scores from the Vulnerability Assessment are combined to generate a single sensitivity scalar for each feature. Features within each dominant substrate class are weighted according to their expected distributions to obtain a sensitivity score for each dominant‐substrate/gear combination. Contact‐adjusted area swept estimates are assigned to each grid cell for all gear types under evaluation. The effort data for each grid cell is then scaled by the appropriate sensitivity coefficient to obtain a sensitivity‐adjusted area swept.

These sensitivity‐adjusted area swept values can be presented by individual gear type or for a combination of gear types, for example to represent a fishery. Effort data over time may be compared, and the FiGSI model can be integrated with other spatially specified projection models to forecast changes in sensitivity‐adjusted area swept resulting from proposed regulations. Because the FiGSI is both disaggregated and quantitative, new information is easily integrated. For example, the contact coefficients for various gear components can be modified to model new gear designs, or to understand how assumptions about the contact indices affect area swept. The relative weightings of various habitat features can be altered to accommodate new habitat data. Susceptibility and recovery values can be modified given a better understanding of how various gears interact with the seabed, or if new information about recovery times becomes available in the scientific literature.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 8 DRAFT April 17, 2009 Figure 1 – FiGSI model flowchart

2.1.2 Document content The FiGSI model is a work in progress, and as such all fishing gear and habitat component combinations have not been fully analyzed. This document discusses the methods and data that will be used to assess all fishing gears and habitat components, but only presents results for the combinations analyzed to date. These combinations include the sensitivity of geological habitat components to New Bedford‐style scallop dredges, groundfish trawls, shrimp trawls, squid trawls, and raised footrope trawls.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 9 DRAFT April 17, 2009 3.0 Background and purpose

3.1.1 Regulatory context The Magnuson Fishery Conservation and Management Act of 1976, (renamed the Magnuson‐Stevens Fishery Conservation and Management Act when amended on October 11, 1996) established a U. S. exclusive economic zone (EEZ) between 3 and 200 miles offshore, and established eight regional fishery management councils that manage the living marine resources within that area. The eighteen (18) member New England Fishery Management Council’s (Council) authority extends from Maine to southern New England and, in some cases, to the mid‐Atlantic because of the range of the species covered under its management plans. The 1996 amendments to the Magnuson‐Stevens Fishery Conservation and Management Act, known as the Sustainable Fisheries Act (SFA), emphasized the importance of habitat protection to developing healthy fisheries by strengthening the ability of the National Marine Fisheries Service (NMFS) and the Councils to protect and conserve the habitat of marine, estuarine, and anadromous finfish, mollusks, and crustaceans. Such habitat is termed ʺessential fish habitatʺ and is broadly defined to include ʺthose waters and substrate necessary to fish for spawning, breeding, feeding, or growth to maturity.ʺ

Requirements of the SFA for NMFS, Councils, and Federal Agencies

To protect fish habitat, the SFA requires or authorizes the Councils, NMFS, and other federal agencies to take new actions. Relevant to the goals of Phase I of the Omnibus Amendment 2, the SFA requires the Council amend its fishery management plans to: • Describe and identify essential fish habitat (EFH) for every fishery • Minimize to the extent practicable the adverse impacts of fishing on EFH • List the major prey species for the species in the FMU and discuss their location • Identify non‐fishing activities that may adversely affect EFH

Essential Fish Habitat Guidelines (Final Rule) The National Marine Fisheries Service promulgated guidelines interpreting the EFH components of the SFA on January 17, 2002. These guidelines: • Require EFH designations for all managed species, including unique descriptions of EFH for each life‐stage for those species, and provide guidance for making such designations • Introduce the concept of habitat areas of particular concern (HAPC) • Requires Councils to review EFH documents every five years • Specifies the requirements for minimizing to the extent practicable the adverse effects from fishing on habitat, specifically:

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 10 DRAFT April 17, 2009 “Each FMP must contain an evaluation of the potential adverse effects of fishing on EFH designated under the FMP, including effects of each fishing activity regulated under the FMP or other Federal FMPs. This evaluation should consider the effects of each fishing activity on each type of habitat found within EFH. FMPs must describe each fishing activity, review and discuss all available relevant information (such as information regarding the intensity, extent, and frequency of any adverse effect on EFH; the type of habitat within EFH that may be affected adversely; and the habitat functions that may be disturbed), and provide conclusions regarding whether and how each fishing activity adversely affects EFH. The evaluation should also consider the cumulative effects of multiple fishing activities on EFH. The evaluation should list any past management actions that minimize potential adverse effects on EFH and describe the benefits of those actions to EFH.”

The Final Rule further specifies that: “Councils must act to prevent, mitigate, or minimize any adverse effects from fishing, to the extent practicable, if there is evidence that a fishing activity adversely affects EFH in a manner that is more than minimal and not temporary in nature. In such cases, FMPs should identify a range of potential new actions that could be taken to address adverse effects on EFH, include an analysis of the practicability of potential new actions, and adopt any new measures that are necessary and practicable. Amendments to the FMP or to its implementing regulations must ensure that the FMP continues to minimize to the extent practicable adverse effects on EFH caused by fishing. FMPs must explain the reasons for the Council’s conclusions regarding the past and/or new actions that minimize to the extent.”

3.1.2 Management history Omnibus Habitat Amendment The Omnibus EFH Amendment #1 was prepared in 1998 to identify and describe the EFH for all species of marine, estuarine, anadromous finfish and mollusks managed by the Council to better protect, conserve, and enhance this habitat. This was done through the following FMP amendments: Northeast Multispecies (11), Atlantic Sea (9), Atlantic Salmon (1), and Atlantic Herring (added to FMP later). The 1998 EFH Amendment also identified the major threats to EFH from both fishing and non‐fishing related activities and conservation and enhancement measures. The Council began implementation of the SFA’s EFH requirements based on guidance provided by NMFS on interpreting the mandate and timelines. Amendments to the FMPs managed by the Council were initiated in 1998 and combined in one management action that was termed the “Habitat Omnibus Amendment of 1998.” The Council approved the final EFH FMP amendments (EA) in September 1998 and the EA was submitted to NMFS in October 1998. The Secretary of Commerce approved the amendments to all FMPs, with the exception of the Monkfish FMP, on March 1999. The EFH requirements of FMPs that were not included in the Omnibus Amendment of 1998 were completed on the

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 11 DRAFT April 17, 2009 following schedule: Monkfish FMP (April 1999), Red Crab FMP (October 2002), and Skate FMP (July 2003).

AOC v. Daley lawsuit A lawsuit brought by several environmental organizations (American Campaign (AOC) et al. v. Daley et al.) resulted in a ruling in 2000 that prevented the Department of Commerce (DOC) from enforcing the EFH amendments challenged in the suit, which included amendments to all of the New England Council’s fishery management plans amended under the Omnibus Habitat Amendment. The Council was required to perform “a new and thorough EA or EIS” for each of the EFH amendments, in compliance with NEPA. Specifically, the DOC agreed to instruct the Councils to: 1) Prepare EISs for all fisheries challenged in the lawsuit. 2) Comply with the requirements of all applicable statues, including NEPA; the Council on Environmental Quality (CEQ) NEPA implementing regulations, 40 C.F.R. Parts 1500‐1508; and the National Oceanic and Atmospheric Administration (NOAA) Administrative Order 216‐6. 3) Include analyses of environmental impacts of fishing on EFH, including direct and indirect effects, as defined in the EFH regulations at 50 C.F.R. 600.810, and analyses of the environmental impacts of alternatives for implementing the requirement of the M‐S Act, that the FMP “minimize, to the extent practicable, adverse effects on [EFH] caused by fishing.” 4) Consider a range of reasonable alternatives for minimizing the adverse effects (as defined by the EFH regulations) of fishing on EFH, including potential adverse effects. This range of alternatives will include “no action” or status quo alternatives and alternatives set forth specifying fishery management actions that can be taken by NMFS under the M‐S Act. The alternatives may include a suite of fishery management measures, and the same fishery management measures may appear in more than one alternative. 5) Identify one preferred alternative, except that, in the draft EIS, NMFS may elect, if it deems appropriate, to designate a subset of the alternatives considered in the draft EIS, as the preferred range of alternatives, instead of designating only one preferred alternative. 6) Present the environmental impacts of the alternatives in comparative form, thus sharply defining the issues and providing a clear basis for choice among the options, as set forth in CEQ regulation 40 C.F.R. 1502.14.

In response to the Stipulation, the Council determined that the analysis and subsequent management alternatives required by the Court Order would be presented within separate NEPA documents currently being developed by NMFS and the Council for the Northeast Multispecies and Atlantic Sea Scallops Fishery Management Plans. These documents were completed and submitted in 2004, and included extensive analysis of the adverse effects from fishing on essential fish habitat and a range of alternatives to address such effects.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 12 DRAFT April 17, 2009

Amendment 13 to the Northeast Multispecies FMP and Amendment 10 the Atlantic Sea Scallop FMP These two amendments included descriptions of fishing gears used in the New England Region, descriptions of existing habitats, and summaries of the existing knowledge on the affects of fishing gears on habitats. Both documents included a gear effects evaluation to assess the vulnerability of each Council‐managed species and life stage’s EFH to mobile bottom‐tending gear.

A simple matrix was developed for each benthic life stage for each species to determine the vulnerability of its EFH to effects from bottom tending mobile gear. Six criteria were qualitatively evaluated for each life stage based upon existing information. Each evaluation consisted of a value based upon a predefined threshold. The first three criteria were related to habitat function and included shelter, food and reproduction. Values for these criteria were determined as follows:

Shelter: (Scored from 0‐2) If the life stage had no dependence upon bottom habitat to provide shelter then a 0 was selected. If the life stage had some dependence upon unstructured or non‐complex habitat for shelter it was scored a 1. For example, flatfishes that rely primarily on cryptic coloration for predator avoidance or small scale sand waves for refuge were scored a 1. If the life stage had a strong reliance on complex habitats for shelter it was scored a 2. For example, species such as juvenile cod and haddock that are heavily reliant on structure or complex habitat for predator avoidance were scored a 2.

Food: (Scored from 0‐2) If the life stage had no dependence on benthic prey it was scored a 0. If the life stage utilized benthic prey for part of its diet but not exclusively a benthic feeder it was scored a 1. For example, species feeding opportunistically on crabs as well as squid or fish were scored a 1. If the life stage feeds exclusively on benthic organisms and cannot change its mode of feeding it was scored a 2.

Reproduction: (Scored from 0‐1) If the species had no dependence upon bottom habitats for spawning or its life stage was not a reproductive stage it was scored a 0. If the species had some dependence upon bottom habitats for spawning it was scored a 1. For example, species that spawn on or over the bottom were scored a 1.

Habitat Sensitivity: (Scored from 0‐2) This criterion looked at EFH‐based relative habitat sensitivity to disturbances. The habitat needed by the species was based primarily upon its EFH designation. If a habitat was not considered sensitive to disturbance it was scored a 0. If the habitat was considered to have a low sensitivity it was scored a 1. If the habitat type was considered highly sensitive it was scored a 2. These values were based upon the existing conceptual models that show a direct

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 13 DRAFT April 17, 2009 relationship between structural complexity of the habitat and recovery time with increasing vulnerability.

Habitat Rank: The habitat rank was determined quantitatively as the sum of the previous values (shelter + food + reproduction + habitat sensitivity). Another way to characterize the habitat rank is the relative vulnerability of the habitat to non‐natural physical disturbance. The rank could range from 0‐7, with 7 being the most vulnerable.

Gear Distribution: (Scored from 0‐2) This criterion factored in the use of a particular gear type (otter trawl, scallop dredge, hydraulic clam dredge) in EFH for a particular life stage. If the gear is not used in the described EFH it was scored a 0. If the gear operated in only a small portion of the described EFH it was scored a 1. If the gear operated in more than a small amount of the described EFH it was scored a 2. Distribution was determined as the qualitative overlap of EFH on the Vessel Trip Report location data which has been described in previous sections of this report.

Gear Rank: The gear rank provides the vulnerability of EFH to a particular gear type and was calculated as the product of the Habitat Rank x Gear Distribution. Based upon natural breaks in the ranking frequency distribution, the following interpretations of the ranking were made: 0 = no vulnerability to the gear; 1 ‐ 6 = low vulnerability to the gear; 7 ‐ 9 = moderate vulnerability to the gear; 10 ‐ 14 = high vulnerability to the gear.

Based upon this species‐by‐species matrix, the Council determined that:

Otter Trawls The use of Otter Trawls may have an adverse effect on the following species (and life stages) EFH as designated in Amendment 11 to the Northeast Multispecies FMP (1998): American plaice (J, A), Atlantic cod (J, A), Atlantic halibut (J, A), Atlantic sea scallops (J), haddock (J, A), pout (E, L, J, A), red hake (J, A), redfish (J, A), white hake (J), silver hake (J), winter flounder (A), witch flounder (J, A), yellowtail flounder (J, A), red crab (J, A), black sea bass (J, A), scup (J), tilefish (J, A), barndoor skate* (J, A), clearnose skate* (J, A), little skate* (J, A), rosette skate* (J, A), smooth skate* (J, A), thorny skate* (J, A), and winter skate* (J, A).

Scallop Dredge (New Bedford style) The use of New Bedford style scallop dredges may have an adverse effect on the following species (and life stages) EFH as designated in Amendment 11 to the Northeast Multispecies FMP (1998): American plaice (J, A), Atlantic cod (J, A), Atlantic halibut (J, A), Atlantic sea scallops (J), haddock (J, A), ocean pout (E, L, J, A), red hake (J, A), redfish (J, A), white hake (J), silver hake (J), winter flounder (J, A), yellowtail flounder (J, A), black sea bass , (J, A), scup (J), barndoor skate* (J, A), clearnose skate* (J, A), little skate* (J, A), rosette skate* (J, A), smooth skate* (J, A), thorny skate* (J, A), and winter skate* (J, A).

Hydraulic Clam Dredges

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 14 DRAFT April 17, 2009 The use of Hydraulic clam dredges may have an adverse effect on the following species (and life stages) EFH as designated in Amendment 11 to the Northeast Multispecies FMP (1998): Atlantic sea scallops (J), ocean pout (E, L, J, A), red hake (J), silver hake (J), winter flounder (A), yellowtail flounder (J, A), black sea bass (J, A), scup (J),clearnose skate* (J, A), little skate* (J, A), rosette skate* (J, A), and winter skate* (J, A).

(Notes: * =, E = eggs lifestage, L = larvae lifestage, J = juvenile lifestage, and A = adult lifestage).

Building on these conclusions, the documents proposed and evaluated a suite of measures designed to minimize the adverse effects of fishing on EFH. Specifically, they included the following management options:

Incidental benefits of other Amendment 10 and 13 measures: Because management measures that were designed to reduce fishing mortality may also provide benefits to fish habitat, such management measures were explicitly considered as part of a formal strategy to reduce impacts on habitat.

Modification of current groundfish closed areas to protect habitat: Modifications to the boundaries of the existing closed areas were proposed to better protect sensitive habitat. Some entirely new closed areas were proposed.

Identification of important habitat areas within current groundfish closures: Areas within currently existing closed area containing important habitat were identified. Such areas may be subject to more severe restrictions in order protect the habitat.

Closed areas designed to protect habitat and minimize impact on fisheries: This alternative was proposed to close areas with important habitat elements that are of low value to the multispecies, scallop, and monkfish fisheries in terms of productivity.

Current closed areas, with the exception of scallop access areas: The then‐current year round closed areas were considered for designation as habitat closures, with the exception of portions of those areas that have been made accessible to the scallop fishery through time‐limited openings.

Expand List of prohibited gears in closed areas: This alternative would have expanded the number of types of fishing gears that may not be used in the closed areas to include shrimp trawls, herring mid‐water trawls, clam dredges, and pots and traps.

Restrictions on the use of rockhopper and roller gear: This alternative was proposed to restrict the use of rockhopper and roller trawl gear. Various alternatives with respect to the maximum size of the gear allowed were evaluated.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 15 DRAFT April 17, 2009

To assess the impacts of management alternatives on fish habitats, Amendments 10 (Sea Scallop FMP) and 13 (Multispecies FMP) used a suite of different metrics including: • Days at Sea use • Days absent, as reported in the Vessel Trip Reports (VTRs) • % of overlap with areas designated EFH • Biomass inside/outside area closure alternatives for five trophic guilds and five spatio‐temporal species assemblages • Biomass inside/outside area closure alternative for six species with high levels of association with benthic habitats: longhorn sculpin, sea raven, redfish, ocean pout, jonah crab and American lobster • Sediment composition inside/outside area closure alternatives based on the Poppe et. Al. (1989) dataset

Alternatives were ranked based primarily on various methods of summing the raw values provided by these metrics. Ultimately, Amendment 13 to the Northeast Multispecies FMP adopted the following measures to minimize the adverse effects of fishing on EFH to the extent practicable: 1) Effort reductions, by significantly reducing DAS reductions and including seasonal closures 2) Area closure, by designating new areas both inside and outside then‐existing year‐round closures as “habitat closure areas” to reduce the effect of fishing on benthic habitats

Amendment 10 to the Atlantic Sea Scallop FMP adopted the following measures: 1) Effort reductions, by significantly reducing DAS reductions and including seasonal closures 2) Area closure, by designating new areas both inside and outside then‐existing year‐round closures as “habitat closure areas” to reduce the effect of fishing on benthic habitats 3) Gear modifications that increased dredge ring size to 4” throughout fishery, which were shown through analysis to be more efficient than 3.5” rings and therefore minimized bottom contact time 4) Mandated a portion of the TAC set‐aside for habitat research

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 16 DRAFT April 17, 2009 Figure 2 Existing management areas and the Habitat Closure Areas established under Amendments 10 and 13 to the Atlantic sea scallop and Northeast multispecies FMPs

Omnibus Habitat Amendment 2 The New England Fishery Management Council (Council) initiated the development of a second Omnibus Essential Fish Habitat (EFH) Amendment in 2005. This process is ongoing. Like the first Omnibus Habitat Amendment, the action will amend all of the fishery management plans (FMPs) managed by the Council and will become Amendment 14 to the Northeast Multispecies FMP, Amendment 14 to the Atlantic Sea Scallop FMP, Amendment 4 to the Monkfish FMP, Amendment 3 to the Herring FMP, Amendment 2 to the Skate FMP, Amendment 2 to the Red Crab FMP and Amendment 3 to the Atlantic Salmon FMP.

The purpose of Omnibus Amendment #2 is to address measures necessary to meet NMFS’ published guidelines for implementation of the Magnuson‐Stevens Act’s EFH provisions to review and revise EFH components of FMPs at least once every five (5) years; and to develop a comprehensive EFH management plan that will successfully

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 17 DRAFT April 17, 2009 minimize adverse effects from fishing on EFH through actions that will apply to all Council‐managed FMPs.

The Council has expressed dissatisfaction with its current practice of evaluating EFH and management measures to minimize to the extent practicable adverse effects from fishing on EFH through individual plans, believing instead that it is preferable to meet the EFH requirements by developing a comprehensive EFH Omnibus Amendment for all its FMPs.

The Goals for the Omnibus Habitat Amendment 2 are as follows: 1) Update the identification and description all EFH for those species of finfish and mollusks managed by the Council 2) Identify all major threats (fishing and non‐fishing) to the EFH of those species managed by the Council 3) Identify and implement mechanisms to protect, conserve, and enhance the EFH of those species managed by the Council to the extent practicable. 4) Defining the measurable thresholds for achieving the requirements to minimize adverse impacts to the extent practicable 5) Integrate and optimize measures to minimize the adverse impacts to EFH across all Council managed FMPs 6) Update research and information needs, including consideration of dedicated habitat research areas. 7) Review and update prey species information as required

The italicized items from these Goals were completed as part of Phase I of the Omnibus Amendment 2, which designated EFH for all fishery management unit (FMU) species and life stages, designated habitat areas of particular concern (HAPCs), identified and updated major non‐fishing activities that may adversely affect EFH, and reviewed and updated prey species information for each species in the FMU. Phase I of the Omnibus Amendment 2 was published as a Draft Environmental Impact Statement in July of 2007.

3.1.3 Purpose of this document This document describes the tool the Council is developing to address these Goals for the Omnibus Amendment 2. Specifically, the FiGSI model represents an objective tool designed to assist the Council in meeting Goals 2, 3, 4, and 5 (above).

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 18 DRAFT April 17, 2009 4.0 Evaluating the effects of fishing gears on habitats (Vulnerability Assessment)

4.1 Overview This section describes the Vulnerability Assessment (VA) used to assess the effects of fishing gears on fish habitats. The VA is comprised of four distinct steps. First, fishing gears used in New England are identified and described. Second, fish habitats are disaggregated into three components (geological, biological, and prey), which are then further subdivided into habitat features. Third, the gear impacts literature is reviewed for relevant work, and these studies are evaluated to determine the relevance of each study to the identified habitat features. The results of this review and summaries of each study are compiled. Fourth, for each fishing gear type, habitat feature, and energy level, susceptibility and recovery values are estimated using information from the literature combined with an understanding of the functional value of the habitat features and the effect of the fishing gear impact. These susceptibility and recovery values are then combined to produce one sensitivity score. The sensitivity values are used in the FiGSI model to modify the contact‐adjusted area swept values produced by the SASI model (discussed later in this document).

4.2 Fishing gear descriptions Forty‐five categories of fishing gear were identified as having been associated with landings of federal or state managed species based on a review of the National Marine Fisheries Service commercial fisheries landings data (Table 1 through Table 6). Although certain gear types are not managed under the auspices of the MSA, some gears utilized in state waters may have adverse impacts to EFH that is designated in nearshore, estuarine and riverine areas. Table 7 provides a list of 60 gears used in both state and federal fisheries, and indicates whether those gears are utilized in estuaries, coastal waters (0‐3 miles), or offshore waters (3‐200 miles). Since the seabed is the location of the habitat types most susceptible to gear disturbances, this table also indicates whether the gear contacts the bottom and if the use of the gear is regulated under a federal FMP.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 19 DRAFT April 17, 2009 Table 1 – Landed pounds by gear type (1,000 lbs, source: NMFS vessel trip reports) GEARNM 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 CARRIER VESSEL 00000000000069 CASTNET 00005101514247960933 DIVING GEAR 443 259 245 181 132 132 82 34 23 12 1 3 1 DREDGE, SCALLOP‐CHAIN MAT000000000371513,9813,529 DREDGE, URCHIN 152 192 206 246 185 151 103 71 72 191 117 25 145 DREDGE, 383352172710000602365706 DREDGE,OCEAN QUAHOG/SURF CLAM 6,377 619 4,704 686 1,845 1,580 1,183 538 1,066 1,079 979 862 533 DREDGE,OTHER 373 438 341 486 468 593 350 370 395 321 148 263 243 DREDGE,SCALLOP,SEA 19,180 18,303 16,985 25,245 31,935 45,529 50,169 54,404 62,008 54,664 53,257 55,352 43,766 FYKE NET 00000000361210 GILL NET,DRIFT,LARGE MESH 86 84 83 66 125 21 25 380 593 904 888 1,290 922 GILL NET,DRIFT,SMALL MESH 409 535 1,018 874 1,352 1,396 1,228 464 604 354 175 357 148 GILL NET,RUNAROUND 161 79 565 448 635 508 538 855 642 685 666 362 354 GILL NET,SINK 50,253 47,034 50,396 44,430 39,060 37,950 37,109 41,421 37,067 32,726 25,083 99,100 38,104 HAND LINE/ROD & REEL 2,353 2,071 2,645 2,337 2,561 3,622 2,935 2,177 1,939 1,402 953 1,441 893 HAND RAKE 0 0 0 0 20 4 0 184 55 115 146 150 70 HARPOON 119 71 93 102 250 107 50 53 15 8 7 6 8 HAUL SEINE 00000010720020 LONGLINE, PELAGIC 430 537 395 130 210 209 241 191 339 87 23 135 100 LONGLINE,BOTTOM 9,245 10,081 9,481 9,626 7,197 6,522 4,267 3,366 4,782 4,326 2,648 3,174 2,768 MIXED GEAR 624487608815500000000 OTHER GEAR 8,2967,2051,9142309563351110140 OTTER TRAWL, BEAM 1 0 2 7 40 144 523 529 1,182 776 269 640 477 OTTER TRAWL,BOTTOM,FISH 235,333 229,592 250,298 220,968 215,631 225,020 200,721 198,906 247,918 196,598 161,113 166,036 164,161 OTTER TRAWL,BOTTOM,OTHER3237908284386342700000032 OTTER TRAWL,BOTTOM,SCALLOP 1,395 935 2,063 2,060 2,395 3,547 3,660 3,367 3,072 1,854 956 1,345 1,039 OTTER TRAWL,BOTTOM,SHRIMP 18,159 15,212 9,162 6,140 9,104 4,447 3,261 3,142 5,080 4,347 4,300 9,820 10,576 OTTER TRAWL,MIDWATER 122,712 107,547 107,606 92,927 93,445 101,565 74,885 67,292 56,550 58,375 56,250 32,207 13,145 PAIR TRAWL,BOTTOM 43 81 127 374 45 49 113 0 9 711 18 0 240 PAIR TRAWL,MIDWATER 1,942 18,231 37,783 45,639 83,675 139,422 136,552 193,334 217,663 199,218 188,610 118,141 145,731 POT, / 464 504 841 1,191 1,817 1,850 1,834 2,210 1,503 1,400 952 3,543 1,632 POT, EEL 0000000000020 POT, HAG 3,447 3,401 2,493 3,759 3,767 3,251 2,416 1,950 3,396 1,479 796 2,541 4,961 POT,CRAB 1,052 1,052 869 698 1,546 3,963 3,517 3,567 4,251 3,953 2,525 3,062 2,317 POT,FISH 1,283 1,643 1,709 2,081 1,668 862 1,239 2,404 1,195 1,442 1,264 1,380 836 POT,LOBSTER 20,362 22,221 21,493 24,847 26,015 24,589 23,321 21,087 21,559 20,577 14,757 20,005 21,197 POT,OTHER 2421013215031581042330169259 POT,SHRIMP 72 18 12 26 574 266 111 286 84 202 129 202 273 POTS, MIXED 105928875500000000 PURSE SEINE 81,689 110,605 58,520 83,012 83,307 78,248 66,817 55,910 47,509 50,838 51,868 101,744 111,240 SEINE, STOP 000000323115540 SEINE,DANISH 6,121 10,444 10,217 7,896 1,950 1,631 4,985 2,294 3,034 8 1,876 755 234 SEINE,SCOTTISH 269 268 221 135 235 278 125 170 104 11 0 0 0 TRAP 2,189 1,684 835 907 492 633 1,273 858 598 334 455 821 203 WEIR 0 0 50 326 262 278 570 271 330 0 0 19 0 total 596,087 612,768 595,234 579,204 613,757 688,438 624,225 662,133 724,832 639,583 571,683 629,617 570,215

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 20 DRAFT April 17, 2009 Table 2 – Percent of total landed pounds by gear type (source: NMFS vessel trip reports)

GEARNM 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 CARRIER VESSEL 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% CASTNET 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.1%0.0%0.0%0.0% DIVING GEAR 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% DREDGE, SCALLOP‐CHAIN MAT 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.6%0.6% DREDGE, URCHIN 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% DREDGE,MUSSEL 0.1%0.1%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.1%0.0% DREDGE,OCEAN QUAHOG/SURF CLAM 1.1% 0.1% 0.8% 0.1% 0.3% 0.2% 0.2% 0.1% 0.1% 0.2% 0.2% 0.1% 0.1% DREDGE,OTHER 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.0% 0.0% 0.0% DREDGE,SCALLOP,SEA 3.2% 3.0% 2.9% 4.4% 5.2% 6.6% 8.0% 8.2% 8.6% 8.5% 9.3% 8.8% 7.7% FYKE NET 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% GILL NET,DRIFT,LARGE MESH 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.1% 0.1% 0.2% 0.2% 0.2% GILL NET,DRIFT,SMALL MESH 0.1% 0.1% 0.2% 0.2% 0.2% 0.2% 0.2% 0.1% 0.1% 0.1% 0.0% 0.1% 0.0% GILL NET,RUNAROUND 0.0%0.0%0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.1% GILL NET,SINK 8.4% 7.7% 8.5% 7.7% 6.4% 5.5% 5.9% 6.3% 5.1% 5.1% 4.4% 15.7% 6.7% HAND LINE/ROD & REEL 0.4% 0.3% 0.4% 0.4% 0.4% 0.5% 0.5% 0.3% 0.3% 0.2% 0.2% 0.2% 0.2% HAND RAKE 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% HARPOON 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% HAUL SEINE 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% LONGLINE, PELAGIC 0.1%0.1%0.1%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% LONGLINE,BOTTOM 1.6%1.6%1.6%1.7%1.2%0.9%0.7%0.5%0.7%0.7%0.5%0.5%0.5% MIXED GEAR 0.1% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% OTHER GEAR 1.4% 1.2% 0.3% 0.0% 0.2% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% OTTER TRAWL, BEAM 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.1% 0.2% 0.1% 0.0% 0.1% 0.1% OTTER TRAWL,BOTTOM,FISH 39.5% 37.5% 42.1% 38.2% 35.1% 32.7% 32.2% 30.0% 34.2% 30.7% 28.2% 26.4% 28.8% OTTER TRAWL,BOTTOM,OTHER 0.1% 0.1% 0.1% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% OTTER TRAWL,BOTTOM,SCALLOP 0.2%0.2%0.3%0.4%0.4%0.5%0.6%0.5%0.4%0.3%0.2%0.2%0.2% OTTER TRAWL,BOTTOM,SHRIMP 3.0%2.5%1.5%1.1%1.5%0.6%0.5%0.5%0.7%0.7%0.8%1.6%1.9% OTTER TRAWL,MIDWATER 20.6% 17.6% 18.1% 16.0% 15.2% 14.8% 12.0% 10.2% 7.8% 9.1% 9.8% 5.1% 2.3% PAIR TRAWL,BOTTOM 0.0% 0.0% 0.0% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.0% 0.0% 0.0% PAIR TRAWL,MIDWATER 0.3% 3.0% 6.3% 7.9% 13.6% 20.3% 21.9% 29.2% 30.0% 31.1% 33.0% 18.8% 25.6% POT, CONCH/WHELK 0.1%0.1%0.1%0.2%0.3%0.3%0.3%0.3%0.2%0.2%0.2%0.6%0.3% POT, EEL 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% POT, HAG 0.6%0.6%0.4%0.6%0.6%0.5%0.4%0.3%0.5%0.2%0.1%0.4%0.9% POT,CRAB 0.2% 0.2% 0.1% 0.1% 0.3% 0.6% 0.6% 0.5% 0.6% 0.6% 0.4% 0.5% 0.4% POT,FISH 0.2% 0.3% 0.3% 0.4% 0.3% 0.1% 0.2% 0.4% 0.2% 0.2% 0.2% 0.2% 0.1% POT,LOBSTER 3.4%3.6%3.6%4.3%4.2%3.6%3.7%3.2%3.0%3.2%2.6%3.2%3.7% POT,OTHER 0.0%0.0%0.1%0.1%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% POT,SHRIMP 0.0% 0.0% 0.0% 0.0% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% POTS, MIXED 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% PURSE SEINE 13.7% 18.1% 9.8% 14.3% 13.6% 11.4% 10.7% 8.4% 6.6% 7.9% 9.1% 16.2% 19.5% SEINE, STOP 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% SEINE,DANISH 1.0% 1.7% 1.7% 1.4% 0.3% 0.2% 0.8% 0.3% 0.4% 0.0% 0.3% 0.1% 0.0% SEINE,SCOTTISH 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% TRAP 0.4% 0.3% 0.1% 0.2% 0.1% 0.1% 0.2% 0.1% 0.1% 0.1% 0.1% 0.1% 0.0% WEIR 0.0% 0.0% 0.0% 0.1% 0.0% 0.0% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 21 DRAFT April 17, 2009 Table 3 – Revenue by gear type (1,000 dollars, all values converted to 2007 dollars; source: NMFS vessel trip reports) GEARNM 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 CARRIER VESSEL 00000000000010 CASTNET 0000310756281123611 DIVING GEAR 371 356 177 175 147 94 81 78 81 58 12 8 5 DREDGE, SCALLOP‐CHAIN MAT0000000003431,41125,50722,934 DREDGE, URCHIN 112 128 127 208 153 114 67 52 57 105 109 22 104 DREDGE,MUSSEL 201292111810000531804083 DREDGE,OCEAN QUAHOG/SURF CLAM 8,075 565 4,002 684 1,450 1,565 880 667 1,549 4,560 5,199 3,933 1,564 DREDGE,OTHER 1,240 1,546 1,307 2,736 1,731 880 401 770 867 931 107 841 1,142 DREDGE,SCALLOP,SEA 131,362 119,704 94,851 145,839 183,848 210,929 241,939 271,784 354,412 441,855 375,956 357,267 294,304 FYKE NET 00000000332110 GILL NET,DRIFT,LARGE MESH 71 165 96 97 113 8 12 294 89 627 419 863 325 GILL NET,DRIFT,SMALL MESH 349 397 870 807 1,144 1,048 872 295 548 239 124 267 64 GILL NET,RUNAROUND 83 48 364 246 368 292 326 508 430 576 230 318 284 GILL NET,SINK 39,512 36,256 41,337 47,440 51,961 48,154 45,766 47,559 41,851 43,885 37,653 40,061 36,401 HAND LINE/ROD & REEL 8,325 5,110 5,580 5,925 6,860 8,996 7,331 4,153 2,885 1,752 1,721 2,088 1,059 HAND RAKE 0 0 0 0 12 2 0 160 26 210 66 400 55 HARPOON 945 509 568 646 1,945 735 315 311 61 31 41 11 28 HAUL SEINE 0000003410010 LONGLINE, PELAGIC 1,213 1,377 819 412 809 592 469 342 807 99 106 199 172 LONGLINE,BOTTOM 8,172 8,228 8,932 8,356 5,446 5,327 4,166 3,296 5,092 5,483 3,916 4,092 2,660 MIXED GEAR 4085013391225000000000 OTHER GEAR 6,8595,4192,7835341,4261076010390 OTTER TRAWL, BEAM 16 0 4 16 50 153 529 743 1,278 1,108 413 449 616 OTTER TRAWL,BOTTOM,FISH 226,763 204,184 219,144 207,375 207,206 218,814 201,782 197,663 208,425 195,431 164,913 161,524 137,823 OTTER TRAWL,BOTTOM,OTHER3888351,4095561,1713400000014 OTTER TRAWL,BOTTOM,SCALLOP 10,700 6,458 8,727 12,013 13,055 15,155 14,690 13,319 13,276 10,163 6,160 5,787 4,176 OTTER TRAWL,BOTTOM,SHRIMP 19,461 20,154 12,458 12,308 17,184 8,906 7,607 5,117 3,922 3,295 3,804 10,393 10,206 OTTER TRAWL,MIDWATER 14,874 13,815 13,853 9,682 10,877 9,085 7,667 7,802 6,541 7,142 9,572 4,299 1,722 PAIR TRAWL,BOTTOM 220 371 162 482 178 182 228 0 22 109 15 3 510 PAIR TRAWL,MIDWATER 146 1,343 3,837 3,581 6,436 10,716 12,850 19,184 23,303 22,325 27,302 12,650 16,625 POT, CONCH/WHELK 179 218 425 791 1,005 1,111 1,261 1,022 724 1,087 825 1,597 649 POT, EEL 0000000000220 POT, HAG 1,492 1,716 1,404 2,300 1,898 2,127 1,459 1,134 894 1,062 613 1,807 2,103 POT,CRAB 716 786 603 681 1,138 2,647 1,697 2,083 2,198 2,613 1,458 2,679 916 POT,FISH 2,078 3,100 3,116 3,539 2,823 1,724 2,337 3,335 2,741 3,415 3,812 3,355 2,041 POT,LOBSTER 85,360 84,729 75,724 98,900 94,390 85,325 83,106 77,726 76,865 82,172 74,433 67,879 51,629 POT,OTHER 178 147 257 285 163 38 16 3 5 16 0 261 175 POT,SHRIMP 49 19 15 34 572 311 147 247 60 158 67 78 132 POTS, MIXED 1932311391281200000000 PURSE SEINE 10,895 13,188 9,672 12,660 13,717 17,850 14,744 12,172 5,925 14,564 9,310 30,185 18,841 SEINE, STOP 00000011094440 SEINE,DANISH 2,219 5,137 4,763 4,228 1,110 1,211 2,670 978 1,364 5 630 437 51 SEINE,SCOTTISH 369 354 334 187 230 265 163 174 110 17 0 0 0 TRAP 1,629 1,001 473 840 582 628 1,021 714 410 519 636 604 181 WEIR 0 0 15 112 135 206 326 202 181 0 0 14 0 total 585,223 538,387 518,697 584,943 631,399 655,332 656,935 673,908 757,099 846,295 731,346 740,364 609,525

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 22 DRAFT April 17, 2009 Table 4 – Percent of total revenues by gear type (source: NMFS vessel trip reports)

GEARNM 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 CARRIER VESSEL 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% CASTNET 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% DIVING GEAR 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% DREDGE, SCALLOP‐CHAIN MAT 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.2%3.4%3.8% DREDGE, URCHIN 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% DREDGE,MUSSEL 0.0%0.1%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.1%0.0% DREDGE,OCEAN QUAHOG/SURF CLAM 1.4% 0.1% 0.8% 0.1% 0.2% 0.2% 0.1% 0.1% 0.2% 0.5% 0.7% 0.5% 0.3% DREDGE,OTHER 0.2% 0.3% 0.3% 0.5% 0.3% 0.1% 0.1% 0.1% 0.1% 0.1% 0.0% 0.1% 0.2% DREDGE,SCALLOP,SEA 22.4% 22.2% 18.3% 24.9% 29.1% 32.2% 36.8% 40.3% 46.8% 52.2% 51.4% 48.3% 48.3% FYKE NET 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% GILL NET,DRIFT,LARGE MESH 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.1% 0.1% 0.1% GILL NET,DRIFT,SMALL MESH 0.1% 0.1% 0.2% 0.1% 0.2% 0.2% 0.1% 0.0% 0.1% 0.0% 0.0% 0.0% 0.0% GILL NET,RUNAROUND 0.0%0.0%0.1%0.0%0.1%0.0%0.0%0.1%0.1%0.1%0.0%0.0%0.0% GILL NET,SINK 6.8% 6.7% 8.0% 8.1% 8.2% 7.3% 7.0% 7.1% 5.5% 5.2% 5.1% 5.4% 6.0% HAND LINE/ROD & REEL 1.4% 0.9% 1.1% 1.0% 1.1% 1.4% 1.1% 0.6% 0.4% 0.2% 0.2% 0.3% 0.2% HAND RAKE 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.0% HARPOON 0.2% 0.1% 0.1% 0.1% 0.3% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% HAUL SEINE 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% LONGLINE, PELAGIC 0.2%0.3%0.2%0.1%0.1%0.1%0.1%0.1%0.1%0.0%0.0%0.0%0.0% LONGLINE,BOTTOM 1.4%1.5%1.7%1.4%0.9%0.8%0.6%0.5%0.7%0.6%0.5%0.6%0.4% MIXED GEAR 0.1% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% OTHER GEAR 1.2% 1.0% 0.5% 0.1% 0.2% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% OTTER TRAWL, BEAM 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.1% 0.2% 0.1% 0.1% 0.1% 0.1% OTTER TRAWL,BOTTOM,FISH 38.7% 37.9% 42.2% 35.5% 32.8% 33.4% 30.7% 29.3% 27.5% 23.1% 22.5% 21.8% 22.6% OTTER TRAWL,BOTTOM,OTHER 0.1% 0.2% 0.3% 0.1% 0.2% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% OTTER TRAWL,BOTTOM,SCALLOP 1.8% 1.2% 1.7% 2.1% 2.1% 2.3% 2.2% 2.0% 1.8% 1.2% 0.8% 0.8% 0.7% OTTER TRAWL,BOTTOM,SHRIMP 3.3%3.7%2.4%2.1%2.7%1.4%1.2%0.8%0.5%0.4%0.5%1.4%1.7% OTTER TRAWL,MIDWATER 2.5% 2.6% 2.7% 1.7% 1.7% 1.4% 1.2% 1.2% 0.9% 0.8% 1.3% 0.6% 0.3% PAIR TRAWL,BOTTOM 0.0% 0.1% 0.0% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% PAIR TRAWL,MIDWATER 0.0% 0.2% 0.7% 0.6% 1.0% 1.6% 2.0% 2.8% 3.1% 2.6% 3.7% 1.7% 2.7% POT, CONCH/WHELK 0.0%0.0%0.1%0.1%0.2%0.2%0.2%0.2%0.1%0.1%0.1%0.2%0.1% POT, EEL 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% POT, HAG 0.3%0.3%0.3%0.4%0.3%0.3%0.2%0.2%0.1%0.1%0.1%0.2%0.3% POT,CRAB 0.1% 0.1% 0.1% 0.1% 0.2% 0.4% 0.3% 0.3% 0.3% 0.3% 0.2% 0.4% 0.2% POT,FISH 0.4% 0.6% 0.6% 0.6% 0.4% 0.3% 0.4% 0.5% 0.4% 0.4% 0.5% 0.5% 0.3% POT,LOBSTER 14.6% 15.7% 14.6% 16.9% 14.9% 13.0% 12.7% 11.5% 10.2% 9.7% 10.2% 9.2% 8.5% POT,OTHER 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% POT,SHRIMP 0.0% 0.0% 0.0% 0.0% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% POTS, MIXED 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% PURSE SEINE 1.9%2.4%1.9%2.2%2.2%2.7%2.2%1.8%0.8%1.7%1.3%4.1%3.1% SEINE, STOP 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% SEINE,DANISH 0.4% 1.0% 0.9% 0.7% 0.2% 0.2% 0.4% 0.1% 0.2% 0.0% 0.1% 0.1% 0.0% SEINE,SCOTTISH 0.1% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% TRAP 0.3% 0.2% 0.1% 0.1% 0.1% 0.1% 0.2% 0.1% 0.1% 0.1% 0.1% 0.1% 0.0% WEIR 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 23 DRAFT April 17, 2009 Table 5 – Days absent by gear type (source: NMFS vessel trip reports) GEARNM 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 CARRIER VESSEL 0000000000001 CASTNET 0 0 0 0 21 3 0 11 13 135 28 53 6 DIVING GEAR 2191311361168011279586428101514 DREDGE, SCALLOP‐CHAIN MAT000000000341194,3203,894 DREDGE, URCHIN 1071151351571319154473217141324 DREDGE,MUSSEL 585434392 1 0 0 0 210321 DREDGE,OCEAN QUAHOG/SURF CLAM 702 396 373 507 468 894 746 336 496 1,979 2,176 2,553 1,865 DREDGE,OTHER 1,624 1,363 2,002 1,973 872 331 190 253 208 216 186 257 220 DREDGE,SCALLOP,SEA 109,552 92,014 117,521 97,355 82,237 75,244 76,528 74,358 70,777 68,084 65,721 78,181 55,904 FYKE NET 00000010284860 GILL NET,DRIFT,LARGE MESH 403 103 434 49 82 10 13 379 658 591 546 809 407 GILL NET,DRIFT,SMALL MESH 360 513 985 1,401 1,276 1,057 666 306 462 206 94 224 103 GILL NET,RUNAROUND 179 70 434 489 685 476 648 800 683 506 429 443 486 GILL NET,SINK 61,044 48,126 53,873 57,506 65,451 69,240 55,734 54,454 50,288 45,468 33,627 41,899 41,166 HAND LINE/ROD & REEL 6,282 6,533 8,559 7,654 7,016 9,065 8,752 7,542 6,609 5,251 4,023 6,243 3,570 HAND RAKE 0 0 0 0 40 35 14 46 25 36 50 43 17 HARPOON 78 88 115 159 225 243 143 93 19 7 7 16 12 HAUL SEINE 00000012450050 LONGLINE, PELAGIC 3,564 2,450 2,061 730 1,675 1,657 1,785 1,271 1,964 704 127 831 914 LONGLINE,BOTTOM 13,108 12,749 16,061 10,894 7,575 6,713 6,832 5,411 5,986 5,881 3,993 5,373 4,355 MIXED GEAR 1,83439850925310400000000 OTHER GEAR 9,698 6,955 5,267 580 1,611 144 24 1 3 2 1 13 0 OTTER TRAWL, BEAM 9 3 162 48 134 347 912 2,121 2,805 1,576 485 522 852 OTTER TRAWL,BOTTOM,FISH 437,190 376,357 400,592 399,583 367,867 394,397 355,604 329,149 314,677 315,865 233,359 266,620 239,546 OTTER TRAWL,BOTTOM,OTHER 1,002 1,838 2,448 381 852 112 0 0 0 0 0 0 16 OTTER TRAWL,BOTTOM,SCALLOP 3,654 4,119 5,802 5,211 3,991 4,327 4,234 3,976 4,395 5,052 3,493 3,656 1,723 OTTER TRAWL,BOTTOM,SHRIMP 13,677 18,956 15,949 17,802 16,790 11,428 9,406 5,178 6,717 4,418 4,611 9,756 10,235 OTTER TRAWL,MIDWATER 4,859 4,475 4,005 2,651 3,219 3,527 2,830 1,733 1,761 2,157 1,475 1,132 784 PAIR TRAWL,BOTTOM 140 478 298 474 151 410 570 0 37 12 52 0 1,317 PAIR TRAWL,MIDWATER 39 419 652 1,191 1,842 3,514 3,118 4,184 4,142 4,626 3,488 2,335 3,331 POT, CONCH/WHELK 212 212 300 326 591 653 620 564 519 524 401 665 618 POT, EEL 0000000000020 POT, HAG 489 591 420 523 615 579 463 257 257 287 197 495 761 POT,CRAB 212 312 341 402 566 822 507 701 1,084 953 706 844 607 POT,FISH 1,603 1,995 2,644 2,705 1,887 1,587 1,882 2,662 2,502 2,932 2,331 3,030 1,967 POT,LOBSTER 39,561 39,198 41,904 43,058 43,225 42,503 38,609 38,713 38,910 33,631 25,351 35,547 32,904 POT,OTHER 89 156 93 202 58 23 8 3 6 3 0 79 84 POT,SHRIMP 78 41 11 16 246 200 95 108 121 76 75 92 89 POTS, MIXED 2562132471742700000000 PURSE SEINE 1,791 2,496 1,599 1,166 1,513 997 1,143 922 968 775 606 1,480 1,768 SEINE, STOP 0000002676430 SEINE,DANISH 36 72 63 60 15 17 27 10 28 4 12 13 2 SEINE,SCOTTISH 442 499 470 479 467 378 229 176 207 34 2 0 0 TRAP 741 561 777 492 221 284 667 1,136 966 855 750 1,272 170 WEIR 0 0 5 60 80 102 119 104 76 0 0 29 0 total 714,892 625,049 687,281 656,866 613,908 631,523 573,266 537,073 518,505 502,937 388,567 468,901 409,733

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 24 DRAFT April 17, 2009 Table 6 – Percent of days absent by gear type (source: NMFS vessel trip reports)

GEARNM 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 CARRIER VESSEL 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% CASTNET 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% DIVING GEAR 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% DREDGE, SCALLOP‐CHAIN MAT 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.9%1.0% DREDGE, URCHIN 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% DREDGE,MUSSEL 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% DREDGE,OCEAN QUAHOG/SURF CLAM 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.4% 0.6% 0.5% 0.5% DREDGE,OTHER 0.2% 0.2% 0.3% 0.3% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.1% DREDGE,SCALLOP,SEA 15.3% 14.7% 17.1% 14.8% 13.4% 11.9% 13.3% 13.8% 13.7% 13.5% 16.9% 16.7% 13.6% FYKE NET 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% GILL NET,DRIFT,LARGE MESH 0.1% 0.0% 0.1% 0.0% 0.0% 0.0% 0.0% 0.1% 0.1% 0.1% 0.1% 0.2% 0.1% GILL NET,DRIFT,SMALL MESH 0.1% 0.1% 0.1% 0.2% 0.2% 0.2% 0.1% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% GILL NET,RUNAROUND 0.0%0.0%0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.1% GILL NET,SINK 8.5% 7.7% 7.8% 8.8% 10.7% 11.0% 9.7% 10.1% 9.7% 9.0% 8.7% 8.9% 10.0% HAND LINE/ROD & REEL 0.9% 1.0% 1.2% 1.2% 1.1% 1.4% 1.5% 1.4% 1.3% 1.0% 1.0% 1.3% 0.9% HAND RAKE 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% HARPOON 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% HAUL SEINE 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% LONGLINE, PELAGIC 0.5%0.4%0.3%0.1%0.3%0.3%0.3%0.2%0.4%0.1%0.0%0.2%0.2% LONGLINE,BOTTOM 1.8%2.0%2.3%1.7%1.2%1.1%1.2%1.0%1.2%1.2%1.0%1.1%1.1% MIXED GEAR 0.3% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% OTHER GEAR 1.4% 1.1% 0.8% 0.1% 0.3% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% OTTER TRAWL, BEAM 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.2% 0.4% 0.5% 0.3% 0.1% 0.1% 0.2% OTTER TRAWL,BOTTOM,FISH 61.2% 60.2% 58.3% 60.8% 59.9% 62.5% 62.0% 61.3% 60.7% 62.8% 60.1% 56.9% 58.5% OTTER TRAWL,BOTTOM,OTHER 0.1% 0.3% 0.4% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% OTTER TRAWL,BOTTOM,SCALLOP 0.5% 0.7% 0.8% 0.8% 0.7% 0.7% 0.7% 0.7% 0.8% 1.0% 0.9% 0.8% 0.4% OTTER TRAWL,BOTTOM,SHRIMP 1.9%3.0%2.3%2.7%2.7%1.8%1.6%1.0%1.3%0.9%1.2%2.1%2.5% OTTER TRAWL,MIDWATER 0.7% 0.7% 0.6% 0.4% 0.5% 0.6% 0.5% 0.3% 0.3% 0.4% 0.4% 0.2% 0.2% PAIR TRAWL,BOTTOM 0.0%0.1%0.0%0.1%0.0%0.1%0.1%0.0%0.0%0.0%0.0%0.0%0.3% PAIR TRAWL,MIDWATER 0.0% 0.1% 0.1% 0.2% 0.3% 0.6% 0.5% 0.8% 0.8% 0.9% 0.9% 0.5% 0.8% POT, CONCH/WHELK 0.0%0.0%0.0%0.0%0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.2% POT, EEL 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% POT, HAG 0.1%0.1%0.1%0.1%0.1%0.1%0.1%0.0%0.0%0.1%0.1%0.1%0.2% POT,CRAB 0.0% 0.0% 0.0% 0.1% 0.1% 0.1% 0.1% 0.1% 0.2% 0.2% 0.2% 0.2% 0.1% POT,FISH 0.2% 0.3% 0.4% 0.4% 0.3% 0.3% 0.3% 0.5% 0.5% 0.6% 0.6% 0.6% 0.5% POT,LOBSTER 5.5%6.3%6.1%6.6%7.0%6.7%6.7%7.2%7.5%6.7%6.5%7.6%8.0% POT,OTHER 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% POT,SHRIMP 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% POTS, MIXED 0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0%0.0% PURSE SEINE 0.3%0.4%0.2%0.2%0.2%0.2%0.2%0.2%0.2%0.2%0.2%0.3%0.4% SEINE, STOP 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% SEINE,DANISH 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% SEINE,SCOTTISH 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% TRAP 0.1% 0.1% 0.1% 0.1% 0.0% 0.0% 0.1% 0.2% 0.2% 0.2% 0.2% 0.3% 0.0% WEIR 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 25 DRAFT April 17, 2009 Table 7 ‐ Fishing Gears Used in Estuaries and Bays, Coastal Waters, and Offshore Waters of the EEZ, from Maine to North Carolina GEAR Estuary or Bay Coastal Offshore Contacts Federally 0‐3 Miles 3‐200 Miles Bottom Regulated Bag Nets X X X X Beam Trawls X X X X X By Hand X X X Cast Nets X X X Clam Kicking X X Diving Outfits X X X Dredge Clam X X X X X Dredge Conch X X Dredge Crab X X X Dredge Mussel X X X Dredge , Common X X Dredge Scallop, Bay X X Dredge Scallop, Sea X X X X Dredge Urchin, Sea X X X Floating Traps (Shallow) X X X X Fyke And Hoop Nets, Fish X X X Gill Nets, Drift, Other X X Gill Nets, Drift, Runaround X X Gill Nets, Sink/Anchor, Other X X X X X Gill Nets, Stake X X X X X Haul Seines, Beach X X X Haul Seines, Long X X X Haul Seines, Long(Danish) X X X X Hoes X X Lines Hand, Other X X X X Lines Long Set With Hooks X X X X Lines Long, Reef Fish X X X X Lines Long, Shark X X X Lines Troll, Other X X X Lines Trot With Baits X X X Otter Trawl Bottom, Crab X X X X Otter Trawl Bottom, Fish X X X X X Otter Trawl Bottom, Scallop X X X X Otter Trawl Bottom, Shrimp X X X X X Otter Trawl Midwater X X X Pots And Traps, Conch X X X Pots and Traps, Crab, Blue Peeler X X X Pots And Traps, Crab, Blue X X X Pots And Traps, Crab, Other X X X X X Pots And Traps, Eel X X X Pots and Traps, Lobster Inshore X X X Pots and Traps, Lobster Offshore X X X Pots and Traps, Fish X X X X X Pound Nets, Crab X X X Pound Nets, Fish X X X Purse Seines, Herring X X X Purse Seines, Menhaden X X Purse Seines, Tuna X X X Rakes X X Reel, Electric or Hydraulic X X X Rod and Reel X X X X Scottish Seine X X X X Scrapes X X Spears X X X Stop Seines X X Tongs and Grabs, Oyster X X Tongs Patent, Clam Other X X Tongs Patent, Oyster X X Trawl Midwater, Paired X X X Weirs X X

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 26 DRAFT April 17, 2009 Five primary gear types individually account for roughly 1% or greater of landings, revenues and/or days absent; these gears are the focus of evaluation in this document. Detailed gear descriptions are included for the following gear categories: demersal otter trawl (with sub‐ categories), New Bedford‐style scallop dredge, hydraulic clam dredge, fish and shellfish trap, and demersal longline.

Unless otherwise noted, the following descriptions are based on Sainsbury (1996), DeAlteris (1998), Everhart and Youngs (1981), and the report of a panel of science and representatives on the effects of fishing gear on in the region (NREFHSC 2002), updated in Stevenson et al. (2004). Descriptions of fishing gears used in state waters are based on Stephan et al. (2000). Additional amplifying information was provided by the Council’s Habitat Advisory Panel. These descriptions are based on the available literature and are necessarily oversimplifications of the wide variety of fishing gear configurations actually in use.

4.2.1 Demersal otter trawls Demersal, or bottom, otter trawls are towed along the seafloor to catch a variety of species throughout the region. They account for a higher proportion of the catch of federally‐managed species than any other gear type. Use of demersal otter trawls in the region is managed under several federal FMPs developed by the NEFMC and MAFMC, including Northeast Multispecies; Atlantic Sea Scallop; Monkfish; Small Mesh Multispecies; Atlantic Mackerel, , and Butterfish; Dogfish; Skates; and Summer Flounder, Scup, and Black Sea Bass. Otter is also managed under various interstate FMPs developed by the ASMFC, including Northern Shrimp.

Trawl gear components include the warps, which attach the gear to the vessel; the doors, which hold the net open under water, the ground cables and bridles, which attach the door to the wings of the net; and the net itself. The top opening of the net, or headrope, is rigged with floats, and the lower opening, or groundrope, is rigged with a sweep, which varies in design depending on the target species (e.g., whether they are found on or off the bottom) as well as the roughness and hardness of the bottom. The net terminates in a codend, which has a drawstring opening that can be untied easily to dump the catch on deck. Three components of the otter trawl typically come in contact with the seafloor: the doors; the ground cables and lower bridles; and the footrope and sweep. The body of the net may also contact the seabed and may be rigged with chafing gear to avoid damage to the net, but this is not believed to be a common occurrence (S. Eayrs, personal communication).

The traditional otter board, or door, is a flat, rectangular wooden structure with steel fittings and a steel “shoe” along the leading and bottom edges that prevents damage as the door drags over the bottom. In the Northeast Region, wooden doors have been largely replaced by heavier, more efficient, steel doors. Two types of steel doors commonly used in the region are the V‐ shaped “Thyboron” door and the cambered (or curved) “Bison” door. Either type of door can be slotted to allow some water to flow through the door, reducing drag in the water. Steel “shoes” can be added at the bottom of the door to aid in keeping it upright and take the wear

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 27 DRAFT April 17, 2009 from bottom contact. The sizes and weights of trawl doors used in the Northeast region vary according to the size and type of trawl, and the size and horsepower of the vessel. Large steel doors 43‐54 ft2 (4‐5 m2) weigh between 1500‐2200 lb (700‐1000 kg).

The attachment point of the warps on the doors creates the towing angle, which in turn generates the hydrodynamic forces needed to push the door outward and downward, thus spreading the wings of the net. The non‐traditional door designs increase the spreading force of the door by increasing direct pressure on the face of the door and/or by creating more suction on the back of the door. On fine‐grained sediments, the doors create a silt cloud that aids in herding fish into the mouth of the net. On rocky or more irregular bottom, trawl doors impact rocks in a jarring manner and can jump distances of 3‐6 ft (1‐2 m) (Carr and Milliken 1998).

Steel ground cables attach the doors to the wings of the net. Each ground cable runs from a door to the upper and lower bridles, which attach to the top and bottom of the net wing. Thus, both the ground cables and the lower bridles contact the bottom. In New England, fixed rubber roller disks (sometimes called cookies) are attached to the ground cables and lower bridles to assist the passage of the trawl over the bottom. Depending upon bottom conditions, towing speed, and fish behavior, ground cables vary in length from 0‐240 ft (0‐73 m), and bridles vary in length from 30‐240 ft (9‐73 m).

As mentioned above, sweep type varies by target species and substrate. In New England, two types of sweep are used on smooth bottom (Mirarchi 1998). In the traditional chain sweep, loops of chain are suspended from a steel cable, with only 2‐3 links of the chain touching bottom. Contact of the chain with the bottom increases the buoyancy of the trawl so that it skims just a few inches above the bottom to catch species such as squid and scup. Another type of smooth bottom sweep uses a heavy chain with rubber cookies instead of a cable, and is used to catch flounder. The cookies vary in diameter from 4 to 16 in (10 to 41) and do not rotate (Carr and Milliken 1998). This type of sweep is always in contact with the bottom.

On rough bottoms, roller and rockhopper sweeps are used (Carr and Milliken 1998). On the roller sweeps, vertical rubber rollers as large as 36 in (91 cm) in diameter are placed at intervals along the sweep. Although the rollers are free to rotate, because the sweep is shaped in a curve, only the rollers that are located at or near the center of the sweep actually “roll” over the bottom; the others are oriented at increasing angles to the direction of the tow and do not rotate freely as they are dragged over the bottom. In New England, roller sweeps have been largely replaced with rockhopper sweeps that use larger diameter fixed rollers, and are designed to “hop” over rocks as large as 1 m in diameter. Small rubber “spacer” disks are placed in between the larger rubber disks in both types of sweep. Rockhopper gear is no longer used exclusively on hard bottom habitats, but is actually quite versatile and used in a variety of habitat types (Carr and Milliken 1998).

A number of different types of bottom otter trawls are designed to catch certain species of fish on specific bottom types and at particular times of year. Some of the major differences in

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 28 DRAFT April 17, 2009 bottom trawl designs designed to catch groundfish, scallops, shrimp, and squid, are described below; these descriptions are necessarily generalizations because there are infinite variations on each basic trawl type. The raised footrope trawl is also described.

4.2.1.1 Groundfish trawls Groundfish trawls can be divided into two classes, those rigged to target flatfish, and those rigged to target fish that rise off bottom. Flatfish trawls are designed with a low net opening between the headrope and the footrope and more ground rigging (i.e., rubber cookies and chain) on the sweep (Mirarchi 1998). This design allows the sweep to follow the contours in the bottom in order to encourage flatfish, which lie in contact with the seafloor, to swim off the bottom and into the net. It is used on smooth mud and sand. A high‐rise or fly net with larger mesh has a wide net opening and is used to catch demersal fish that rise higher off the bottom, e.g. haddock and cod (NREFHSC 2002). Trawls used on gravel or rocky bottom, or on mud or sand bottom with occasional boulders, may be rigged with rockhopper gear, intended to get the sweep over irregularities in the bottom without damaging the net.

4.2.1.2 Scallop trawls Scallop trawls are used on sandy bottoms, typically in waters from Long Island south to the Virginia coast. Vessels typically use wooden doors, and fishing usually occurs in waters less than 40 fathoms (approximately 75 m) deep. Cable lengths vary from 3:1 to 5:1 ratios of cable to depth. Typical scallop trawls are 55 or 65 ft (17 or 20 m) two seam nets with body and wings constructed of 5 in2, 4mm or 5mm braided poly webbing. Wings are 20 to 25 ft (6‐8 m) long cut on an 8:1 or 10:1 taper, while the body and belly sections are 20 to 23 ft (6‐7 m) long and are cut on a 10:1 taper. Body and belly sections are identical with no over hang and both top and bottom lines are hung on 5/8 inch combination cable. Varying numbers of 8 inch (20 cm) hard plastic floats are used on the headrope, while the footrope is lined with 0.375 in to 0.5 in (1‐1.3 cm) loop chain either single or double looped along the entire length. Some fishermen also use tickler chains ahead of the trawl to help kick up scallops. No trawl extensions are used and the tailbag sections are 60 meshes around by 50 meshes deep and are constructed of 5 in2, 4mm or 5mm, braided, double poly webbing. A whisker‐type chaffing gear is used along the underside of the trawl and bag to reduce wear. Scallop trawls are not disaggregated in the Vulnerability Assessment; scallop trawl effort is evaluated together with groundfish trawls under the groundfish trawl matrix.

4.2.1.3 Shrimp trawls The northern shrimp trawl fishery is prosecuted primarily in the western Gulf of Maine on mud and muddy sand substrates in depths between 20 and 100 fathoms (37‐183 m). The fishery is seasonal, beginning in December and extending as late as May. Gear used in the northern shrimp fishery is required by regulation to include a Nordmore grate to minimize bycatch of other bottom dwelling species, and is generally thought to be rigged for lighter contact on bottom (also for bycatch reduction). Footropes range in length from 140‐100 ft (12‐30 m), but most are 50‐90 ft (15‐27 m). Regulations require that northern shrimp trawls may not be used with ground cables and that the “legs” of the bridles not exceed 90 ft (27 m). Shrimp trawls use

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 29 DRAFT April 17, 2009 12 in (30.5 cm) or greater rockhoppers and 1 in (2.5 cm) mesh. Trawling is generally restricted to daylight hours, when shrimp are lower in the water column. Tow times may typically be two hours.

4.2.1.4 Squid trawls Bottom otter trawls used to catch species like scup and squid that swim over the bottom are rigged very lightly, with loops of chain suspended from the sweep (Mirarchi 1998). This gear is designed to skim along the seafloor with only two or three links of each loop of chain touching the bottom.

4.2.1.5 Raised footrope trawls The raised‐footrope trawl is designed capture small mesh species (silver hake, red hake, and dogfish). Raised‐footrope trawls can be rigged with or without a chain sweep. If no sweep is used, drop chains must be hung at defined intervals along the footrope. In trawls with a sweep, chains connect the sweep to the footrope. Both configurations are designed to make the trawl fish about 0.45 ‐ 0.6 m (1.5 ‐ 2 ft) above the bottom (Carr and Milliken 1998). Although the doors of the trawl still ride on the bottom, underwater video and observations in flume tanks have confirmed that the sweep in the raised footrope trawl has much less contact with the sea floor than does the traditional cookie sweep that it replaces (Carr and Milliken 1998).

Floats of approx 8 in (20 cm) in diameter are attached to the entire length of the headrope, with a maximum spacing of 4 ft (1.2 m) between floats. The ground gear is bare wire. The top and bottom legs are equal in length, and net fishes with no extensions. The total length of ground cables and legs must not be greater than 240 ft (73 m) from the doors to wing ends. The sweep and its rigging, including drop chains, must be made entirely of bare chain with a maximum diameter of 0.3 in (0.8 cm). No wrapping or cookies are allowed on the drop chains or sweep.

A raised footrope trawl fishery for shrimp in the inshore Gulf of Maine requires the use of a grate to exclude finfish bycatch, but otherwise, the net is rigged and fished in a manner similar to other raised footrope trawls.

4.2.2 New Bedford‐style scallop dredges The New Bedford‐style scallop dredge is the primary gear used in the and Mid‐ Atlantic sea scallop fishery. The use of scallop dredges in federal waters of the Northeast Region is managed under the federal Atlantic Sea Scallop FMP, developed by the NEFMC in consultation with the MAFMC.

In the Northeast Region, scallop dredges are used in high‐ and low‐energy sand environments, and high‐energy gravel environments. Although gravel exists in low‐energy environments of deepwater banks and ridges in the GOM, the fishery is not prosecuted there.

A New Bedford‐style scallop dredge consists of a chain bag and a steel towing frame. The bag is made of two sheets of 4 in (10 cm) metal rings. The upper portion of the bag includes a 10 in

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 30 DRAFT April 17, 2009 mesh twine top designed to allow fish to escape, and the lower portion is rigged with chafing gear. During fishing, the bag drags on the substrate. The frame consists of a flat steel cutting bar and a pressure plate mounted above it which run parallel to the direction of the tow, and a triangular frame which connects the cutting bar and pressure plate to the single towing wire. The pressure plate generates hydrodynamic pressure, while the cutting bar rides along the surface of the substrate. Shoes on the right and left sides of the cutting bar ride along the substrate surface and are intended to take much of the wear. A sweep chain is attached to each shoe and to the forward portion of the bottom panel of the ring bag (Smolowitz 1998). Tickler chains run from side to side between the frame and the ring bag, and, in hard‐bottom scalloping, a series of rock chains run from front to back to prevent large rocks from getting into the bag.

New Bedford‐style dredges are typically 15 ft (4.5 m) wide; one or two of them are towed by single vessels at speeds of 4‐5 knots (7.4‐9.3 km∙hr‐1). Towing times are highly variable, depending on the density of marketable‐sized sea scallops at any given location, and may be as short as 10 minutes or as long as an hour. New Bedford‐style dredges used along the Maine coast are smaller, and dredges used on hard bottoms are heavier and stronger than dredges used on sand.

4.2.3 Hydraulic clam dredges Hydraulic clam dredges have been used in the Atlantic surfclam (Spisula solidissima) fishery for over five decades, and in the ocean quahog () fishery since its inception in the early 1970s. Use of this gear in the region is managed under the federal FMP for surf and ocean quahogs developed by the MAFMC. The gear is also used in state waters in the Mid‐ Atlantic region.

Hydraulic clam dredges can be operated in areas of large‐grain sand, fine sand, sand with small‐grain gravel, sand with small amounts of mud, and sand with very small amounts of clay. Most tows are made in large‐grain sand. Surfclam/ocean quahog dredges are not fished in clay, mud, pebbles, rocks, coral, large gravel >0.5 in (> 1.25 cm), or seagrass beds.

The typical dredge is 12 ft (3.7 m) wide and about 22 ft (6.7 m) long, and uses pressurized water jets to wash clams out of the seafloor. Towing speed at the start of the tow is about 2.5 knots (4.6 km∙hr‐1), and declines as the dredge accumulates clams. The dredge is retrieved once the vessel speed drops below about 1.5 knots (2.8 km∙hr‐1), which can be only a few minutes in very dense beds. However, a typical tow lasts about 15 minutes. The water jets penetrate the sediment in front of the dredge to a depth of about 8‐10 in (20‐25 cm) and help to “drive” the dredge forward. The water pressure required to fluidize the sediment varies from 50 lb∙in‐2 (psi) in coarse sand to 110 psi in finer sediments. The objective is to use as little pressure as possible since too much pressure will blow sediment into the clams and reduce product quality. The “knife” (or “cutting bar”) on the leading bottom edge of the dredge opening is 5.5 in (14 cm) deep for surfclams and 3.5 in (9 cm) for ocean quahogs. The knife “picks up” clams that

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 31 DRAFT April 17, 2009 have been separated from the sediment and guides them into the body of the dredge (“the cage”).

4.2.4 Fish and shellfish traps Traps are used to capture lobsters, crabs, black sea bass, eels, and other bottom‐dwelling species seeking food or shelter. Trap fishing can be divided into two general classifications: 1) inshore trapping in estuaries, lagoons, inlets, and bays in depths up to about 75 m (250 ft); and 2) offshore trapping using larger and heavier vessels and gear in depths up to 730 m (2400 ft) or more.

4.2.4.1 Lobster traps Originally, traps used to harvest American lobster (Homarus americanus) are constructed of wooden laths with single, and later, double, funnel entrances made from net twine. Today, roughly 95% are made from coated wire mesh. They are rectangular and are divided into two sections, the “kitchen” and the “parlor.” The kitchen has an entrance on both sides of the pot and is baited. Lobsters enter either chamber then move to the parlor through a long, sloping tunnel to the parlor. Escape vents are installed in both areas of the pot to minimize the retention of sub‐legal‐sized lobsters. Rock crabs (Cancer spp.) are also harvested in lobster pots.

Lobster traps are fished as either a single trap per buoy, 2 or 3 traps per buoy, or strung together in “trawls” of up to 100 traps Trawls are used on flatter types of bottom. Traps in trawls are connected by “mainlines” which either float off the bottom, or, in areas where they are likely to become entangled with marine mammals, sink to the bottom. Single traps are often used in rough, hard bottom areas where lines connecting traps in a trawl line tend to become entangled in bottom structures.

Soak time depends on season and location, ranging from 1‐3 days in inshore waters in warm weather, up to several weeks in colder waters. Offshore traps are larger (>1.2 m (4 ft) long) and heavier (~45 kg (100 lb)) than inshore traps with an average of about 40 traps per trawl. They are usually deployed for a week at a time. Although the offshore component of the fishery is regulated under federal rules, American lobster is not managed under a federal fishery management plan.

4.2.4.2 Deep‐sea red crab traps Deep‐sea red crab (Geryon quinquedens) traps are traditionally wood and wire traps that are 48 in long, 30 in wide, and 20 in high (1.20 x 0.75 x 0.5 m) with a top entry. A second style of trap used in this fishery is conical in shape, 4 ft (1.3 m) in diameter at the base and 22 in (0.45 m) high with a top entry funnel. Six large (90‐150 ft, 27‐46 m) vessels are engaged in the deep‐sea red crab fishery, which is managed by the NEFMC (NEFMC 2002). Vessels use an average of 560 traps that are deployed in trawls of 75‐180 traps per trawl along the continental slope at depths of 1300‐2600 ft (400‐800 m). The traps are transported to and from the fishing grounds during each trip and are generally hauled daily.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 32 DRAFT April 17, 2009 4.2.5 Demersal longlines A longline is a long length of line, often several miles long, to which short lengths of line (“gangions”) carrying baited hooks are attached. Demersal longlining is used to catch a wide range of species on continental shelf areas and offshore banks. Bottom longlines are also referred to as “trot” lines and are used in the Mid‐Atlantic to harvest blue crabs (Callinectes sapidus).

Bottom longline fishing in the Northeast Region is conducted using hand‐baited gear that is stored in tubs before the vessel goes fishing and by vessels equipped with automated “snap‐on” or “racking” systems. The gangions are 15 in (38 cm) long and spaced 3‐6 ft (0.9‐1.8 m) apart. The mainline, hooks, and gangions all contact the bottom. In the Cape Cod longline fishery, up to six individual longlines are strung together, for a total length of about 1500 ft (460 m), and are deployed with 20‐24 lb (9‐11 kg) anchors. Each set consists of 600 to 1200 hooks. In tub trawls, the mainline is parachute cord; stainless steel wire and monofilament nylon gangions are used in snap‐on systems (Leach 1998). The gangions are snapped on to the mainline as it pays off a drum and removed and rebaited when the wire is hauled. In New England, longlines are usually set for only a few hours at a time in areas with attached benthic epifauna. Longlines used for tilefish are deployed in deep water, may be up to 25 mi (40 km) long, are stainless steel or galvanized wire, and are set in a zigzag fashion. These activities are managed under federal fishery management plans.

4.2.6 Sink gill nets A gill net is a large wall of netting which may be set at or below the surface, on the seafloor, or at any depth between. They are equipped with floats at the top and lead weights along the bottom. Sink, or bottom gill nets are anchored or staked in position. Fish are caught as they try to pass through the net meshes. Gill nets are highly selective because the species and sizes of fish caught are highly dependant on the mesh size of the net. They are used to catch a wide range of species, including many federally‐managed species. Bottom gill net fishing occurs in the Northeast Region in nearshore coastal and estuarine waters as well as offshore on the continental shelf. The use of sink gill nets in federal waters is managed under federal fishery management plans. The use of gill nets is restricted or prohibited in some state waters in the region.

Gill nets have three components: leadline, netting, and floatline. Leadlines used in New England are 65 lb (30 kg) per net; leadlines used in the Mid‐Atlantic are slightly heavier. The netting is monofilament nylon, and the mesh size varies, depending on the target species. Nets are anchored at each end, using materials such as pieces of railroad track, sash weights, or Danforth anchors. Anchors and leadlines have the most contact with the bottom. Individual gill nets are typically 300 ft (91 m) long and 12 ft (3.6 m) high. Strings of nets may be set out in straight lines, often across the current, or in various other configurations (e.g., circles), depending upon bottom and current conditions.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 33 DRAFT April 17, 2009 In New England, bottom gill nets are fished in strings of 5‐20 nets attached end to end. They are fished in two different ways, as “stand up” and “tie‐down” nets (Williamson, 1998). Stand‐up nets are used to catch cod, haddock, pollock, and hake and are soaked for 12‐24 hrs. Tie‐down nets are set with the float line tied to the lead line at 1.8 m (6 ft) intervals so the float line is close to the bottom and the net forms a limp bag in between each tie. They are left in the water for 3‐ 4 days and used to catch flounders and monkfish. Bottom gill nets in New England are set in relation to changes in bottom topography or bottom type where fish are expected to congregate. Other species caught in bottom gill nets in New England are spiny dogfish, and skates.

In the Mid‐Atlantic, sink gill nets are fished singly or in strings of just 3‐4 nets. The Mid‐ Atlantic fishery is more of a “strike” type fishery in which nets are set on schools of fish or around distinct bottom features and retrieved the same day, sometimes more than once. They catch species such as bluefish (Pomatomus saltatrix), Atlantic croaker (Micropogonias undulates), striped bass (Morone saxatilis), spot (Leiostomus xanthurus), mullet (Mugii spp.), spiny dogfish (Squalus acanthias), smooth dogfish (Mustelus canis), and skates (Leucoraja ocellata, Leucoraja erinacea, Raja eglanteria, Leucoraja garmani).

4.3 Habitat component descriptions Fish habitats are considered to be comprised of three distinct components: geological, biological, and prey. Each of these three components offer functional value as fish habitat, providing the waters and substrates necessary for food, shelter and growth to maturity for managed species. Each of these habitat components are subdivided into features to facilitate evaluation of their susceptibility to and recovery from fishing gear impacts. These features are selected based on their (1) importance to managed species and (2) the potential for differential effects from fishing gears. These components feed directly into the matrix‐based assessment described later in this document.

The following sections describe the habitat features evaluated in the Vulnerability Assessment, highlighting (1) the characteristics of the features that would likely influence their susceptibility and recovery values, (2) the importance of natural disturbance (i.e. high or low energy environment) in creating or maintaining features, and (3) the use of the features by managed species.

4.3.1 Geological Geological habitat features include non‐living seafloor structures that provide shelter for managed species (Table 8). They are grouped by substrate class. These features are created and maintained via either physical oceanographic processes or benthic organisms.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 34 DRAFT April 17, 2009 Table 8 ‐ Geological habitat component classes, subclasses, and features. Asterisk (*) indicates level used for mapping purposes Substrate Substrate class subclass Feature Feature description Featureless Biogenic depressions Fish, crab, lobster depressions Clay‐silt Biogenic burrows Fish, crab, lobster burrows Special case biogenic burrows Clay pipes, tilefish burrows Bedforms Ripples Mud* Featureless Biogenic depressions Fish, crab, lobster, scallop depressions Muddy‐sand Biogenic burrows Fish, crab, lobster burrows Special case biogenic burrows Clay pipes, tilefish burrows Bedforms Megaripples, dunes Featureless Sand* Biogenic depressions Fish, crab, lobster, scallop depressions Bedforms Megaripples, dunes Granule‐pebble Granule‐pebble* Granule‐pebble pavement Cobble Cobble* Cobble pavement Cobble piles Boulders Boulder* Piled boulders

The following sections further describe the geological habitat features listed above, including any characteristics likely to influence their susceptibility to and their recovery from fishing gear effects or natural disturbance.

Biogenic structures Biogenic structures in the substrate include shallow depressions and deeper burrows. These may be formed and used by fishes or crustaceans (Stanley 1971, Auster et al. 1995).

Tilefish create complex burrows in soft sediments. Various authors, including Twichell et al. (1985), Able et al. (1982, 1993), Grimes et al. (1986, 1987), and Cooper et al (1987), have studied the burrows and their use by the tilefish; this research is summarized in Steimle et al. 1999. Tilefish burrow may be tubular or funnel shaped. They range in size, but the largest are up to 5 meters wide and several meters deep. It is believed that either tilefish (Grimes et al. 1986, 1987) or crustaceans (Grimes et al. 1986, 1987, Cooper et al. 1987) form the burrows initially. The burrows may be created over the lifetime of the tilefish (Twichell et al. 1985); the maximum observed ages for female and male tilefish respectively are 46 and 39 years (Nitschke 2006).

Clay pipes (or pipe clay) are hard, hollow tubes or more irregularly shaped sedimentary structures that are likely originally created by crabs, fish, or burrowing anemones and then eroded from the seabed (Valentine 2001). They generally are found in muddy habitats, but they can be found in settings where sand and gravel overlie mud or clay (P. Valentine, personal communication).

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 35 DRAFT April 17, 2009 Gear effects on these features include filling and crushing. As they are of biological origin, recovery depends on the continued presence of the organisms that created the features, with timing dependent on the complexity of the feature: shorter for depressions, longer for burrows, and longest for tilefish burrows. Clay pipes will not recover on management timescales.

Bedforms From smallest to largest, sedimentary bedforms include ripples, megaripples, and waves. These are formed by the action of waves and tides over the seabed. Twichell (1983) defines these features by size (Table 9).

Table 9 ‐ Bedform classification (after Twichell 1983) Bedform Wavelength Height Featureless seafloor none None Ripple < 0.6 m Megaripple 1‐15 m Less than 1 m Wave 50‐1000 m 1‐25 m

‘Featureless’ clay‐silt, muddy‐sand, or sand can also be considered a bedform in the sense that the lack of seabed relief is related to the physical oceanography of the area.

Gravel and gravel pavements Areas where pebble or cobbles completely cover the seabed are considered pavement. Possible gear effects on pavements are the burial of pebbles or cobbles in underlying softer substrates.

Piles Larger size classes of gravel (i.e. cobbles and boulders) may be aggregated into piles on the seafloor. Gear effects on these piles include smoothing or displacement.

4.3.2 Biological Biological habitat features include organisms that provide structure for managed species. They are mostly , although macroalgae and sea grasses are also identified as important providers of structure. There is little overlap between biological and prey habitat components (see next section), with the exception of anemones, amphipods, and mollusks. Biological habitat components are grouped at higher taxonomic levels (phylum, order, class), although individual species common to the region are noted below (Table 10). The example species lists and regional descriptions are developed from the fishing impacts and ecological literature; they are by no means exhaustive.

During earlier stages of the Vulnerability Assessment, a structure/life history‐based list of biological habitat components is considered. This list would have organized example species in terms of whether they are erect or encrusting, soft or hard, and r‐selected or k‐selected, based on the theory that these attributes would influence sensitivity to fishing gears. Susceptibility and recovery values for organisms with these characteristics would then have been generated based

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 36 DRAFT April 17, 2009 on the scientific literature. However, while attempting to implement this approach, it became clear that susceptibility and recovery values are being considered based largely on the example species in each structure/life history category. Thus, a taxon, rather than structure/life history‐ based approach is determined to be more straightforward. In addition, a taxon‐based approach is more consistent with the literature, which tends to group structure‐forming organisms in this way (typically at family and higher levels). In the case of sponges, anemones, bryozoans, and macroalgae, a distinction is made between emergent and burrowing or encrusting forms, as shown below.

Table 10 ‐ Biological habitat components Phylum Feature (taxon evaluated) Habit Example species Sponges Emergent Haliclona oculata, Isodictya palmate Porifera Sponges Encrusting Polymastia robusta Hydroids Emergent Sertularia cupressina, Sertularia argentea Anemones Emergent Anemones Emergent, burrowing Primnoa resedaeformis, Paragorgia Soft corals: Gorgonians and soft Cnidaria Emergent arborea, Gersemia rubiformis, Alcyonium corals digitatum Sea pens Emergent Pennatula aculeate, Stylatula elegans Hard corals: stony corals and black Encrusting Astrangia poculata corals Annelida Colonial tube worms Emergent Filograna implexa Arthropoda Tube‐building amphipods Emergent , Modiolus Bivalves Emergent modiolus Ectoprocta Bryozoans Emergent Flustra foliacea Bryozoans Encrusting Brachiopoda Brachiopods Emergent Chordata Tunicates Emergent Boltenia ovifera Macroalgae Emergent (leafy) Macroalgae Encrusting Sea grass Emergent

The following sections describe each feature, including the taxonomic bounds of the group and example species, and any attributes that would contribute to the feature’s susceptibility and recovery values. Such attributes include habit, growth rate, and reproductive biology.

Sponges ‐‐To be completed‐‐

Hydroids ‐‐To be completed‐‐

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 37 DRAFT April 17, 2009 Anemones ‐‐To be completed‐‐

Deep‐sea corals ‐‐To be completed‐‐ Soft corals Hard corals

Colonial tube worms ‐‐To be completed‐‐

Tube‐building amphipods ‐‐To be completed‐‐

Bivalves ‐‐To be completed‐‐

Bryozoans ‐‐To be completed‐‐

Brachiopods ‐‐To be completed‐‐

Tunicates ‐‐To be completed‐‐

Macroalgae ‐‐To be completed‐‐

Sea grass ‐‐To be completed‐‐

4.3.3 Prey Prey habitat components include infaunal and epifaunal invertebrate prey commonly consumed by managed species. This list was developed using the Essential Fish Habitat Source Documents.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 38 DRAFT April 17, 2009 Table 11 ‐ Prey habitat components Phylum Feature (taxon evaluated) Example species Cnidaria Anemones Amphipods Isopods Arthropoda Mysids Decapod crabs Decapod shrimp Annelida Polychaetes Mollusca Mollusks Brittle stars Ophiopholis aculeata Strongylocentrotus droebachiensis Echinodermata Sea urchins and sand dollars

Sea stars Asterias spp.

The following sections describe each prey feature, including the taxonomic bounds of the group and example species, and any attributes that would contribute to the feature’s susceptibility and recovery values. Such attributes include habit, growth rate, and reproductive biology. Associations between specific features and substrates are also described.

Anemones ‐‐To be completed‐‐

Crustaceans ‐‐To be completed‐‐. Will include amphipods, isopods, mysids, decapod crabs, and decapod shrimp.

Polychaetes ‐‐To be completed‐‐

Mollusks ‐‐To be completed‐‐

Brittle stars ‐‐To be completed‐‐

Sea urchins and sand dollars ‐‐To be completed‐‐

Sea stars ‐‐To be completed‐‐

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 39 DRAFT April 17, 2009 4.4 Gear impacts literature review For each of the habitat features described above, susceptibility to and recovery from fishing gears is scored based on information found in the scientific literature, to the extent possible. A list was compiled containing over 400 candidate studies. This was narrowed to include only those studies deemed relevant to the Vulnerability Assessment. Review papers, studies that examined gear types very different from those used in the Northeast Region, or studies conducted in habitats very different from those found in the Northeast region were not evaluated. The literature reviewed included mostly peer‐reviewed scientific literature and peer‐reviewed conference papers. In addition, the literature included a few reports and one thesis. In total, 82 studies were evaluated.

Due to the large number of studies and the specificity of the various habitat features, a database was constructed to catalog the gear types and habitat features examined in each study. The database included fields to code for study design, relevance and appropriateness, gear type, energy environment, whether recovery was addressed, depth, and which of the various habitat features were evaluated (Table 12). The database was constructed in Microsoft Access, and analysts interacted with the database via a form (Figure 3).

Table 12 ‐ Literature review database fields Database field Coding options Coding guidelines Study design Choice of (1) comparative or (2) Comparative refers to studies that assessed impacts to fished and unfished experimental areas, or to studies that observed the effects of gear in fished areas. Experimental refers to studies that either: (1) evaluated the experimental use of fishing gear in comparison with an unfished control, or (2) used a before‐after control‐impact design to study the effects of either experimental use of fishing gear or actual fishing effort. Study relevance (1) Similar gears or habitats but All studies reviewed used or observed the effects of gears fairly similar to geographically remote study area (2) those used in the Northeast U.S. in temperate habitats. A score of (1) would Geographically similar (though non‐ indicate that the study meets these basic criteria. A score of (4) would NE) study area, similar gears/habitats indicate that they study is conducted in Northeast U.S. waters and (3) Study area overlaps with NE area evaluated the impacts of Northeast U.S. gear types. Values of (2) and (3) fall (incl. CA side of Georges) and uses between these two extremes. similar gears (4) Study performed in NE area with NE gears Study appropriateness (to Choice of study (1) tangentially Regardless of relevance, studies that specifically examine the effects of Vulnerability Assessment) supports, (2) supports, or (3) is particular gear types on particular habitat components should receive the perfectly aligned with the highest appropriateness values. Studies that are more general, perhaps vulnerability assessment aggregating multiple gear types or impacts, or that do not provide clear information on the substrate, depth, energy, or other background, would receive lower values. Evaluation of recovery could help a study achieve a higher score. Gear type One or more of the following: Multiple gear types could be checked as applicable, with details summarized groundfish trawl, shrimp trawl, squid in the comments section. trawl, raised footrope trawl, pelagic trawl, New Bedford scallop dredge, surf clam/ocean quahog dredge, lobster trap, deep‐sea red crab trap, longline, gillnet Energy High, low, both, or not specified Both high and low could be checked if applicable to multiple study sites. An energy environment was only checked if stated by the authors or could be reasonably inferred. Depth Choice of four ranges: 0‐50m, 51‐ Additional space provided for minimum and maximum study depths 100m, 101‐200 m, deeper than 200m

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 40 DRAFT April 17, 2009 Database field Coding options Coding guidelines Recovery addressed True/false Did the study address the recovery of habitat components from disturbance? Deep‐sea coral habitats True/false True indicates that the study referred to any deep‐sea coral species, whether impacts to corals are evaluated separately or if they are simply mentioned as a biological habitat component in the study area. In the Northeast, deep‐sea corals include five Anthozoan orders: Scleratinia (stony corals), Octocorallia (soft corals), Antipatharia (black corals), Gorgonacea (sea fans), and Pennatulacea (sea pens). Geological habitat True/false This section included both substrate subclasses and substrate features as components shown in Table 8, including an additional checkbox for geochemical effects. Subclasses and features are noted as either referenced (i.e. present) or evaluated. ‘Geological_habitat_components’ was checked when the study assessed impacts to substrate subclasses or features. Biological habitat True/false This section included biological habitat component taxa as shown in Table components 10. ‘Biological_habitat_components’ was checked if fishing effects to the various taxa were studied. Prey habitat components 256 character text box This section included prey habitat component taxa as shown in Table 11. In contrast to biological habitat components, ‘Prey_habitat_components’ was checked if various prey species were mentioned in the study. Comments 256 character text box This section was used to detail any gear or energy characteristics, note the study location, summarize key results, and note concerns or caveats as to the usefulness of the study for the Vulnerability Assessment.

Figure 3 ‐ Literature review database form. Data field descriptions provided in Table 12.

With few exceptions, one analyst coded each study. There was limited overlap (8 studies) between the team of analysts and the pool of study authors. After initial coding, the process was reevaluated and updates were made to the database form. The second, final, version is shown above.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 41 DRAFT April 17, 2009 Each of the three major habitat components was scored somewhat differently. A particular biological feature was only checked if gear impacts to that features were assessed, while a particular prey feature was checked if it was mentioned in the study. For geological features, a distinction was made between whether features were simply present in the study area or if impacts to the features were evaluated. Because the base grid for the spatial model is predicated on the substrate subclasses (see section 6.1.1), it is important to know which subclasses occurred in the study area, regardless of whether impacts to geological features were evaluated explicitly.

Coding of most fields was decided to be fairly straightforward and objective, with the exceptions of study relevance and study appropriateness, which might be somewhat subjective. However, these two fields are only used to characterize the literature used for the Vulnerability Assessment, and are not directly used for the matrix evaluations described later in this document. Study relevance averaged 2.3, with a standard deviation of 1.2. A score of 2 indicated that the study occurred in similar habitats with similar gears. Study appropriateness averaged 2.0, with a standard deviation of 0.8. A score of 2 indicated that the study supported the Vulnerability Assessment. Twenty two studies received a relevance score of 4 (Northeast gears, Northeast habitats), 27 studies received an appropriateness score of 3 (perfectly aligned with Vulnerability Assessment), and 11 received the highest values for both metrics.

Recovery was addressed by 37 of the 82 studies. The literature was evenly split between experimental studies and comparative studies. Impacts to geological habitat components were addressed by 34 studies; 44 addressed impacts to biological habitat components and 59 addressed impacts to prey habitat components. Most of the literature is fairly recent: all but nine studies were published in 1990 or later, and 56 of the 82 studies were published in 2000 or later.

Of the gear types evaluated in the Vulnerability Assessment, sink gill net and demersal longline gear impacts on habitat are not addressed by the literature. By author/year and gear type, the habitat components evaluated, as well as whether or not the study addressed recovery, are shown in Table 13.

Table 13 ‐ Studies reviewed by gear type(s), substrate type(s), and habitat component(s) evaluated. Habitat components are denoted as G (geological), B (biological), and P (prey). Substrate classes are denoted as M (mud), S (sand), and G (gravel). Studies are listed alphabetically. Citation # Habitat Recovery Otter trawl New Bedford Surf‐clam Lobster and components evaluated substrate scallop ocean deep sea red evaluated classes dredge quahog crab traps (G, B, P) evaluated substrates substrates substrates (M, S, G) evaluated evaluated evaluated (M, S, G) (M, S, G) Asch and Collie 2007 404 B, P R G G ‐ ‐ Auster et al. 1996 11 G, B, P ‐ M, S, G M, S, G ‐ ‐ Ball et al. 2000 17 G, P R M ‐ ‐ ‐ Bergman and VanSantbrink 2000 21 P ‐ M, S ‐ ‐ ‐ Blanchard et al. 2004 24 P ‐ M ‐ ‐ ‐

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 42 DRAFT April 17, 2009 Citation # Habitat Recovery Otter trawl New Bedford Surf‐clam Lobster and components evaluated substrate scallop ocean deep sea red evaluated classes dredge quahog crab traps (G, B, P) evaluated substrates substrates substrates (M, S, G) evaluated evaluated evaluated (M, S, G) (M, S, G) Boat Mirarchi and CR Env. 2003 408 G, B, P ‐ M, S ‐ ‐ ‐ Boat Mirarchi and CR Env. 2005 409 G, B, P ‐ M, S ‐ ‐ ‐ Brown et al. 2005a 34 B, P ‐ S ‐ ‐ ‐ Brown et al. 2005b 35 G ‐ S ‐ ‐ ‐ Caddy 1968 42 B, P ‐ ‐ M, S ‐ ‐ Caddy 1973 43 B, P ‐ ‐ S, G ‐ ‐ Collie et al. 1997 69 G, B, P ‐ S, G S, G ‐ ‐ Collie et al. 2000 70 B, P ‐ S, G S, G ‐ ‐ Collie et al. 2005 71 G, B, P R S, G S, G ‐ ‐ De Biasi 2004 88 G, P R M ‐ ‐ ‐ de Juan et al. 2007a 89 P ‐ M ‐ ‐ ‐ de Juan et al. 2007b 90 P ‐ M ‐ ‐ ‐ DeAlteris et al. 1999 92 G R M, S ‐ ‐ ‐ Dellapenna et al. 2006 406 G, P R M ‐ ‐ ‐ Drabsch et al. 2001 97 G, P ‐ M, S ‐ ‐ ‐ Engel and Kvitek 1998 101 G, B, P ‐ M, S, G ‐ ‐ ‐ Eno et al. 2001 102 B ‐ M, G ‐ ‐ Freese 2001 110 B R G ‐ ‐ ‐ Freese et al. 1999 111 G, B, P ‐ G ‐ ‐ ‐ Frid et al. 1999 113 P ‐ M, S ‐ ‐ ‐ Gibbs et al. 1980 119 G, P ‐ M, S ‐ ‐ ‐ Gilkinson et al. 1998 120 G, P ‐ S ‐ ‐ ‐ Gilkinson et al. 2003 121 G ‐ ‐ ‐ S ‐ Gilkinson et al. 2005a 122 B, P ‐ ‐ ‐ S ‐ Gilkinson et al. 2005b 123 B ‐ ‐ ‐ S ‐ Gordon et al. 2005 128 G, P R S ‐ ‐ ‐ Hall et al. 1990 140 B, P R ‐ ‐ S ‐ Hall et al. 1993 141 B, P ‐ S ‐ ‐ ‐ Hansson et al. 2000 149 P ‐ M ‐ ‐ ‐ Henry et al. 2006 157 B ‐ G ‐ ‐ ‐ Hermsen et al. 2003 158 G, B, P ‐ S, G S, G ‐ ‐ Hixon and Tissot 2007 164 B, P ‐ S, G ‐ ‐ ‐ Kaiser et al. 2000 184 S, G ‐ ‐ ‐ Kenchington et al. 2001 192 P ‐ S ‐ ‐ ‐ Kenchington et al. 2005 193 P ‐ G ‐ ‐ ‐ Kenchington et al. 2006 194 B, P ‐ G ‐ ‐ ‐ Knight 2005 203 B, P ‐ M, S, G ‐ ‐ ‐ Koulouri et al. 2005 211 P ‐ M ‐ ‐ ‐ Kutti et al. 2005 214 B, P ‐ S, G ‐ ‐ ‐ Langton and Robinson 1990 217 B, P ‐ ‐ S,G ‐ ‐ Lindholm et al. 2004 225 G, B, P ‐ S S ‐ ‐ Link et al. 2005 228 B, P ‐ S, G S,G ‐ ‐ MacKenzie 1982 232 G, B, P ‐ ‐ ‐ S ‐

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 43 DRAFT April 17, 2009 Citation # Habitat Recovery Otter trawl New Bedford Surf‐clam Lobster and components evaluated substrate scallop ocean deep sea red evaluated classes dredge quahog crab traps (G, B, P) evaluated substrates substrates substrates (M, S, G) evaluated evaluated evaluated (M, S, G) (M, S, G) Mayer et al. 1991 236 G ‐ M ‐ ‐ ‐ McConnaughey et al. 2000 238 B ‐ S ‐ ‐ ‐ McConnaughey et al. 2005 239 B, P ‐ S ‐ ‐ ‐ Medcof and Caddy 1971 244 G R ‐ ‐ S ‐ Meyer et al. 1981 245 G R ‐ ‐ S ‐ Morais et al. 2007 247 G, B, P ‐ M, S. G ‐ ‐ ‐ Moran and Stephenson 2000 248 G, B ‐ S ‐ ‐ ‐ Morello et al. 2005 249 P ‐ ‐ ‐ S ‐ Murawski and Serchuk 1989 256 G, P R ‐ ‐ S ‐ Nilsson and Rosenberg 2003 407 G ‐ M ‐ ‐ ‐ Palanques et al. 2001 277 G R M ‐ ‐ ‐ Pilskaln et al. 1998 283 G ‐ M ‐ ‐ ‐ Pranovi and Giovanardi 1994 287 G, B R ‐ ‐ S ‐ Prena et al. 1999 291 G, B, P ‐ S ‐ ‐ ‐ Rosenburg et al. 2003 313 G ‐ M ‐ ‐ ‐ Sanchez et al. 2000 320 G, B, P R M ‐ ‐ ‐ Schwinghamer et al. 1998 325 G, P R S ‐ ‐ ‐ Sheridan and Doerr 2005 330 G, P ‐ M,S ‐ ‐ ‐ Simpson and Watling 2006 333 G, P ‐ M ‐ ‐ ‐ Smith et al. 1985 334 G, P R M, S ‐ ‐ ‐ Smith et al. 2000 335 G, B, P R M ‐ ‐ ‐ Smith et al. 2003 336 G ‐ M ‐ ‐ ‐ Sparks‐McConkey and Watling 2001 338 G, P R M ‐ ‐ ‐ Stokesbury and Harris 2006 352 G, B, P ‐ ‐ S, G ‐ ‐ Stone et al. 2005 355 G, B, P ‐ S ‐ ‐ ‐ Sullivan et al. 2003 359 G, B, P R ‐ ‐ ‐ ‐ Tanner 2003 360 G, B R S ‐ ‐ ‐ Tuck et al. 1998 372 G, P R M ‐ ‐ ‐ Tuck et al. 2000 373 G, B, P R ‐ ‐ S ‐ Van Dolah et al. 1987 382 B, P R S ‐ ‐ ‐ Wassenberg et al.l 2002 387 B ‐ S, G ‐ ‐ ‐ Watling et al. 2001 391 G, P R ‐ S ‐ ‐

The following sections provide a short summary of each study reviewed, and also describe generally the body of literature related to each gear type. This review builds on but is distinct from NOAA Technical Memorandum NMFS‐NE‐181 and subsequent updates. The studies are grouped by gear type (demersal otter trawl, New Bedford‐style scallop dredge, surf clam/ocean quahog dredge, or lobster trap). Although demersal otter trawls were disaggregated for the purpose of evaluating susceptibility and recovery (Section 4.5), the literature did not generally consider the four trawl types separately, with the exception of eight studies that used shrimp trawls and one study that used a raised footrope trawl.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 44 DRAFT April 17, 2009 The summary table for each gear type includes study information, study approach, impacts/effects, and recovery, where evaluated. Specifically, the study information column gives study location, depth, substrate, energy, and habitat components evaluated. The substrate is given as the Vulnerability Assessment substrate class (mud, sand, and/or gravel) followed parenthetically by the authors’ substrate description. The energy (high or low) is either specified by the authors or could be reasonably inferred based on knowledge of the location’s physical oceanography. Evaluated refers to the three endpoint matrices used in the VA: geological, biological, and prey.

The study approach column defines the work as comparative, experimental, or observational. Although observational is not coded in the database, this level of detail is added in this phase of the review. During the database evaluations, observational studies are coded as comparative. The study approach column also provides a brief description of the methods. Some comparative studies evaluated mobile gear impacts generally, such that the impacts could not be attributed solely to either otter trawls or scallop dredges. These studies are included in both the otter trawl and scallop dredge tables, and the use of multiple gear types in the study area is noted. If specific details about gear configuration are provided, these are also noted in this column.

The gear effects/impacts column describes either effects (slicing, crushing, etc.) or impacts (decrease in epifaunal abundance, increase in the number of opportunistic species, etc.), generally ordered as geological, biological, prey. Statistically significant findings are denoted with an ‘S’.

If recovery is evaluated, the recovery column describes briefly the extent to which various habitat components did or did not recover, including the time steps at which recovery is assessed. Generally speaking, recovery is infrequently evaluated, and when it was, the timescales varied widely.

A common feature of this literature is the reporting of multiple distinct field studies within one publication; where this has occurred, the study components are denoted with (A), (B), (C), etc. in the tables. In other cases, multiple publications shared authors, study sites, and/or data; these are combined into one row with the various publications denoted as (A), (B), and (C). For many of these studies, the results are more complex than could be summarized here; these summaries served as a tool during matrix‐based sensitivity evaluations but did not substitute for close reading of the literature.

4.4.1 Demersal otter trawls This section presents the gear impacts literature relevant to demersal otter trawls (Table 14). Although the otter trawl gear category is disaggregated for the Vulnerability Assessment into groundfish, shrimp, raised footrope, and squid trawl types, all references are combined in the table, with any details about gear configurations noted in the ‘study approach’ column.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 45 DRAFT April 17, 2009 Out of 81 studies reviewed, 62 evaluated the impacts of various otter trawl configurations, with eight of these related specifically to shrimp trawls (Dellapenna et al. 2006; Drabsch et al. 2001; Gibbs et al. 1980; Sheridan and Doerr 2005; Simpson and Watling 2006; Sparks‐McConkey and Watling 2001; Tanner 2003; and Nilsson and Rosenberg 2003). Moran and Stephenson (2000) compared a raised‐footrope trawl with a groundfish trawl. No studies examined the effects of squid trawls.

By substrate class, 27 groundfish trawl studies were conducted in mud habitats, with 15 of these addressing impacts to geological habitat components. Disaggregating mud into the clay‐silt and muddy‐sand subclasses, 17 studies were conducted in silt‐clay habitats and 18 in muddy‐ sand habitats, with 11 and 8 studies respectively evaluating impacts to geological habitat components. Thirty‐four groundfish trawl studies were conducted in sand habitats; 19 examined impacts to geological habitat components. Nineteen groundfish trawl studies are conducted in gravel habitats; 9 of these examined impacts to geological habitat components. Disaggregating gravel into the granule‐pebble, cobble, and boulder subclasses, 18 study sites contained granule‐pebble (9 examined impacts), 13 contained cobble (5 examined impacts), and 9 contained boulder (5 examined impacts). Many study areas contained multiple substrate classes.

All of the shrimp trawl studies evaluated impacts to geological habitat components and were conducted in mud habitats (or a combination of mud, sand, and gravel) except for Tanner (2003), which was conducted in sand habitat.

A total of 31 otter trawl studies (30 groundfish, 1 shrimp) examined the effects of fishing on biological habitat components. Effects of fishing on prey habitat components were examined in 46 studies (40 groundfish, 6 shrimp). With the exception of biological habitat impacts of shrimp trawls on mud substrates, these studies covered all three substrate classes (mud, sand, gravel) for both gear types.

Eight studies (Asch and Collie 2007; Auster et al. 1996; Collie et al. 1997; Collie et al. 2000; Collie et al. 2005; Hermsen et al. 2003; Lindholm et al. 2004; and Link et al. 2005) considered fishing gear impacts in areas that had been fished by both otter trawls and New Bedford‐style scallop dredges.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 46 DRAFT April 17, 2009 Table 14 ‐ Impacts of otter trawls to geological, biological, and prey habitat components on mud, sand, and gravel substrates. (Where multiple field studies are included in one publication, or where multiple publications are summarized together, individual components are denoted as (A), (B), (C), etc. ‘S’ indicates statistical significance) Study information Study approach Effects/Impact Recovery Ref #: 404 Comparative. Multiple gear types At shallow sites, cover of all epifauna except At shallow sites, several Citation: Asch and Collie (2007) fished in study area (see New hydroids S differed by disturbance regime. taxa showed changes in Location: Northern edge, eastern Bedford scallop dredge). Still Sponges and bushy bryozoans showed S abundance beginning 2 Georges Bank, U.S. and Canada photographs (N=454) are analyzed higher % cover at undisturbed sites, years after closed area Depth: 40‐50, 80‐90 m for percent cover of colonial encrusting bryozoans and Filograna implexa established. Increase in Substrate: Gravel (pebble and epifauna and the abundance of showed S higher % cover at disturbed sites. abundance of P. cobble pavements with some non‐colonial organisms at shallow Also at shallow sites, generally S between magellanicus, Pagurus overlying sand) and deep disturbed and year variations. At deep sites, % cover of F. spp., S. droebachiensis, Energy: High undisturbed sites in and around implexa and hydroids is S higher in Asterias spp. between Evaluated: Biological, prey Georges Bank Closed Area II. undisturbed areas; other taxa showed no closure (1994) and differences by disturbance regime. For non‐ 2000. colonial epifauna, depth contributed more to differences in species composition than disturbance. Higher spp. richness at undisturbed sites (S at shallow sites). Ref #: 11 Amount of fishing effort and types (A) In cobble habitat (N=12‐13 transects per Not addressed Citation: Auster et al. 1996 of mobile gear used in study areas treatment), S lower cover of emergent Location: 3 sites, Gulf of Maine, not well defined. See New Bedford epifauna, sea cucumbers in fished area; in USA scallop dredge. Three sites: sand habitat (N=17‐18 transects per Depth: See study approach (A) Swans Island: closed 10 yr, sand treatment), S lower cover of sea cucumbers Substrate: See study approach and cobble, depth not specified, and biogenic depressions in fished area. Energy: Not specified comparative: inside‐outside video (B) Qualitative; loss of mud veneer, Evaluated: Geological (B), transects reduction in epifaunal species, incl. sponges biological (A,B,C), prey (A,B) (B) Jeffreys Bank: boulders (quantified but no statistical tests), prevented fishing, then fishing, movement of boulders gravel and mud, depth 94m, (C) Positive relationship between hydrozoan comparative: one pair of before/6 Corymorpha penduala and shrimp in 1993, years after submersible dives fewer areas with hydrozoans and wide (C) Stellwagen Bank: daily fishing distribution of tunicate Molgula arenata in evidenced by trawl/dredge tracks, 1994 gravel and sand, depth 32‐43m, observational: n=4 (?) video transects over 2 years Ref #: 17 (A) Experimental. Used (A) Number of species, species richness, (A) Not addressed. Citation: Ball et al. (2000) ‘commercial pattern trawl’ to diversity, and biomass declined 24 hr after (B) Physical disturbance Location: (A) Irish Sea (B) Loch impact heavily fished (75 m) and fishing at the lightly fished site. No of seabed could be Gareloch, Firth of Clyde, Scotland lightly fished (35 m) Nephrops difference before and after fishing could be identified 5 but not 18 Depth: (A) 35m, 75 m (B) not grounds and measured benthos determined for the heavily fished site. The months after fishing. specified with grab samples before and 24 h heavily fished site is more different from its Some biological Substrate: (A,B) Mud (sandy‐silt) after. Also compared grabs to unfished control than the lightly fished site differences between Energy: (A,B)Low unfished sites (near wrecks) at both from its control. experimental and Evaluated: (A) Prey, (B) Geological, depths to determine medium‐term (B) Over time, increase in number of species control sites remained prey impacts. and individuals, no difference in biomass, after 18 mo. (B) Experimental. Used decrease in species diversity and evenness. ‘commercial pattern trawl’ monthly Increase in abundance of ‘opportunistic’ for 60 mo in an area closed to species. fishing for 25 y. Compared fished and unfished areas using infaunal sampling, underwater television, and acoustic sampling.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 47 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref #: 21 Experimental trawling (one‐ and‐a‐ High (20‐50%) mortalities for six sedentary Not evaluated Citation: Bergman and van half tows per unit of area) in and/or immobile megafaunal (>1 cm) Santbrink (2000) commercially trawled area; effects species, <20% for 10 others, from a single Location: Southern North Sea, assessed after 24‐48 hr. pass of the trawl; S effects on 11 of 54 Dutch coast occasions. Depth: 20‐45m Substrate: Mud, sand (silty sand and sand) Energy: ‐ Evaluated: Prey Ref #: 24 Comparative. Tested hypothesis Species diversity and the largest body mass Not evaluated Citation: Blanchard et al. (2004) that high fishing effort (1) reduces class of invertebrates are smaller in H areas Location: Bay of Biscay, France diversity and evenness, (2) reduces than in M ones. H areas characterized by a Depth: 106‐129 m the observed maximum body mass, comparatively large biomass of small Substrate: Mud (muddy sand and (3) favors a few body mass classes, invertebrates. Dominant species in M areas sandy mud (10‐35% silt)) (4) increases the steepness of the are a commercial crustacean species and a Energy: Low slope of number–size spectra, (5) disturbance‐sensitive echinoderm; in H Evaluated: Prey shifts abundance and biomass areas an opportunistic crustacean and distributions among species toward carnivores of minor or no commercial those typical of a disturbed interest. No fragile species are found in the community, and (6) changes H areas, whereas 6 fragile species are found species composition by sampling in the M areas. benthic megafauna with a 2 m beam trawl at fishing grounds (8 stations within 4 areas) subject to heavy (H) and moderate (M) levels of exploitation. 60% of effort in areas by Nephrops twin bottom trawls, fish twin bottom trawls, fish bottom otter trawls, and sole twin bottom trawls. Ref #: 408, 409 (A) Mud Hole. Experimental. (A) Trawl doors caused deep furrows and Not evaluated. Citation: Boat Kathleen A. Mirarchi Evaluated immediate effects of 6 ridges, net sweep smoothed seabed and Inc. and CR Environmental Inc. repetitive tows in each of two lanes exposed polychaete tubes. No difference in (2003, 2005) within a heavily trawled area, with grain size attributable to chronic trawling. Location: Gulf of Maine, adjacent control lanes. No S immediate effects on benthic Massachusetts Coast, USA Experimental tows repeated a yr macrofaunal community, no difference in Depth: (A) 43 m (B) 36m later in same lanes to evaluate infaunal density, richness, or composition Substrate: (A) Mud (muddy sand) chronic effects. after experimental tows in yrs 1 or 2. (B) sand (B) Little Tow. Experimental. (B) Physical impacts of trawling less visible Energy: High Evaluated immediate effects of 6 than at (A); Little tow is more subject to Evaluated: Geological, biological, repetitive experimental tows in natural disturbance. No S immediate effects prey each of two lanes within a lightly on benthic macrofaunal community, no trawled area, with adjacent control difference in infaunal density, richness, or lanes. Experimental tows repeated composition after experimental tows in a year later in same lanes to years 1 or 2, no difference in grain size evaluate chronic effects. attributable to chronic trawling.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 48 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref #: 34, 35 (A) Experimental. Compared two (A) Fished area is characterized by reduced Not evaluated Citation: (A) Brown et al. (2005a) areas, one closed for 10 years and macrofaunal density, biomass, and richness; (B) Brown et al. (2005b) one open to commercial trawling, sessile taxa (e.g., polychaetes) are prevalent Location: Bering Sea and also examined the immediate in the closed area, and mobile scavengers Depth: 20‐30 m effects of experimental trawling on (e.g., amphipods) are more common in the Substrate: Fine sand benthic community structure in the fished area. Immediate responses of Energy: High closed area, using grab samples macrofauna to experimental trawling are Evaluated: (A) Biological, prey (B) and video transects. Initial subtle (i.e., reduced richness, absence of Geological responses of the benthic rare taxa, and patchy changes in assemblage community are determined within biomass), but no differences are detected 1 week. Experimental trawling is relative to controls for density, diversity, or over a 30 hour period and total biomass. consisted of ten tows within a 4 (B) Surficial sediments (top 3 cm) of the km2 block inside the closed area. fished area are slightly better sorted, less (B) Experimental. In same study variable, and contained fewer fine grains areas as Brown et al. (2005a), than those of the closed area, but study is examined the immediate effects of conducted during a period of low wave experimental trawling (see above) disturbance. and long‐term impacts of commercial trawling on sediment properties, and compared bottom trawling to natural disturbance by waves at the same study sites, using ROV video transects and push cores collected by divers. Ref #: 69, 70, 71 Comparative. Multiple gear types S higher total densities, biomass, and species (C) 5 years after fishing Citation: (A) Collie et al. 1997, (B) fished in study area (see scallop diversity in undisturbed sites, but also in eliminated from area Collie et al. 2000, (C) Collie et al. dredge). Benthic sampling, video, deeper water (i.e., effects of fishing could (Closed Area II), 2005 and still photos in 2 shallow (42‐47 not be distinguished from depth effects); 6 observed S shifts in Location: Eastern Georges Bank m) and 4 deep (80‐90 m) sites species abundant at U sites, rare or absent species composition and (Northern edge), U.S. and Canada disturbed (D) and undisturbed (U) at D sites; percent cover of tube‐dwelling S increases in Depth: 42‐90 m by trawls and scallop dredges. polychaetes, hydroids, and bryozoans S abundance, biomass, Substrate: Sand, gravel (pebble‐ higher in deepwater, but no disturbance production, and cobble “pavement” with some effect. epifaunal cover. overlying sand) Energy: High Evaluated: (A, C) Geological, (A,B,C) biological, prey Ref #: 88 Experimental. Trawling (14, 1 hr Door tracks; S increase in clay, decrease in Door tracks less distinct Citation: De Biasi (2004) tows in a 24 hr period at each silt at control sites located landward of after 48 hrs, almost Location: Tyrrhenian Sea station) in an unfished area (5 treatment sites suggest that trawling re‐ invisible after 1 month; (Mediterranean) stations), effects evaluated relative suspended and re‐distributed finer abundance of benthic Depth: 32‐34 m to landward and seaward control sediments. Some infaunal taxa S more infaunal taxa recovered Substrate: Mud (fine silt) sites immediately after, 24/48 hrs abundant at treatment sites after 48 hrs, within same time frame. Energy: ‐ after, and one month after trawling some less so (clearest changes for mollusks), Evalcuated: Geological, prey with sidescan sonar and box core but no S changes immediately after, 24 hrs samples. after, or 1 mo after disturbance.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 49 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref #: 90, 89 (A) Compared diets of two (A) Feeding of both species generally Not evaluated Citation: (A) de Juan et al. (2007a) epifaunal species – a starfish and a increased with fishing activity; trawling (B) de Juan et al. (2007b) flatfish – collected at fished and appeared to modify the relative abundance Location: Coast of Spain, unfished locations relative to the of ingested prey although increased Mediterranean Sea abundance of their prey. opportunism due to trawling disturbance is Depth: 30‐80 m (B) Comparative. Benthic taxa that not detected and the density of these Substrate: Mud contributed most to dissimilarity predators over the fishing grounds is more Energy: ‐ between a chronically trawled closely related to their vulnerability to Evaluated: Prey location and an unfished location trawling. (for 20 yrs) with similar habitat (B) The fished area had a higher abundance characteristics are grouped into of burrowing epifaunal scavengers and functional categories based on motile burrowing infauna, while the their expected response to fishing undisturbed area is characterized by higher disturbance; changes in functional abundance of surface infauna, epifaunal components of benthos are suspension feeders, and predatory fish; analyzed relative to seasonal patterns in the responses of the functional variability and variations in fishing groups to fishing are apparent despite the intensity during year‐long study. dominance of organisms not considered especially vulnerable to trawling. Ref #: 92 (A,B) Observational. Diver (A) Doors produced tracks 5‐10 cm deep and (A) Hand dug trenches Citation: DeAlteris et al. (1999) observations and modeling of adjacent berm 10‐20 cm high. Model lasted >60 days. Location: Narragansett Bay, Rhode bottom hydrodynamic and analysis indicated that sediment transport (B) Hand dug trenches Island, USA sediment transport processes. occurs <5% of the time in mud substrate. not visible after 1‐4 Depth: (A) 14 m, (B) 7m (B) No tracks found. Model analysis days. Substrate: (A) Mud (B) Sand indicated that sediment transport occurs Energy: High daily in sand substrate. Evaluated: Geological Ref #: 406 Experimental. Pre‐ and post‐trawl Trawl doors, net, and tickler chains excavate Plume settling and Citation: Dellapenna et al. (2006) sediment and water column seabed to maximum depth of 1.5 cm dispersion caused Location: Galveston Bay (Gulf of profiling in small, heavily fished (considerably less in most areas); fine suspended sediment Mexico) area (for shrimp) subjected to sediment suspended behind net an order of inventories to return to Depth: 2‐5 m experimental trawling (3 tows) on magnitude higher than before trawling and pre‐trawl values about Substrate: Mud two occasions. Shrimp trawl gear. comparable to amount produced by 9‐10 14 min after trawl Energy: High m/s wind event; no S increase in shear passage. Evaluated: Geological strength of sediment surface. Ref #: 97 (A,B) Experimental trawling (two (A,B) Trawl door tracks, smoothing of Not evaluated Citation: Drabsch et al. (2001) tows per unit of area in 1 day) in topographic features such as biogenic Location: Gulf of St. Vincent, South area with very little to no trawling mounds. Removal of 28% of epifauna Australia for 15 yrs; effects on infauna (unpublished results). No S impacts of Depth: 20 m evaluated before study and after 1 trawling on infauna when three sites are Substrate: (A) Mud (fine silt, their wk by analyzing sediment cores pooled site 2) (B) Sand (medium coarse collected by SCUBA divers. Shrimp (A) Change in infauna before/after trawling sand and shell, their sites 1 and 3) trawl gear with chain sweeps. when site 2 is examined individually at high Energy: High taxonomic resolution. Evaluated: Geological, biological, (B) No S impacts of trawling on infauna at focus on prey sites 1 and 3 individually; for site 3 greater variation in control areas. Ref #: 101 Comparative. Used a submersible S fewer rocks and biogenic mounds, S less Not evaluated Citation: Engel and Kvitek (1998) and grab samples (3 yr) to compare flocculent material, and S more exposed Location: California, USA lightly trawled (LT) and heavily sediment and shell fragments in HT area; Depth: 180 m trawled (HT) lower densities of large epibenthic taxa in HT Substrate: Mud, sand, gravel sites with same sediments and area (S for sea pens, starfish, anemones, and Energy: ‐ depth. sea slugs); higher densities of nematodes, Evaluated: Geological, biological, oligochaetes, brittle stars and one species of prey polychaete in HT area; no differences between areas for crustaceans, mollusks, or nemerteans.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 50 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref #: (A) 111 (B) 110 Experimental (A) with (A) Boulders displaced; groundgear left (B) Furrows still Citation: (A) Freese et al. 1999 (B) observational follow‐up (B). Video furrows 1‐8 cm deep in less compact prominent after 1 yr, no Freese 2001 observations from a submersible 2‐ sediment; layer of silt removed in more recruitment of new Location: Gulf of Alaska 5 hr after single trawl tows in area compact sediment; S reductions in sponges and no repair Depth: 206‐274 m exposed to little or no commercial abundance of sponges, anemones, and sea or re‐growth of Substrate: Gravel (93% pebble, 5% trawling for about 20 yr; towed and whips; damage to sponges, sea whips and damaged sponges, but cobble, 2% boulder) un‐towed sites re‐visited a year brittle stars. sponges that had been Energy: Low later knocked over, or pieces Evaluated: (A,B) Biological, (A) of sponge lying on Geological, prey bottom, are still viable. Ref #: 113 (A) Comparative. Examined (A) S increase in total number of individuals Not evaluated Citation: Frid et al. (1999) changes in benthic fauna in ‘positive‐response’ taxa. No effect of Location: Northeast England (North macrofauna in a heavily trawled increasing effort on total number of Sea) (A) heavily fished site (B) location. 27 yr study period individuals in ‘negative response’ taxa, lightly fished site divided into periods of low, high, except for S increase in number of errant Depth: (A) 80 m (B) 55m and moderate fishing activity. polychaetes and S decline in sea urchins Substrate: (A) Mud (silt‐clay) (B) Annual grab sampling of benthic (both Sand macrofauna. Macrofauna divided ‘negative response’ taxa). Energy: ‐ into taxa likely to respond (B) No correlation between total abundance Evaluated: Prey negatively and positively to of benthic macrofauna and fishing effort increases in fishing effort. (previous analysis of data indicated positive correlation with phytoplankton abundance). Ref #: 119 Comparative. Sampling before, Sediment plume; very little disturbance of Not evaluated Citation: Gibbs et al. (1980) immediately after, and 6 mos after seafloor; no consistent effects on benthic Location: Botany Bay, New South 1 wk of experimental trawling in a community diversity. Wales, Australia fished location; control area Depth: Shallow estuary located 200 km away. Shrimp trawl Substrate: Mud, sand (sand with 0‐ gear. 30% silt‐clay ) Energy: High Evaluated: Geological, prey Ref #: 120 Observed effects of commercial Trawl door created 5.5‐cm berm adjacent to Not evaluated Citation: Gilkinson et al. (1998) otter door model in test tank. 2‐cm furrow; bivalves displaced, but little Location: Test tank to simulate damage. Grand Banks of Newfoundland Depth: not applicable Substrate: Sand Energy: High Evaluated: Geological, prey Ref #: 128 Experimental. Three year study to Except for Chionoecetes opilio and All available evidence Citation: Gordon et al. (2005) examine repetitive otter trawling. Gorgonocephalus arcticus, direct removal of suggests that the Location: Grand Banks of Epibenthos sampled with sled, epibenthic fauna appeared to be biological community Newfoundland infauna sampled with large insignificant. Immediately after trawling, the recovered from the Depth: 120‐146m videograb. Repeated study #120 mean biomass of epibenthic organisms is annual trawling Substrate: Sand with live bivalves. Otter trawl with reduced by an average 24%. The most disturbance in a year or Energy: Low rockhopper gear (well specified). affected species are snow crabs, basket less, and no significant Evaluated: Geological, prey stars, Echinarachnius parma, Ophiura sarsi, effects could be seen on Strongylocentrotus pallidus and Gersemia benthic community spp. The immediate impacts of otter structure after 3 years trawling on the infauna appeared to be of otter trawling. minor and limited to a few species of polychaetes. Significant effects could not be detected on the majority of species found at the study site, including all mollusk species. Ref #: 141 Comparative. Sampled infauna at Abundance of infauna related to changes in Not evaluated Citation: Hall et al. (1993) increasing distance from a sediment type and organic content, not Location: North Sea shipwreck. Design assumed that distance from shipwreck. Authors thought Depth: 80 m effort would increase with that either natural linear bands of sediment Substrate: Sand (coarse sand) increasing distance from the or wreck‐induced concentric rings might Energy: ‐ wreck.. have been responsible for observed Evaluated: Biological (tube building patterns. amphipods), prey

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 51 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref #: 149 Experimental trawling for 1 yr (two Abundance of 61% infaunal species Not evaluated Citation: Hansson et al. (2000) tows per wk, twenty‐four tows per negatively affected and S reductions in Location: Fjord on the west coast unit of area) in area closed to abundance of brittle stars during last 5 mo of Sweden fishing for 6 yr (three treatment of disturbance period; S reductions in total Depth: 75‐90 m and three control sites); effects biomass at 3 of 3 trawled sites and 1 of 3 Substrate: Mud (clay) evaluated during last 5 mo of control sites, and in number of individuals at Energy: Low experiment. See refs #407, 313. 2 of 3 trawled sites and 1 of 3 control sites; Evaluated: Prey abundance of polychaetes, amphipods, and mollusks not affected. Ref #: 157, 193, 194 (A) Experimental study of trawling (A) Short‐term effects detected as NS Not evaluated Citation: (A) Henry et al. 2006 (B) impacts on colonial epifauna, decreases in number of taxa, total biomass, Kenchington et al. 2005 (C) videograb sampling up to 20 days and total hydroid biomass, with associated Kenchington et al. 2006 before and again 1‐5 days after changes in community composition; no Location: Western Bank (Scotian trawling along trawled and multiple cumulative effects, no differences on control shelf) control lines. and impacted lines at end of experiment. Depth: 70 m (B) Experimental trawling in area Effects are small relative to natural inter‐ Substrate: Gravel (pebbles or closed to bottom trawling for 10 yr; annual changes at control sites. cobbles overlaying medium to 12‐14 sets along same line (during (B) S changes in abundance (usually more in gravelly sand) 15‐19 hr) on three separate time 2) of prey consumed by haddock, cod, Energy: High occasions in 3 yrs, stomach winter flounder, plaice, and yellowtail and Evaluated: Biological (A,C), prey contents compared between first qualitative changes resulting from (B,C) two sets (time 1) and subsequent opportunistic feeding by cod, plaice, and YT sets (time 2). on prey made more available by trawling (C) Experimental trawling study. (i.e., infauna and epifauna living on or near Macro‐ and megafaunal data the sediment surface). collected with videograb samples (C) Trawling had few detectable immediate and a mounted camera system. effects on the abundance or biomass of individual taxa and none on community composition; a few taxa, primarily polychaetes and amphipods, decreased S after trawling, some because of scavenging by demersal fish. 15 taxa showed S decreases after trawling when the cumulative effects of trawling are considered; the species affected are primarily high turn‐over species, such as polychaetes and amphipods, and . Mussels, a tube‐dwelling polychaete, and a brachiopod are more visibly damaged by trawl gear than other species. Ref #: 158 Comparative. Multiple gear types Production remained markedly lower at See impacts/effects Citation: Hermsen et al. 2003 used in study area (see scallop shallow disturbed site over course of study Location: Northern edge, eastern dredge). Benthic macrofauna than at nearby recovering site, where it Georges Bank, U.S. and Canada sampled at deep and shallow sites increased over 12‐fold from before closure Depth: 47‐90 m disturbed and undisturbed (by to 5 yrs after closure; at deep sites, Substrate: Sand, gravel (pebble‐ fishing) using Naturalists dredge production remained S higher at cobble “pavement” with some with a 6.4 mm liner 8 times over 7 undisturbed sites. Sea scallops and sea overlying sand) yr period, 2 yrs prior to closure, just urchins dominated production at shallow Energy: High after closure, and 5 yrs after recovering site; a soft‐bodied tube‐building Evaluated: Geological, biological, closure. polychaete dominated production at the prey deep, undisturbed site. Ref #: 164 Submersible observations in a Trawl door tracks; fish (individuals and Not evaluated Citation: Hixon and Tissot (2007) trawled and an un‐trawled area species) more abundant on untrawled Location: Oregon Coast, USA (trawled area is shallower). bottom; benthic invertebrates six times Depth: 183‐361 m more numerous in untrawled area, but there Substrate: Mud are fewer taxa. Dominant fishes and Energy: Low macroinvertebrates on trawled seafloor are Evaluated: Biological, prey mobile scavengers.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 52 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref# 184 Comparative, multiple gear types No S effect of high fishing effort on numbers Not evaluated Citation: Kaiser et al 2000 used in study area. Sampled of infaunal or epifaunal species or Location: South Devon Coast, infaunal and epifaunal individuals; in high‐effort areas there were: England communities in areas of high, 1) a lower reduced abundance of larger, less Depth: 15‐70 m medium, and low fishing effort by mobile, and emergent epifauna; 2) a higher Substrate: Sand fixed and mobile gears; each area abundance of more epifauna; and 3) fewer Energy: ‐ with three sites (shallow, fine sand, high‐biomass species of epifauna and Evaluated: Biological, prey deep medium sand, and deep infauna; infauna in deeper coarse‐medium coarse‐medium sand). sand habitat most affected by fishing. Ref #: 192 Experimental trawling (3‐6 tows After trawling in 1995, the percentages of Benthic organisms that Citation: Kenchington et al. (2001) per unit of area) in closed area 1, 2, organic carbon and nitrogen in the surface are reduced in Location: Grand Banks, and 3 yrs after closure; lightly sediment S decreased. S short‐term abundance in 1994 Newfoundland exploited for >10 yrs; effects reductions in total abundance and recovered prior to the Depth: 120‐146 m evaluated within several hours or abundance of 15 infaunal and epifaunal taxa next sampling event one Substrate: Sand (fine to medium days after trawling and after 1 yr. (mostly polychaetes) in only 1 of 3 yr; no year later. grained sand) short‐term effects on biomass or taxonomic Energy: Low diversity; no long‐term effects. Evaluated: Geological, prey Ref #: 203 Comparative. Multiple gear types Although 2 open sites are S different in For epifauna, 4‐year Citation: Knight (2005) fished in study area, predominantly terms of the infaunal families present, they closed sites are more Location: Gulf of Maine otter trawls. Epifaunal are more similar to each other than to the 2 similar to the 6‐year Depth: 100‐130 m communities in closed areas closed sites (same is true for closed sites). closed sites than the 2‐ Substrate: Mud, sand, gravel (clay‐ (WGOMC) are compared to open For epifauna, Molgula sp. (solitary tunicate) year closed sites. silt, muddy‐sand, sand, granule fishing areas (the Kettle) using dominated all sites, Pandalus borealis is Pairwise tests indicated pebble) video and grab samples. Sampling common at 2 year closed and open sites, weakly significant Energy: Low in 2002 and 2004. Cerianthus borealis dominated 4‐year differences between the Evaluated: Biological, prey closed, sponges dominated 6‐year closed. 2‐ and 4‐year closed Epifaunal communities between open and sites, and closed sites could be distinguished. it appeared that the 2‐ year closed sites are more similar to both Open sites than the 4‐ year closed sites. Ref #: 211 Comparative. 3‐level hyperbenthic Trawling cause S changes in structure of Not evaluated Citation: Koulouri et al. (2005) sledge used to collect disturbed hyperbenthic communities. Groundrope Location: Crete, Mediterranean Sea (groundrope present) and disturbed hyperbenthic and zooplankton Depth: 50 m undisturbed (no groundrope) fauna living on or a few cm above the Substrate: Mud samples before and during trawling sediment surface during daylight; changes in Energy: ‐ season in an actively fished area. abundances of certain taxa persisted for a Evaluated: Prey week and can not be attributed to natural disturbances. Ref #: 214 Experimental, BACI design. Area Trawling re‐suspended surface sediment and Not evaluated Citation: Kutti et al. (2005) closed to fishing for 22 yrs, single caused re‐location of shallow, burrowing, Location: Barents Sea, Norway transect trawled 10 times in 20 hr; infaunal species to sediment surface; which Depth: 85‐100 m benthic organisms collected with led to S increase in abundance and biomass Substrate: Sand, gravel (shell an epibenthic sled equipped with a of most infaunal bivalves in sled samples 25 debris and finer sediment with video camera before trawling, hours after trawling. After trawling, there is some boulders) immediately after, and at 6 mo no change in the number of species sampled Energy: ‐ intervals for 2 yrs. or in the numerical diversity of species, but Evaluated: Biological (bivalves), there is S higher biomass diversity. There are prey slight but not dramatic changes in species composition. Ref #: 225 Comparative. Multiple gear types S higher incidence of rare sponge and shell Not evaluated. Citation: Lindholm et al. (2004) fished in study area (see scallop fragment habitats inside closed area, NS Location: Eastern Georges Bank dredge). Compared relative differences for 6 more common habitat Depth: 50‐100 m abundance of 7 microhabitats at 32 types in fished and unfished areas in mobile Substrate: Sand stations located inside and outside (<60m) or immobile (>60m) sand habitats, Energy: High an area closed for 4.5 yrs to bottom sponges and biogenic depressions Evaluated: Geological, biological, trawls and dredges (Closed Area II) numerically more abundant in immobile prey using video and still photos taken sand habitats inside closed area. along transects.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 53 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref #: 228 Comparative. Multiple gear types Benthic macroinvertebrate species richness After 5 years of closure, Citation: Link et al. (2005) fished in study area (see scallop did not vary by inside/outside closure but by generally did not see a Location: Georges Bank dredge). Fished inside and outside habitat type. notable increase in Depth: 35‐90 m of Closed Areas I and II with a #36 biomass and abundance Substrate: Sand, gravel Yankee otter trawl to sample inside the closed area Energy: High nekton and benthic community. for most species. Evaluated: Biological, prey Ref #: 236 Experimental trawling (single tow); An otter trawl that largely remained above Not evaluated Citation: Mayer et al. (1991) examined immediate effects on the sediment water interface caused little Location: Gulf of Maine, Maine sediment composition and food change it organic matter profiles, although coast, USA value to sediment depth of 18 cm. 7Be profiles suggested an export of the Depth: 20 m surficial horizon. Doors made furrows Substrate: Mud several cm deep; some planing of surface Energy: High, low features, but no plowing of bottom or burial Evaluated: Geological of surface sediments. Ref #: 238 Compared abundance of epifauna Reduced abundance (S for sponges and Not evaluated. Citation: McConnaughey et al. caught in small‐mesh trawl inside anemones); more patchy distribution; S (2000) and outside an area closed to decrease in species diversity of sedentary Location: Eastern Bering Sea, trawling for almost 40 yr. epifauna; mixed responses of motile taxa Alaska and bivalves. Depth: 44‐52 m Substrate: Sand Energy: High Evaluated: Biological Ref #: 239 Compared mean sizes (weights) of On average, 15 taxa are smaller in the HT Not evaluated. Citation: McConnaughey et al. 16 invertebrate taxa in 42 paired area with S differences between HT and UT (2005) samples from heavily trawled (HT) body sizes, but, individually, only a whelk Location: Bristol Bay, Eastern and untrawled (UT) areas. and sea anemones are S smaller. Overall, Bering Sea, Alaska these comparisons indicate natural Depth: 44‐52 m variability of body size in UT areas is large Substrate: Sand relative to the observed HT‐UT differences Energy: High due to chronic bottom trawling. Evaluated: Biological, prey Ref #: 247 Observational, pilot study without Marks on seafloor produced by trawl doors, Not evaluated Citation: Morais et al. (2007) controls. Observed 5 transects ground rope, and tickler chains observed Location: Southern Portugal near the head and in the flanks of a during 14‐60% of dive times, mostly in fine Depth: 121.5‐286.2 m canyon subject to trawling (for sediment, but also in coarse sediment and Substrate: Mud, sand, gravel crustaceans) with a submersible, rock‐boulder areas; furrows left by doors 40 (mostly fine sediment) continuously recording video, and cm wide and about 20 cm deep. Energy: Low occasionally collection biological Evaluated: Geological, biological, samples. Physical disturbance by prey trawl gear are coded as 1 (recent) to 3 (most eroded). Organisms in videos are identified to lowest taxonomic level possible, and sediment type is recorded. Ref #: 248 Experimental – multiple BACI When fished 15 cm above the seafloor, Not evaluated Citation: Moran and Stephenson design. Video surveys before and semi‐pelagic trawls had no measureable (2000) after four experimental trawling effects on macrobenthos (> 20 cm high, all Location: Northwest Australia events (one tow per unit area) at 2‐ species pooled). A single tow with a Depth: 50‐100 m day intervals in unexploited area. demersal trawl reduced density of Substrate: Sand (assumed) Compared demersal vs. semi‐ macrobenthos (>20 cm) by 15.5%, while Energy: ‐ pelagic trawl effects. after 4 tows macrobenthic density is Evaluated: Biological. reduced by 50%.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 54 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref #: 407 Evaluation of benthic habitat Variation in mean BHQ values greater in Not evaluated Citation: Nilsson and Rosenberg quality index (BHQ) – sediment trawled transects than natural variation in (2003) surface and subsurface variables experimental area; BHQ values lower in Location: Fjord on the west coast plus redox conditions – in trawled transects than in control transects of Sweden sediment profile images (SPIs) from after trawling started; a severe mechanical Depth: 75‐90 m 3 control and 3 experimentally disturbance observed in about 43% of the Substrate: Mud (clay) trawled transects sampled 10 times SPIs which S increased spatial variance of Energy: ‐ on 3 occasions before and during BHQ‐indices in trawled areas compared to Evaluated: Geological trawling in a area not exposed to control areas. commercial shrimp trawling for 6 yr (see Hansson et al 2000). Shrimp trawl gear. See Refs #313, 149. Ref #: 277 Experimental trawling along two Door tracks; footrope removed 2‐3 cm fine Door tracks still visible 1 Citation: Palanques et al. (2001) transects (7 and 14 sets) in summer sediment in study area, but silt settled on to yr after trawling, but are Location: NW Mediterranean Sea (period of low natural disturbance) sediment surface within 1 hr after trawling; not as sharply defined; Depth: 30‐40 m during two different yrs; before turbidity increased 3x 4‐5 d after trawling, surface sediments Substrate: Mud (>80% clay and silt) and after changes in bottom first near the bottom and within 2‐5 hrs mixed within a day after Energy: Low morphology monitored with further off the bottom; at least 10% of trawling (i.e., same grain Evaluated: Geological sidescan sonar; also measured sediment disturbed by trawling remained in size distribution) currents, turbidity, vertical profiles water column 4‐5 d after trawling, up to 90% of temperature, salinity, light accumulated on the bottom. transmission, and sediment composition in trawl lines, before and at various times after trawling. Ref #: 283 Deployed sediment traps in fishing Sediment resuspension due to trawling Not evaluated Citation: Pilskaln et al. (1998) grounds 25 m above substrate. inferred based on greater abundance of Location: Gulf of Maine, USA suspended (25 m off bottom) infaunal Depth: 250 m polychaetes in more heavily trawled area. Substrate: Mud Energy: ‐ Evaluated: Geological Ref #: 291 Experimental trawling (3‐6 tows 24% average decrease in epibenthic Not evaluated Citation: Prena et al. (1999) per unit of area) in closed area 1, 2 biomass; S reductions in total and mean Location: Grand Banks, and 3 yr after closure, lightly individual epifaunal biomass, and biomass of Newfoundland exploited for >10 yr. five of nine dominant species; damage to Depth: 120‐146 m echinoderms. Substrate: Fine to medium grained sand Energy: Low Evaluated: Biological, prey Ref #: 313 (A) Experimental. 8 sediment (A) Bottom scraped by net, furrows Not evaluated Citation: Rosenberg et al. (2003) profile images analyzed from each produced by doors, mud clasts on sediment Location: (A) Fjord on the west of 3 trawled and 3 control areas surface (could be caused by weights on coast of Sweden (B) Gulf of Lions, during experimental trawling; each ground rope or the net), disturbed sediment NW Mediterranean area trawled 80 times during year‐ surface, reduced number of polychaete Depth: (A) 73‐96 m (B) 35‐88 m long study. See Refs #407, 149. tubes. Substrate: Mud ((A) Clay (B) Mud (B) Observational. 76 sediment (B) Evidence of physical disturbance, e.g. and some sand) profile images analyzed from 26 presence of mud clasts, greater sediment Energy: ‐ stations in 4 coastal locations relief (furrows produced by trawl doors); Evaluated: Geological exposed to unknown levels of absence of epifauna (crinoids) and lower fishing activity, no controls. number of tubes.

Ref #: 320 Experimental trawling in trawled Door tracks in sediment; no change in Door tracks still clearly Citation: Sanchez et al. (2000) area at two sites swept once and number of infaunal individuals or taxa, or in visible after 150 hr. Location: Coast of Spain, twice in a single day; effects abundance of individual taxa; no changes in Mediterranean Sea evaluated after 24, 72, 102, and community structure. Depth: 30‐40 m 150 hr. Substrate: Mud Energy: Low Evaluated: Geological, biological, prey

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 55 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref #: 325 Experimental trawling (3‐6 tows Tracks in sediment; increased bottom Tracks last up to 1 yr; Citation: Schwinghamer et al. per unit area) in closed area 1, 2 roughness; sediment resuspension and recovery of seafloor (1998) and 3 yr after closure, lightly dispersal; smoothing of seafloor and topography within 1 yr. Location: Grand Banks, exploited for >10 yr. removal of flocculated organic material; Newfoundland organisms and shells organized into linear Depth: 120‐146 m features. Substrate: Fine to medium grained sand Energy: Low Evaluated: Geological Ref #: 330 Compared sediments and benthos Predicted accumulation of fine sediments Not evaluated Citation: Sheridan and Doerr (2005) in two adjacent areas, one of which during the 7 mo closure did not occur, Location: Gulf of Mexico, Texas is closed to shrimp trawling for 7 probably because natural disturbance coast mo using divers to collect cores in 5 affects long‐term sediment structure more Depth: 5‐20 m sediment texture classes. Shrimp so than shrimp trawling. Densities and Substrate: Mud, sand trawl gear. biomasses of most abundant taxa and major Energy: High taxonomic groups are similar between Evaluated: Geological, prey zones. Ref #: 333 Experimental. Block design Disturbance caused by shrimp trawling Not evaluated Citation: Simpson and Watling comparing habitat and during study at inshore site produced (2006) macrofaunal community structure changes in macrofaunal community Location: Gulf of Maine, USA at two sites before, during, and structure (but not in total abundance), Depth: 84‐102 m after shrimp trawling season, each whereas effects of older (at least 1 yr Substrate: Mud with trawled and un‐trawled areas, previous) trawling activity could not be Energy: ‐ using video quadrat surveys, an detected at offshore site. Disturbance‐ Evaluated: Geological, prey ROV, and box core samples. tolerant polychaetes more abundant in Shrimp trawl gear, generally with trawled area at inshore site, more sensitive 21m sweeps, 5cm mesh, bivalve mollusks in un‐trawled area. rockhoppers. Untrawled areas had fixed gear. Ref #: 334 Video and diver observations. Tracks in sediment (<5 cm in sand, 5‐15 cm Tracks "naturalized" by Citation: Smith et al. (1985) in mud); attraction of predators; suspension tidal currents. Location: Long Island Sound, New of epibenthic organisms. York, USA Depth: < 50 m Substrate: Mud, sand Energy: High Evaluated: Geological, prey Ref #: 335 Compared two stations inside a Higher numbers of epifauna (especially Closed season did not Citation: Smith et al. (2000) trawling lane and two outside; echinoderms) observed outside trawling seem to allow enough Location: North coast of Crete, video survey and epifaunal lane; species number, biomass, and time for recovery to pre‐ Mediterranean sampling over 11 months starting abundance generally S lower in lane during season levels. Depth: 200 m before the 8 month trawling trawling season, esp echinoderms, Substrate: Mud (soft, silty clay) season and ending well after. sipunculids, and, to a lesser degree, Energy: Low polychaetes; impacts on less mobile fauna Evaluated: Geological, biological, are more pronounced, but degree of prey robustness is also important,

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 56 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref #: 336 (A) Experimental trawling (13 tows (A) Trawling S increased sediment Not evaluated Citation: Smith et al. (2003) in 2 d) with adjacent control area in compaction and reduced bottom roughness Location: Aegean Sea (site B – Dia un‐trawled area between two with greater impact within first 2 mo after Island – is same site as ref# 335) commercially‐trawled lanes; trawling; no differences between areas 5 mo Depth: (A) 80‐90 m (B) 200m impacts analyzed using sediment after trawling impact. Coarse sediments at Substrate: (A) Mud (muddy sand profile imagery. Data collected this site naturally more compacted than soft with maerl fragments) (B) Mud immediately after trawling and 2, 5, sediments, so primary effect is smoothing of (soft, silty clay) and 7 mo afterwards. bottom. Energy: ‐ (B) Used sediment profile imagery (B) No S differences in sediment compaction Evaluated: Geological to analyze sediment penetration or roughness (with or without substrate and roughness, plus a number of attributes), or in substrate attributes alone, substrate attributes (e.g., sediment in trawled and non‐trawled areas, due to type, small‐scale topography, heterogeneous nature of trawling impacts degree of bioturbation) in a (e.g. furrows produced by doors “cancel out” commercially‐ trawled lane and smoothing and scraping action of ground adjacent non‐trawled control areas rope and net). during and after trawling season in two consecutive years. Ref #: 338 Experimental trawling (four tows in S decline in porosity, increased food value, All geochemical Citation: Sparks‐McConkey, and 1 day) in untrawled area; pre‐trawl and increased chlorophyll production of sediment properties and Watling (2001) sampling of sediments and infauna surface sediments; S reductions in number all but one Location: Penobscot Bay, Maine, for a year; recovery monitored for of infaunal individuals and species, species polychaete/bivalve USA 5 mo. Shrimp trawl gear. diversity, and abundances of 6 polychaete species recovered Depth: 60 m and bivalve species, S increase in within 3.5 mo, Substrate: Mud nemerteans. nemerteans still more Energy: Low abundant after 5 mo. Evaluated: Geological, prey Ref #: 355 Experimental. Spatial distribution Species richness is lower in open areas; Not evaluated Citation: Stone et al. (2005) and abundance of epifauna are species dominance is greater in one open Location: Central Gulf of Alaska examined at two sites that area, while the other site had significantly Depth: 105‐157 m overlapped areas open to trawling fewer epifauna in open areas; both sites had Substrate: Sand and areas where bottom trawling decreased abundance of low‐mobility taxa Energy: ‐ had been prohibited for 11‐12 and prey taxa in the open areas. Prey taxa Evaluated: Geological, biological, years. Scallop dredging at one of are highly associated with biogenic and prey the sites. Video strip transects of biotic structures; biogenic structures are the seafloor are collected at each significantly less abundant in open areas. site from a manned submersible. Ref #: 360 Experimental, multiple BACI design Epifauna (a bivalve, ascidians, bryozoans, Further (8%) declines in Citation: Tanner (2003) (multiple after assessments, one and sponges) at trawled sites decreased in epifaunal abundance 2– Location: Gulf of St. Vincent, impact event). Video photos of abundance by 28% within 2 weeks of 3 months post trawling. Australia sessile epifauna taken by SCUBA trawling compared with control sites; Recruitment rates of Depth: 20 m divers in treatment and control persistence of most taxa declined several taxa into visible Substrate: Sand quadrats before and 1 wk and 3 mo significantly in trawled areas compared with size classes increased Energy: High after experimental trawling in 3 untrawled areas. Sea grasses less likely to after trawling (authors Evaluated: Geological, biological locations (two tows per unit area); colonize trawled sites. suggest this might have study area not fished for 15 yrs been due to a increase (see Ref #97). Shrimp trawl gear in available space). with ground chain (specific description given). Ref #: 372 Experimental trawling for 1 day/mo Tracks in sediment, increased bottom Door tracks still evident Citation: Tuck et al. (1998) (one and a half tows per unit of roughness; no effect on sediment after 18 mo; bottom Location: West coast of Scotland area) for 16 mo in area closed to characteristics; S increase in number of roughness recovered Depth: 30‐35 m fishing for >25 years; recovery infaunal species at end of 16 mo disturbance after 6 mo; nearly Substrate: Mud (fine silt) monitored after 6, 12, and 18 mo. period and during 18 mo recovery period; no complete recovery of Energy: Low change in biomass or number of individuals infaunal community Evaluated: Geological, prey at end of recovery period; S increase in within 12 mo, complete polychaetes, S decrease in bivalves; mixed after 18 mo. results of analyses of community structure, S reduction in diversity during first 22 mo.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 57 DRAFT April 17, 2009 Study information Study approach Effects/Impact Recovery Ref #: 382 Experimental study using diver Damage to sponges and corals, mostly to Full recovery of Citation: Van Dolah et al. (1987) counts of large sponges and corals sponges; S reductions in density of damaged organisms and Location: Georgia, SE USA before, immediately after, and 12 undamaged barrel sponges in high‐density density within 12 Depth: 20 m mo after, a single tow of a “roller” transects; no S effects on densities of vase months. Substrate: Sand (smooth rock with trawl in an unexploited area. sponges, finger sponges, or stony corals. thin layer of sand and attached epifauna). Energy: ‐ Evaluated: Biological Ref #: 387 Catch and damage to sponges and Net removed 14% sponges and 3% Not evaluated Citation: Wassenberg et al. (2002) gorgonian corals observed with gorgonians from bottom (average per tow); Location: Northwest Australia video camera mounted in trawl sponges higher than 500 mm impacted the Depth: 25‐358 m (average 78.3m, during 108 individual tows; the most (only 30‐60% passed under net); 68‐ sponges mostly <100) sponges are divided into two types 100% sponges <500mm passed under net; Substrate: Sand, gravel (coarse (lump = broad base; branched = <3% of 300‐500mm sponges are broken up sand with 10‐30% gravel) narrow base) and three height as they passed under net, but all large Energy: ‐ classes (< 300, 301–500 and > 500 branched sponges that did not pass into net Evaluated: Biological mm). Gear is a McKenna fish trawl are either removed by footrope or crushed (very specific configuration given). under it; 90% gorgonians passed under net.

4.4.2 New Bedford‐style scallop dredges A total of 15 studies examined the effects of New Bedford‐style scallop dredge gear on seafloor habitat components (Table 15). Of these, eight (Asch and Collie 2007; Auster et al. 1996; Collie et al. 1997; Collie et al. 2000; Collie et al. 2005; Hermsen et al. 2003; Lindholm et al. 2004; and Link et al. 2005) examined aggregated impacts of scallop dredges and otter trawl gears, and seven (Caddy 1968; Caddy 1973; Langton and Robinson 1990; Mayer et al. 1991; Stokesbury and Harris 2006; Sullivan et al. 2003; and Watling et al. 2001) examined scallop dredging only. Most studies were conducted on sand or sand/gravel habitats; although Auster et al. (1996), Caddy (1968), and Mayer et al. (1991) examined the effects of dredging on mud substrates. Of the 15 studies, 11, 11, and 14 addressed impacts to geological, biological, and prey habitat components, respectively. The studies were conducted primarily in U.S. waters in the Gulf of Maine, on Georges Bank, or in the Mid‐Atlantic Bight, generally in high energy areas; two were conducted in the Gulf of St. Lawrence. Three of the 15 (Stokesbury and Harris 2006; Sullivan et al. 2003; and Watling et al. 2001) were experimental studies. Six of the studies addressed recovery.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 58 DRAFT April 17, 2009 Table 15 ‐ Impacts of New Bedford‐style scallop dredges on geological, biological, and prey habitat components. (Where multiple field studies are included in one publication, or where multiple publications are summarized together, individual components are denoted as (A), (B), (C), etc. ‘S’ indicates statistical significance) Study information Study approach Effects Recovery Ref #: 404 Comparative. Multiple gear types At shallow sites, cover of all At shallow sites, several taxa Citation: Asch and Collie (2007) fished in study area (see New epifauna except hydroids S differed showed changes in abundance Location: Northern edge, Bedford scallop dredge). Still by disturbance regime. Sponges beginning 2 years after closed area eastern Georges Bank, U.S. and photographs (N=454) are analyzed and bushy bryozoans showed S established. Increase in abundance Canada for percent cover of colonial higher % cover at undisturbed of P. magellanicus, Pagurus spp., S. Depth: 40‐50, 80‐90 m epifauna and the abundance of sites, encrusting bryozoans and droebachiensis, Asterias spp. Substrate: Gravel (pebble and non‐colonial organisms at shallow Filograna implexa showed S higher between closure (1994) and 2000. cobble pavements with some and deep disturbed and % cover at disturbed sites. Also at overlying sand) undisturbed sites in and around shallow sites, generally S between Energy: High Georges Bank Closed Area II. year variations. At deep sites, % Evaluated: Biological, prey cover of F. implexa and hydroids is S higher in undisturbed areas; other taxa showed no differences by disturbance regime. For non‐ colonial epifauna, depth contributed more to differences in species composition than disturbance. Higher spp. richness at undisturbed sites (S at shallow sites). Ref #: 11 Comparative. Amount of fishing (A) In cobble habitat (N=12‐13 Not addressed Citation: Auster et al. 1996 effort and types of mobile gear transects per treatment), S lower Location: 3 sites, Gulf of Maine, used in study areas not well cover of emergent epifauna, sea USA defined. See otter trawl. Three cucumbers in fished area; in sand Depth: See study approach sites: habitat (N=17‐18 transects per Substrate: See study approach (A) Swans Island: closed 10 yr, sand treatment), S lower cover of sea Energy: Not specified and cobble, depth not specified, cucumbers and biogenic Evaluated: Geological (B), comparative: inside‐outside video depressions in fished area. biological (A,B,C) transects (B) Qualitative; loss of mud veneer, (B) Jeffreys Bank: boulders reduction in epifaunal species, incl. prevented fishing, then fishing, sponges (quantified but no gravel and mud, depth 94m, statistical tests), movement of comparative: one pair of before/6 boulders years after submersible dives (C) Positive relationship between (C) Stellwagen Bank: daily fishing hydrozoan Corymorpha penduala evidenced by trawl/dredge tracks, and shrimp in 1993, fewer areas gravel and sand, depth 32‐43m, with hydrozoans and wide observational: n=4 (?) video distribution of tunicate Molgula transects over 2 years arenata in 1994 Ref #: 42 Observed impacts during a scallop Drag tracks (3 cm deep) produced Not addressed Citation: Caddy 1968 dredge efficiency study. Divers by skids; smooth ridges between Location: Northumberland examined physical effects of two them produced by rings in drag Strait, Gulf of St. Lawrence, tows. belly; dislodged shells in dredge Canada tracks. Depth: 20 m Substrate: Mud, sand Energy: High Evaluated: Geological Ref #: 43 Submersible observations of tow Suspended sediment; flat track, Not addressed Citation: Caddy 1973 tracks made <1 hr after single marks left by skids, rings, and tow Location: Chaleur Bay, Gulf of dredge tows. bar; gravel fragments less frequent St. Lawrence, Canada (many overturned); rocks dislodged Depth: 40‐50 m or plowed along bottom. Substrate: Sand, gravel (gravel over sand, with occasional boulders) Energy: Low Evaluated: Geological

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 59 DRAFT April 17, 2009 Study information Study approach Effects Recovery Ref #: 69, 70, 71 Comparative. Multiple gear types S higher total densities, biomass, C) 5 years after fishing eliminated Citation: (A) Collie et al. 1997, fished in study area (see scallop and species diversity in from area (Closed Area II), (B) Collie et al. 2000, (C) Collie dredge). Benthic sampling, video, undisturbed sites, but also in observed S shifts in species et al. 2005 and still photos in 2 shallow (42‐47 deeper water (i.e., effects of fishing composition and S increases in Location: Eastern Georges Bank m) and 4 deep (80‐90 m) sites could not be distinguished from abundance, biomass, production, (Northern edge), U.S. and disturbed (D) and undisturbed (U) depth effects); 6 species abundant and epifaunal cover. Canada by trawls and scallop dredges. at U sites, rare or absent at D sites; Depth: 42‐90 m percent cover of tube‐dwelling Substrate: Sand, gravel polychaetes, hydroids, and (pebble‐cobble “pavement” bryozoans S higher in deepwater, with some overlying sand) but no disturbance effect. Energy: High Evaluated: (A, C) Geological, (A,B,C) biological, prey Ref #: 158 Comparative. Multiple gear types Production remained markedly See impacts/effects Citation: Hermsen et al. 2003 used in study area (see otter lower at shallow disturbed site over Location: Northern edge, trawl). Benthic macrofauna course of study than at nearby eastern Georges Bank, U.S. and sampled at deep and shallow sites recovering site, where it increased Canada disturbed and undisturbed (by over 12‐fold from before closure to Depth: 47‐90 m fishing) using Naturalists dredge 5 yrs after closure; at deep sites, Substrate: Sand, gravel with a 6.4 mm liner 8 times over 7 production remained S higher at (pebble‐cobble “pavement” yr period, 2 yrs prior to closure, just undisturbed sites. Sea scallops and with some overlying sand) after closure, and 5 yrs after sea urchins dominated production Energy: High closure. at shallow recovering site; a soft‐ Evaluated: Geological, bodied tube‐building polychaete biological, prey dominated production at the deep, undisturbed site. Ref #: 217 Submersible observations made 1 Three species dominated both sites Not addressed Citation: Langton and Robinson yr apart, before and after – Placopecten magellanicus, 1990 commercial dredging of area. Myxicola infundibulum, Cerianthus Location: Jeffreys and borealis. After dredging densities Fippennies Ledges, Gulf of of all three are reduced. Maine, USA Depth: 80‐100 m Substrate: Sand, gravel (gravelly sand with some gravel, shell hash, and small rocks) Energy: ‐ Evaluated: Biological, prey Ref #: 225 Comparative. Multiple gear types S higher incidence of rare sponge Not addressed Citation: Lindholm et al. (2004) fished in study area (see otter and shell fragment habitats inside Location: Eastern Georges Bank trawl). Compared relative closed area, NS differences for 6 Depth: 50‐100 m abundance of 7 microhabitats at 32 more common habitat types in Substrate: Sand stations located inside and outside fished and unfished areas in mobile Energy: High an area closed for 4.5 yrs to bottom (<60m) or immobile (>60m) sand Evaluated: Geological, trawls and dredges (Closed Area II) habitats, sponges and biogenic biological, prey using video and still photos taken depressions numerically more along transects. abundant in immobile sand habitats inside closed area. Ref #: 228 Comparative. Multiple gear types Benthic macroinvertebrate species After 5 years of closure, generally Citation: Link et al. (2005) fished in study area (see scallop richness did not vary by did not see a notable increase in Location: Georges Bank dredge). Fished inside and outside inside/outside closure but by biomass and abundance inside the Depth: 35‐90 m of Closed Areas I and II with a #36 habitat type. closed area for most species. Substrate: Sand, gravel Yankee otter trawl to sample Energy: High nekton and benthic community. Evaluated: Biological, prey

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 60 DRAFT April 17, 2009 Study information Study approach Effects Recovery Ref #: 236 The effect of commercial dragging A heavy scallop dredge caused two Not addressed. Citation: Mayer et al. (1991) on sedimentary organic matter is types of organic matter Location: Gulf of Maine, Maine examined in two field experiments translocation – some of the surficial coast, USA using different gear types. organic matter is exported from Depth: 20 m the drag site and the remaining Substrate: Mud material is mixed into subsurface Energy: High, low sediments. Phospholipid analysis Evaluated: Geological indicated decreases in various classes of microbiota, with relative increases in the contribution of anaerobic bacteria to the microbial community. Ref #: 352 BACI study (counts of fish and Changes in density in areas Not addressed Citation: Stokesbury and Harris marcoinvertebrates <40 mm in impacted by limited fishing are 2006 video images) in areas that are similar to changes in control areas; Location: Georges Bank open to scallop fishing in 2000/01 in both experiments Depth: 52‐70 m (means at 4 and control areas that have bryozoans/hydrozoans increased S sites) remained closed since 1994; exp 1 after fishing, while sponges Substrate: Sand, gravel (Sand, compared northern portion of CAII decreased in impact and control shell debris, granule/pebbles, (closed) with NLCA (open), exp 2 areas (S so in exp1), and sand cobbles and boulders) compared open and closed dollars decreased NS in impact Energy: High portions of CAI; both sites in each portion of CAI, with NS increases in Evaluated: Geological, experiment had similar tidal closed area. Temporal changes in biological, prey current velocities, impact areas in open and closed areas (before‐ both experiments deeper with before and after‐after) and shifts in more sand than control areas. sediment composition between surveys indicate that fishing affected the epibenthic community less than natural environmental conditions. Ref #: 359 Experimental. Effects of Dredging reduced physical No evidence of a dredging impact Citation: Sullivan et al. 2003 experimental dredging on habitat heterogeneity such that the of any kind apparent after 3 Location: New York Bight structure for YOY yellowtail frequency of sand waves, tube months and 1 year; however major Depth: 45‐88 m flounder evaluated using a mats, and biogenic depressions disturbance of seabed at two Substrate: Sand submersible to conduct pre‐dredge decreased relative to control plots; shallower sites caused by Energy: High and post‐dredge surveys (2d, 3mo, typical post‐dredge landscapes hurricanes 2 months after Evaluated: Geological, 1yr after impact) at 3 sites (2 within (<2d) consisted of extensive experimental dredging. biological, prey Hudson Canyon closed area), with patches of clean, silty sand, multiple control and dredge interspersed with regular striations treatments at each site. of shell hash; abundant mobile epifauna such as sand dollars typically dislodged or buried under a thin layer of silt. Despite the vigorous reworking of surficial sediments, the overall impact of the dredge appeared to extend no deeper than 2‐6 cm below the sediment surface. A significant decrease in available benthic prey is observed at 3 months following a series of major natural perturbations (Hurricanes Dennis, Floyd, and Gert). Ref #: 391 Experimental study (23 tows in 1 Loss of fine surficial sediments; No recovery of fine sediments, full Citation: Watling et al. 2001 day); effects on macrofauna lowered food quality of sediment; recovery of benthic fauna and food Location: Damariscotta River, (mostly infauna) evaluated 1 day reduced abundance of some taxa; value within 6 months. Maine, USA and 4 and 6 months after dredging no changes in number of taxa; S Depth: 15 m in an unexploited area. reductions in total number of Substrate: Sand (Silty sand) individuals 4 months after Energy: ‐ dredging. Evaluated: Geological, prey

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 61 DRAFT April 17, 2009 4.4.3 Hydraulic clam dredges Eleven studies examined the effects of hydraulic clam dredges on habitat components (Table 16). These studies were conducted predominantly off the U.S. east coast in the Mid‐Atlantic Bight, and also off the coasts of Scotland, Nova Scotia, and Italy.

Table 16 ‐ Effects of hydraulic clam dredges on geological, biological, and prey habitat components. (‘S’ indicates statistical significance.) Study information Study approach Effects Recovery Ref #: (A) 121, (B) 122, (C) 123 Experimental. BACI design. Same study (A) Dramatic change in topography, (A) Slow degradation of Citation: Gilkinson et al. (A) evaluated recovery of physical habitat loss of burrows, tubes, and shells dredge furrow margins 2003, (B) 2005a, (C) 2005b features (A), macrobenthic community (B), through destruction or burial; over time, 90% reduction Location: Scotian Shelf and soft coral Gersemia rubiformis (C). sedimentation created smooth in large burrow density Depth: 70‐80 m Monitored 1, 2, and 3 years after initial surface. due to reductions in Substrate: Sand with shell disturbance in a previously un‐dredged area. (B) Most macrofauna decreased in commercially harvested deposits Three treatments (dredging and discarding, abundance immediately after propellerclam (Cyrtodaria Energy: Low dredging, and discarding) and control areas. dredging; marked increases in siliqua) abundance. Evaluated: Geological, polychaetes and amphipods over (B) Average taxonomic biological, prey time. Some macrofauna increased distinctness is still in abundance in both dredged and decreased two years post‐ controlled areas (possibly due to impact; also, sustained natural increases in recruitment). reduction in biomass of (C) No detectable impact of target bivalves dredging on soft coral abundance, (C) No impacts detected but ANOVA is relatively low power and capture rates of corals in dredges are low and variable. Ref #: 140 Experimental study in unexploited area to Shallow trenches (25 cm deep) and Complete recovery of Citation: Hall et al. (1990) evaluate effects of simulated commercial large holes; sediment “almost physical features and Location: Loch Garloch, escalator dredging activity; recovery fluidized”; median sediment grain benthic community after Scotland evaluated after 40 days. size S higher in fished area; S 40 days; filling of trenches Depth: 7 m reductions in numbers of infaunal and holes accelerated by Substrate: Fine sand organisms; no S effect on winter storms. Energy: High abundance of any individual Evaluated: Biological, prey species, but mean abundances of 10 most common species are all lower 1 d after fishing than in controls and difference for whole group is S; some mortality (not assessed) of large polychaetes and crustaceans retained on conveyor belt or returned to sea surface. Ref #: 232 Comparison of actively fished, recently Resorting of sediments (coarser at Not evaluated Citation: MacKenzie (1982) fished, and never fished areas on the bottom of dredge track); no effect Location: East of Cape May, continental shelf; dredging conducted with on number of infaunal individuals New Jersey, USA hydraulic cage dredges. or species, nor on species Depth: 37 m composition. Substrate: Very fine to medium sand Energy: ‐ Evaluated: Geological, biological, prey Ref #: 244 Observational. SCUBA and submersible Smooth tracks with steep walls, 20 Sediment plume lasted 1 Citation: Medcof and Caddy observations of the effects of individual tows cm deep; sediment cloud. min; dredge tracks still (1971) with a cage dredge. clearly visible after 2‐3 Location: Southern Nova Scotia, days. Canada Depth: 7‐12 m Substrate: Sand and sand‐mud Energy: High Evaluated: Geological

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 62 DRAFT April 17, 2009 Study information Study approach Effects Recovery Ref #: 245 Observational. SCUBA observations during >20‐cm‐deep trench; mounds on Trench nearly indistinct, Citation: Meyer et al. (1981) and following two tows with a cage dredge either side of trench; silt cloud, and predator abundance Location: Long Island, New in a closed area; effects evaluated after 24 attraction of predators. normal, after 24 hr; silt York, USA hr. settled in 4 min. Depth: 11 m Substrate: Fine to medium sand, covered by silt layer Energy: High Evaluated: Geological Ref #: 249 Experimental beyond‐BACI design. In a Upon multivariate analysis of the The authors note that the Citation: Morello et al. (2005) heavily fished area that is closed to entire sampled macrozoobenthic lack of response to impact Location: Coastal Adriatic Sea commercial dredging for 6 months, they assemblage, the Polychaeta, the could indicate a possible Depth: 6m sampled the macrobenthic community Crustacea, and also detritivorous historical effect of Substrate: Sand hours, days, and weeks before and after and suspensivorous guilds, no persistent disturbance Energy: High experimental fishing (2 control plots and 2 impact or effect of the due to fishing; they cite Evaluated: Prey dredged plots). Selected three subplots for experimental fishing tows is dominance by suction sampling during each sampling discernible over natural variability. polychaetes as evidence event. However, the bivalve Abra alba of a disturbed state. (which has a fragile shell) is found to be very susceptible to dredging and suggested as a disturbance indicator species. Ref #: 256 Observational. Submersible observations Trench cut, temporary increase in Trenches filled quickly in Citation: Murawski & Serchuk following hydraulic cage dredge tows. turbidity, disruption of benthic coarse gravel, but took (1989) organisms in dredge path, several days in fine Location: Mid‐Atlantic Bight, attraction of predators. sediments. USA Depth: Not given Substrate: Sand, mud and coarse gravel Energy: ‐ Evaluated: Geological, prey Ref #: 287 Experimental dredging with a cage dredge 8‐10 cm deep trench; S decrease in After 2 months dredge Citation: Pranovi and (single tows) in previously dredged and total abundance, biomass, and tracks still visible; Giovanardi 1994 undredged areas in coastal lagoon; recovery diversity of benthic macrofauna in densities (especially of Location: Venice Lagoon, monitored every 3 wk for 2 months. fishing ground; no S effects outside small species and Adriatic Sea, Italy fishing ground. epibenthic species) in Depth: 1.5‐2 m fishing ground recovered, Substrate: Sand biomass did not. Energy: Low Evaluated: Geological, biological Ref #: 373 Experimental dredging with cage dredge Steep‐sided trenches (30 cm deep); Trenches no longer visible Citation: Tuck et al. 2000 (individual tows at 6 sites) in area closed to sediments fluidized up to 30 cm; S but sand still fluidized Location: Sound of Ronay, commercial dredging, effects evaluated 1 decrease in number of infaunal after 11 wk; species Outer Hebrides, Scotland day, 5 days, and 11 wk after dredging. species and individuals within a day diversity and total Depth: 2‐5 m of dredging; S decrease in abundance recovered Substrate: Medium to fine sand proportion of polychaetes and S within 5 days; proportions Energy: High increase in proportion of of polychaetes and Evaluated: Geological, amphipods 5 days after dredging; S amphipods, and biological, prey increases in abundance of some abundances of individual species and S decreases in species, returned to pre‐ abundance of other species. dredge levels after 11 wk.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 63 DRAFT April 17, 2009 4.4.4 Lobster and deep‐sea red crab traps One study (Eno et al. 2001, Table 17) addresses the impacts of crustacean traps on biological habitat components including soft corals, sea pens, sponges, emergent bryozoans, and tunicates.

Table 17 ‐ Impacts of crustacean traps on biological habitat components (“S” indicates statistical significance) Study information Study approach Effects Recovery Ref #: 102 Examined effects of crustacean trap fisheries (A) Sea pens bent under (A) Full recovery of uprooted Citation: Eno et al. (2001) on species thought to be sensitive. pressure wave before pot sea pens providing peduncle Location: Great Britain (A) Off (A) Manipulative, observational. Assessed contact, sea pens are uprooted gained mud contact, full Scotland (B) Lyme Bay (C) (with diving and video) sea pen recovery and (B) Some detachment of large recovery of smothered sea Greenala Point, West Wales survival following dragging, uprooting, and sponges and damage to large pens Depth: (A) Not given (B) 10‐ smothering by lobster pots ross coral (an emergent (B) Not evaluated 20m (C) < 23m (B) Observational. Divers surveyed deployed bryozoan); not sure if this is (C) Not evaluated Substrate: (A) Mud (B) Gravel pot lines and then watched as lines are related to fishing. Sea fans bent (exposed bedrock covered with hauled. under pots. sand, boulders interspersed (C) Experimental. At two sites with control (C) Track is left in sediments by with coarse sediment (C) Rocky and experimental blocks, pots are deployed pots when hauled; soft corals Energy: High and then hauled every 2‐3 days for four bent under pots during hauling; Evaluated: (A,B,C) Biological weeks. No S decrease in abundance of sponges, soft corals, bryozoans, and tunicates due to fishing (increases in abundance of some taxa in some plots)

4.5 Estimating susceptibility and recovery Vulnerability of habitats to fishing gear impacts is being evaluated using a matrix‐based approach. Susceptibility and recovery values are assigned to each individual habitat component using knowledge of the gear and habitat components combined with the findings of the literature described above. A generic matrix is shown below (Table 18). It includes the features, effects, susceptibility for each feature, recovery for each feature at both high and low energy, and the literature deemed relevant during the literature database review. There is one matrix for each habitat component and gear type combination (e.g. geological habitat components subject to New Bedford‐style scallop dredge impacts).

Table 18 ‐ Generic matrix construction Habitat component, Gear type Recovery – high Recovery – low Studies Feature Effects Susceptibility energy energy considered Feature_1 Effects 1 ‐ x 0‐3 0‐3 0‐3 #, #, #... Feature_2 Effects 1 ‐ x 0‐3 0‐3 0‐3 #, #, #... Feature_3 Effects 1 ‐ x 0‐3 0‐3 0‐3 #, #, #...

Susceptibility is defined as the change in functional value of a habitat component due to a gear effect. Recovery is defined as the speed with which the functional value of that unit of habitat is restored. Both susceptibility and recovery are scored from 0‐3 as shown in Table 19 and Table 20 below. The values for susceptibility and recovery are intended to be comparable in magnitude. Although both susceptibility and recovery would, in reality, vary continuously for a particular interaction, they are considered categorical variables in this analysis.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 64 DRAFT April 17, 2009

Table 19 ‐ Susceptibility values Code Description Quantitative 0 None 0 1 Low >0 ‐ 25% 2 Medium 25 ‐ 50% 3 High > 50%

Table 20 ‐ Recovery values Code Description Quantitative 0 Fast < 1 year 1 Moderate 1 – 2 years 2 Slow 2 – 5 years 3 Very slow > 5 years

Susceptibility is evaluated for the entire swath of seabed affected by the gear during one tow. No baseline state of the seabed is assumed. Effects that might extend beyond the width of the trawl (sediment re‐suspension, etc.) are considered in sensitivity evaluations.

Susceptibility of the features is assumed to depend on the effects of the gear. These effects are summarized in each individual matrix, and might include crushing, scraping, compressing, slicing, creasing, breaking, plowing, burying, and fluidization. Quantitative percentages in Table 19 indicate the proportion of features in the path of the gear likely to be modified to the point that they no longer provide the same functional value. In most cases, a habitat component is small in comparison with the path of the gear. However, in other cases, large geological bedforms, for example, the effects are considered on the portion of the bedform in the path of the gear. All gear components (e.g. doors, ground cables, sweep) are considered together for the purpose of evaluating susceptibility. For a particular habitat feature (e.g. sand bedforms), different susceptibility values for different gear types are based on the differences in gear components and their potential effects on the seabed. Knowledge of the various gear types allows the analyst to distinguish between them for the purpose of assigning susceptibility values. For a given gear type, the component most in contact with the seabed is weighted more heavily. For example, when evaluating the effects of raised footrope trawl gear, the sweep, which has very little contact with the seabed (roughly 5%), would not be weighted heavily, and the effects of the doors and ground cables would be weighted more heavily. For conventional groundfish trawl gear, the effects of all three gear components would be considered.

If the susceptibility indicates that 25% of the features were modified, recovery would mean that all 25% would come back within the time period specified by the recovery value. Recovery does not necessarily mean a restoration of the exact same features, but that after recovery the habitat would have the same functional value. For example, a recovery value of zero for biogenic depressions in sand would indicate that in the path of the gear, the same number of biogenic depressions that were lost would be recreated within a one year timeframe.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 65 DRAFT April 17, 2009

Recovery values were determined to be more dependent on the intrinsic characteristics of the features themselves, rather than on the gear type causing the effects, except to the extent that gear effects will vary by gear type due to the specific gear components that contact the bottom. There were assumed to be differences between the speeds at which habitat features would recover in high and low energy environments. Such differences allow the modulating effects of natural disturbance to be combined with fishing disturbance in the spatial model. Recovery values under a low energy regime are used when critical sheer stress is below the threshold value, and recovery values under the high energy regime are used if critical sheer stress is above the critical value and depth is less that the critical value (20 m).

The matrix approach is intended to explicitly disaggregate gear impacts by gear type, substrate type, habitat feature, and energy environment. A benefit of this disaggregation is that the susceptibility or recovery values for a particular interaction can be updated as new information becomes available, or varied to allow for sensitivity analyses of the final spatially‐based product. In addition, the disaggregated approach clearly identified gaps in the current knowledge about fishing gear impacts on habitat components.

A key point about the matrix evaluations is that they were intended to represent hypothetical interactions. While some habitat feature‐gear type‐energy combinations are highly likely to occur, others will occur very infrequently. In the spatial model, these uncommon interactions are expected be represented rarely or not at all.

However, for the purposes of estimating susceptibility and recovery values, some habitat feature‐gear type‐energy combinations (as defined in the literature review database) are rarely represented in the literature. These included both combinations expected to be rare in nature, and those that are not commonly studied. For this reason, the suite of studies used to inform a particular susceptibility or recovery value is somewhat broad. The matrix approach allows for judgment on the part of the analyst when the literature does not consider a specific interaction between a habitat feature, gear type, and energy environment. The studies used for evaluation of any particular cell are specified, allowing the end users to make their own judgments about the relationship between the literature and the estimated values.

Separate matrices are used to evaluate impacts to geological, biological, and prey habitat components. For this preliminary document, the geological matrix was evaluated for bottom otter trawls (disaggregated into generic bottom trawl, shrimp, raised footrope, and squid) and New Bedford‐style scallop dredges. The geological matrix for surf clam/ocean quahog dredges, lobster traps, deep‐sea red crab traps, sink gill nets, and bottom longlines will be evaluated at a later date, as will the biological and prey matrices for all gear types.

4.5.1 Geological A generic geological matrix is shown below (Table 21). The geological matrix matches closely with the geological features table, adding columns for gear effects, S, Rhigh, and Rlow values, and

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 66 DRAFT April 17, 2009 for the literature considered. Because susceptibility and recovery of rock outcrop is not really applicable, this feature was removed from the matrix evaluations. Although roughly 40% of the literature reviewed considered gear impacts to geological features, such impacts were often discussed very generally. Impacts to specific features, e.g. cobble piles, were often not addressed in the literature. Therefore, the ‘literature considered’ column of the matrices includes studies that evaluated gear impacts to any geological components in that substrate subclass. This gives the analyst a wider body of studies to draw from, but may overstate the availability of research results addressing particular feature/gear/energy combinations.

Table 21 ‐ Generic geological matrix Gear type Substrate R high R low Literature class/subclass Features Gear effects S energy energy considered resuspension, featureless clay‐silt compression, 0‐3 0‐3 0‐3 geochemical fish, crab, biogenic lobster, scallop filling 0‐3 0‐3 0‐3 depressions Clay‐Silt depressions fish, crab, biogenic burrows filling, crushing 0‐3 0‐3 0‐3 lobster burrows special case clay pipes, filling, crushing 0‐3 0‐3 0‐3 biogenic burrows tilefish burrows bedforms ripples smoothing 0‐3 0‐3 0‐3 Mud resuspension, featureless muddy sand 0‐3 0‐3 0‐3 compression fish, crab, biogenic lobster, scallop filling 0‐3 0‐3 0‐3 depressions depressions Muddy fish, crab, biogenic burrows filling, crushing 0‐3 0‐3 0‐3 Sand lobster burrows special case clay pipes, filling, crushing 0‐3 0‐3 0‐3 biogenic burrows tilefish burrows small (ripples) smoothing 0‐3 0‐3 0‐3 bedforms large (waves) smoothing 0‐3 0‐3 0‐3 resuspension, featureless sand 0‐3 0‐3 0‐3 compression fish, crab, biogenic lobster, scallop filling 0‐3 0‐3 0‐3 Sand depressions depressions small (ripples) smoothing 0‐3 0‐3 0‐3 bedforms large (waves) smoothing 0‐3 0‐3 0‐3 burial, mixing, granule‐pebble 0‐3 0‐3 0‐3 Granule‐Pebble homogenization pebble pavement burial, mixing 0‐3 0‐3 0‐3 cobble displacement 0‐3 0‐3 0‐3 smoothing, Cobble cobble piles 0‐3 0‐3 0‐3 displacement cobble pavement mixing 0‐3 0‐3 0‐3 boulders displacement 0‐3 0‐3 0‐3 Boulder piled boulders displacement 0‐3 0‐3 0‐3 dispersal, Shell deposits 0‐3 0‐3 0‐3 crushing

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 67 DRAFT April 17, 2009 The geological matrices for groundfish trawl, shrimp trawl, squid trawl, raised footrope trawl, and New Bedford‐style scallop dredge are shown below, with S and R values (mean ± sd). Eight analysts evaluated each matrix. The ‘literature considered’ numbers refer to study numbers from the literature review section.

Generally, susceptibility values were higher than recovery values, and within recovery, low energy values were always greater than or equal to high energy values. Recovery values were similar between gear types. Susceptibility values were highest for scallop dredges, followed by groundfish trawls. Shrimp and squid trawls had similar susceptibility values, and raised footrope trawls had the lowest susceptibility values.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 68 DRAFT April 17, 2009 Table 22 – Groundfish trawl matrix with summary S, Rhigh, and Rlow values Gear type: trawl_generic_otter Features Reconciled values Dominant Mean Mean StDev R StDev R Substrate Literature substrate Features Effects Mean R high R low StDev high low subclass considered class S energy energy S energy energy resuspension, compression, featureless clay‐silt geochem, excavation, mixing 2 1 1 0.89 0.74 0.71 biogenic fish, crab, lobster , scallop 17, 92, 97, filling depressions depressions 119, 236, 277, 2 0 1 0.83 0.35 0.92 Clay‐Silt biogenic burrows fish, crab, lobster burrows filling, crushing 333, 335, 336, 2 0 0 0.76 0.71 0.74 special case 338, 372, 406 clay pipes, tilefish burrows filling, crushing, displacement biogenic burrows 2 2 2 0.9 1.21 0.95 bedforms ripples smoothing 3 1 2 0.93 0.76 0.93 Mud resuspension, compression, featureless muddy sand geochem, excavation, mixing 2 0 1 0.83 0.52 0.53 biogenic fish, crab, lobster, scallop 35, 97, 119, filling Muddy depressions depressions 236, 247, 313, 2 0 1 0.83 0.71 1.07 Sand biogenic burrows fish, crab, lobster burrows filling, crushing 320, 330, 336, 2 0 1 0.76 0.71 1.07 special case 360 clay pipes, tilefish burrows filling, crushing, displacement biogenic burrows 2 2 2 0.9 1.21 0.98 bedforms megaripples, waves smoothing 2 0 1 0.89 0.71 0.92 resuspension, compression, 35, 43, 71, 92, featureless sand excavation, mixing 97, 120, 128, 2 0 1 0.99 0.41 0.93 Sand biogenic fish, crab, lobster, scallop 141, 158, 225, filling depressions depressions 247, 291, 325, 2 0 1 0.83 0.71 1.36 bedforms megaripples, waves smoothing 330, 336, 360 2 0 1 0.99 0.35 0.64 Granule‐ granule‐pebble burial, mixing, homogenization 43, 71, 158, 2 1 2 0.71 0.99 1.16 Pebble pebble pavement burial, mixing 225, 247, 336 2 1 2 0.89 0.99 0.93 cobble displacement 2 2 3 0.76 1.16 0.46 Gravel Cobble cobble piles smoothing, displacement 43, 101, 158 3 3 3 0.74 1.06 0 cobble pavement mixing 2 2 2 0.89 1.16 0.92 boulders displacement 2 3 3 0.76 0.71 0 Boulder 101, 247 piled boulders displacement 2 3 3 0.74 0 0 Shell shell deposits displacement, burial, crushing 101, 225 2 1 2 0.93 1.13 1.75

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 69 DRAFT April 17, 2009 Table 23 ‐ Raised footrope trawl matrix with summary S, Rhigh, and Rlow values Gear type: trawl_raised_footrope Features Reconciled values Dominant Mean R Mean R StDev R StDev R substrate Substrate Literature high low StDev high low class subclass Features Effects considered Mean S energy energy S energy energy resuspension, compression, featureless clay‐silt geochem, excavation, mixing 1 1 1 0.74 0.74 0.53 biogenic fish, crab, lobster , scallop 17, 92, 97, 119, filling depressions depressions 236, 277, 333, 1 0 0 0.52 0.35 0.74 Clay‐Silt biogenic burrows fish, crab, lobster burrows filling, crushing 335, 336, 338, 1 0 0 0.46 0.71 0.74 special case 372, 406 clay pipes, tilefish burrows filling, crushing, displacement biogenic burrows 1 2 2 0.53 1.21 0.95 bedforms ripples smoothing 2 0 1 0.64 0.52 0.64 Mud resuspension, compression, featureless muddy sand geochem, excavation, mixing 1 0 1 0.74 0.52 0.53 biogenic fish, crab, lobster, scallop filling 35, 97, 119, Muddy depressions depressions 236, 247, 313, 1 0 0 0.52 0.71 0.74 Sand biogenic burrows fish, crab, lobster burrows filling, crushing 320, 330, 336, 1 0 0 0.46 0.71 0.74 special case 360 clay pipes, tilefish burrows filling, crushing, displacement biogenic burrows 1 2 2 0.53 1.21 0.95 bedforms megaripples, waves smoothing 2 0 1 0.53 0.35 0.64 resuspension, compression, 35, 43, 71, 92, featureless sand excavation, mixing 97, 120, 128, 1 0 1 0.64 0.41 0.49 Sand biogenic fish, crab, lobster, scallop 141, 158, 225, filling depressions depressions 247, 291, 325, 1 0 0 0.52 0.71 0.74 bedforms megaripples, waves smoothing 330, 336, 360 2 0 1 0.53 0.35 0.64 Granule‐ granule‐pebble burial, mixing, homogenization 43, 71, 158, 1 1 2 0.35 0.74 1.07 Pebble pebble pavement burial, mixing 225, 247, 336 1 1 2 0.35 0.76 0.71 cobble displacement 1 2 2 0.53 1.16 1.06 Gravel Cobble cobble piles smoothing, displacement 43, 101, 158 2 3 3 1.04 1.06 1.06 cobble pavement mixing 1 2 2 0.53 1.25 1.2 boulders displacement 1 3 3 0.53 1.06 1.06 Boulder 101, 247 piled boulders displacement 2 3 3 1.07 1.06 1.06 Shell shell deposits displacement, burial, crushing 101, 225 1 1 2 0.46 1.13 1.75

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 70 DRAFT April 17, 2009 Table 24 ‐ Shrimp trawl matrix with summary S, Rhigh, and Rlow values Gear type: trawl_shrimp Features Reconciled values Dominant Mean R Mean R StDev R StDev R substrate Substrate Dominant high low StDev high low class subclass Features Effects substrate class Mean S energy energy S energy energy resuspension, compression, featureless clay‐silt geochem, excavation, mixing 1 1 1 0.92 0.89 0.71 biogenic fish, crab, lobster , filling 17, 92, 97, 119, depressions scallop depressions 2 0 0 0.74 0.35 0.74 236, 277, 333, Clay‐Silt fish, crab, lobster biogenic burrows filling, crushing 335, 336, 338, burrows 2 0 0 0.74 0.71 0.74 372, 406 special case clay pipes, tilefish filling, crushing, displacement biogenic burrows burrows 2 2 2 0.9 1.21 0.95 bedforms ripples smoothing 2 0 1 0.83 0.52 0.71 Mud resuspension, compression, featureless muddy sand geochem, excavation, mixing 1 0 1 0.92 0.52 0.74 biogenic fish, crab, lobster, filling 35, 97, 119, depressions scallop depressions 2 0 0 0.74 0.71 0.74 Muddy 236, 247, 313, fish, crab, lobster Sand biogenic burrows filling, crushing 320, 330, 336, burrows 2 0 0 0.74 0.71 0.74 360 special case clay pipes, tilefish filling, crushing, displacement biogenic burrows burrows 2 2 2 0.90 1.21 0.95 bedforms megaripples, waves smoothing 2 0 1 0.74 0.35 0.71 resuspension, compression, 35, 43, 71, 92, featureless sand excavation, mixing 97, 120, 128, 1 0 1 0.46 0.41 1.00 Sand biogenic fish, crab, lobster, 141, 158, 225, filling depressions scallop depressions 247, 291, 325, 2 0 1 0.74 0.71 1.16 bedforms megaripples, waves smoothing 330, 336, 360 2 0 1 0.76 0.35 0.76 Granule‐ granule‐pebble burial, mixing, homogenization 43, 71, 158, 1 1 2 0.46 0.71 1.07 Pebble pebble pavement burial, mixing 225, 247, 336 1 1 2 0.74 0.76 0.74 cobble displacement 1 2 3 0.35 1.13 0.46 Gravel Cobble cobble piles smoothing, displacement 43, 101, 158 2 3 3 0.92 1.06 0.00 cobble pavement mixing 1 2 2 0.74 1.25 0.92 boulders displacement 1 3 3 0.53 0.71 0.00 Boulder 101, 247 piled boulders displacement 2 3 3 0.76 0.00 0.00 Shell shell deposits displacement, burial, crushing 101, 225 2 1 2 1.07 1.28 1.06

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 71 DRAFT April 17, 2009 Table 25 ‐ Squid trawl matrix with summary S, Rhigh, and Rlow values Gear type: trawl_squid Features Reconciled values Dominant Mean R Mean R StDev R StDev R substrate Substrate Dominant high low StDev high low class subclass Features Effects substrate class Mean S energy energy S energy energy resuspension, compression, featureless clay‐silt geochem, excavation, mixing 2 1 1 0.92 0.74 0.52 biogenic fish, crab, lobster , 17, 92, 97, 119, depressions scallop depressions filling 2 0 0 0.76 0.46 0.74 236, 277, 333, Clay‐Silt fish, crab, lobster biogenic burrows 335, 336, 338, burrows filling, crushing 1 0 0 0.74 0.46 0.74 372, 406 special case clay pipes, tilefish biogenic burrows burrows filling, crushing, displacement 2 2 2 0.79 1.07 0.95 bedforms ripples smoothing 2 1 1 0.89 0.53 0.64 Mud resuspension, compression, featureless muddy sand geochem, excavation, mixing 2 0 1 0.92 0.52 0.53 biogenic fish, crab, lobster, 35, 97, 119, depressions scallop depressions filling 2 0 0 0.76 0.46 0.74 Muddy 236, 247, 313, fish, crab, lobster Sand biogenic burrows 320, 330, 336, burrows filling, crushing 1 0 0 0.74 0.46 0.76 360 special case clay pipes, tilefish biogenic burrows burrows filling, crushing, displacement 2 2 2 0.79 1.07 0.95 bedforms megaripples, waves smoothing 1 0 1 0.74 0.52 0.64 resuspension, compression, 35, 43, 71, 92, featureless sand excavation, mixing 97, 120, 128, 2 0 1 0.76 0.49 0.46 Sand biogenic fish, crab, lobster, 141, 158, 225, depressions scallop depressions filling 247, 291, 325, 2 0 0 0.76 0.46 0.74 bedforms megaripples, waves smoothing 330, 336, 360 1 0 1 0.74 0.46 0.76 Granule‐ granule‐pebble burial, mixing, homogenization 43, 71, 158, 1 1 2 0.74 0.71 1.07 Pebble pebble pavement burial, mixing 225, 247, 336 1 1 2 0.74 0.76 0.74 cobble displacement 1 2 3 0.89 1.19 1.07 Gravel Cobble cobble piles smoothing, displacement 43, 101, 158 2 3 3 0.99 1.06 1.06 cobble pavement mixing 1 2 2 0.89 1.25 1.20 boulders displacement 1 3 3 0.89 1.06 1.06 Boulder 101, 247 piled boulders displacement 2 3 3 0.99 1.06 1.06 Shell shell deposits displacement, burial, crushing 101, 225 2 1 2 0.76 1.25 1.07

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 72 DRAFT April 17, 2009 Table 26 ‐ New Bedford‐style scallop dredge matrix with summary S, Rhigh, and Rlow values Gear type: dredge_scallop Features Reconciled values Dominant Mean R Mean R StDev R StDev R substrate Substrate Dominant high low StDev high low class subclass Features Effects substrate class Mean S energy energy S energy energy resuspension, compression, featureless clay‐silt geochem, excavation, mixing 2 1 1 0.71 0.92 0.64 biogenic fish, crab, lobster , depressions scallop depressions filling 3 0 1 0.53 0.74 1.07 Clay‐Silt fish, crab, lobster 11, 236, 391 biogenic burrows burrows filling, crushing 3 0 1 0.53 0.74 1.07 special case clay pipes, tilefish biogenic burrows burrows filling, crushing, displacement 3 2 2 0.49 0.95 0.98 bedforms ripples smoothing 3 1 2 0.76 0.76 0.92 Mud resuspension, compression, featureless muddy sand geochem, excavation, mixing 2 0 1 0.71 0.74 0.83 biogenic fish, crab, lobster, depressions scallop depressions filling 3 0 1 0.53 0.74 1.07 Muddy fish, crab, lobster 11, 236, 391 Sand biogenic burrows burrows filling, crushing 3 0 0 0.53 0.74 1.13 special case clay pipes, tilefish biogenic burrows burrows filling, crushing, displacement 3 2 2 0.49 0.95 0.98 bedforms megaripples, waves smoothing 2 1 1 0.92 0.74 0.92 resuspension, compression, featureless sand excavation, mixing 2 0 1 0.99 0.49 0.83 69, 71, 158, Sand biogenic fish, crab, lobster, 225, 352 depressions scallop depressions filling 2 0 1 0.52 0.74 1.36 bedforms megaripples, waves smoothing 3 0 1 0.76 0.46 0.64 Granule‐ granule‐pebble burial, mixing, homogenization 69, 71, 158, 2 1 2 0.74 0.93 0.89 Pebble pebble pavement burial, mixing 225, 352 2 1 2 0.74 0.99 0.83 cobble displacement 2 2 3 0.93 1.16 0.46 Gravel Cobble cobble piles smoothing, displacement 158, 352 3 3 3 0.35 1.06 0.00 cobble pavement mixing 2 2 2 0.52 1.16 0.89 boulders displacement 2 3 3 0.99 0.93 0.71 Boulder 352 piled boulders displacement 3 3 3 0.74 0.00 0.00 Shell shell deposits displacement, burial, crushing 225, 352 2 1 2 0.74 1.13 1.63

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 73 DRAFT April 17, 2009 4.5.2 Biological A generic biological matrix is shown below (Table 27). It follows directly from the biological feature table presented earlier. Biological matrices have not been evaluated yet. However, the gear impacts literature is, generally speaking, more biologically focused (including biological and prey components as defined for this analysis) than geologically focused, so matrix evaluation is anticipated to be relatively straightforward.

Table 27 Generic biological matrix Gear type R high R low Literature Feature Gear Effect S energy energy considered Sponges (emergent) slicing, crushing, detachment 0‐3 0‐3 0‐3 Sponges (encrusting) scraping 0‐3 0‐3 0‐3 Hydroids (emergent) detachment 0‐3 0‐3 0‐3 Anemones (emergent) slicing, crushing, detachment 0‐3 0‐3 0‐3 Anemones (emergent, burrowing) slicing, crushing, detachment 0‐3 0‐3 0‐3 Soft corals: Gorgonians and soft corals (emergent) slicing, crushing, detachment 0‐3 0‐3 0‐3 Sea pens (emergent) slicing, crushing, detachment 0‐3 0‐3 0‐3 Hard corals: stony corals and black corals (encrusting) slicing, crushing, detachment 0‐3 0‐3 0‐3 Colonial tube worms (emergent) slicing, crushing, detachment 0‐3 0‐3 0‐3 Tube‐building amphipods (emergent) slicing, crushing, detachment 0‐3 0‐3 0‐3 Bivalves (emergent) crushing, detachment 0‐3 0‐3 0‐3 Bryozoans (emergent) detachment 0‐3 0‐3 0‐3 Bryozoans (encrusting) scraping 0‐3 0‐3 0‐3 Brachiopods (emergent) slicing, crushing, detachment 0‐3 0‐3 0‐3 Tunicates (emergent) slicing, crushing, detachment 0‐3 0‐3 0‐3 Macroalgae (emergent) slicing, detachment 0‐3 0‐3 0‐3 Macroalgae (encrusting) slicing, detachment 0‐3 0‐3 0‐3 Sea grass (emergent) slicing,detachment 0‐3 0‐3 0‐3

4.5.3 Prey A generic prey matrix is shown below (Table 28). It follows directly from the prey feature table presented earlier. Prey matrices have not been evaluated yet. However, the gear impacts literature is, generally speaking, more biologically focused (including biological and prey components as defined for this analysis) than geologically focused, so matrix evaluation is anticipated to be relatively straightforward.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 74 DRAFT April 17, 2009

Table 28 ‐ Generic prey matrix Gear type R high R low Literature Prey Gear effect S energy energy considered Amphipods crushing 0‐3 0‐3 0‐3 Anemones slicing, crushing, detachment 0‐3 0‐3 0‐3 Brittle stars slicing, crushing 0‐3 0‐3 0‐3 Decapod crabs crushing 0‐3 0‐3 0‐3 Decapod shrimp crushing 0‐3 0‐3 0‐3 Isopods crushing 0‐3 0‐3 0‐3 Molluscs crushing, detachment 0‐3 0‐3 0‐3 Mysids crushing 0‐3 0‐3 0‐3 Polychaetes slicing, crushing 0‐3 0‐3 0‐3 Sea urchins/sand dollars crushing 0‐3 0‐3 0‐3 Starfish slicing, crushing 0‐3 0‐3 0‐3

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 75 DRAFT April 17, 2009 5.0 Estimating effective fishing effort (Swept Area Seabed Impact model)

5.1.1 Methods In order to (1) quantify fishing effort in like terms and (2) compare the relative effects of different fishing gears, fishing effort is converted to area swept units (km2) using the Swept Area Seabed Impact (SASI) model. The SASI model provides a measure of the area of seabed contacted by one unit of effort of a particular fishing gear (e.g. one tow, gillnet set, line of hooks, traps etc.). It uses area swept as a proxy for direct seabed impact. The output of the model is a contact‐adjusted area swept, measured in km2.

The SASI model incorporates the components of different fishing gears individually, allowing the user to tease out the relative contribution that each component makes to the area swept by the gear as a whole. By modifying the contact parameters for each component, the model can ‘reward’ gears that are modified to reduce seabed contact (e.g. those designed to skim over the seabed, or with raised ground gear). The model estimates contact‐adjusted area swept at the level of the individual tow. Regardless of gear type, the SASI model requires three things:

(1) total distance towed, or, in the case of fixed gears, total length of the gear; (2) width of the individual gear components; and (3) contact indices for the various gear components.

The contact index (c) is a measure of the overall contact width of the various gear components. It makes an allowance for the fact that the entire width of the gear may not be in contact with the seabed. Contact indices are categorically specified by gear type, and may be revised in the future to accommodate additional data (e.g. for new or modified gear types, or for a better understanding of how a gear interacts with the sea bed while fishing).

The SASI model can be further modified to compare the impact of a particular fishing gear configuration on various substrate types via adjustment of the sensitivity indices for each gear component. This is approach is a vital component of the FiGSI model, and is discussed in greater detail later in this document.

The following sections describe the parameterization of SASI models for five styles of fishing gear: (1) demersal otter trawls, with four otter trawl subtypes differentiated by varying contact indices; (2) New Bedford‐style scallop dredges; (3) hydraulic clam/quahog dredges; (4) lobster and deep‐sea red crab traps; and (5) sink gill nets and demersal longlines.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 76 DRAFT April 17, 2009 5.1.1.1 Demersal otter trawl A demersal trawl has three key components that potentially contribute to seabed impact: the otter boards, the ground cables, and the sweep. These components are identified based on their potentially differing impacts on habitat features. A single model is used for all otter trawl types, including groundfish, shrimp, squid, and raised footrope. The SASI model for a demersal trawl is:

SASI (m2) = dt[(2∙wo∙co) + (2∙wc∙cc) + (ws∙cs)]

dt = distance towed in one tow (m) wo = effective width of an otter board (m), which equals otter board length (m)∙sin (αo), where αo = angle of attack (ranging from 30o to 50 o) co = contact index, otter board wc = effective width of a ground cable (km), which equals ground cable length (m)∙sin(αc), where αc = angle of attack (ranging from 10o to 20 o) cc = contact index, ground cables ws = effective width of sweep (m) cs = contact index, sweep

The angle of attack (α) of an otter board can be determined at sea by measuring the scratch marks on the shoe of the otter board at the completion of a tow. If this is not possible, an assumed value of α can be utilized ranging between 30o and 50 o (Gomez and Jimenez 1994). The angle of attack of a ground cable varies along its length, and cannot be accurately measured at sea. This angle is typically assumed to range between 10o and 20 o (Gomez and Jimenez 1994, Baranov 1969). The effective width of a sweep can only be measured at sea using acoustic mensuration sensors; typically this width is about 40% of total sweep length.

The demersal trawl SASI model assumes the following: • Fishing gear impact is constant within a tow • There is constant impact along the entire length of a gear component • The impact of each gear component is cumulative • A gear component has the same impact on the epibethos and infauna irrespective of its size, length, weight, design and rigging, unless it translates to reduced seabed impact (contact index) • Seabed topography and composition are consistent within a tow • The abundance of epibenthos and infauna within a tow is uniform • Otter board angle of attack is constant during a tow • Ground cables are straight along their entire length • Seabed contact does not change within a tow • The effect of towing speed on seabed contact is accommodated by dt

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 77 DRAFT April 17, 2009 5.1.1.2 New Bedford‐style scallop dredge A scallop dredge has five key components that potentially contribute to seabed impact. They are: the contact shoes; the dredge bale arm including cutting bar; the bale arm rollers; the chain sweep; and the ring bag and club stick. These components impact the seabed and benthic fauna differently. However, if each gear component were to be modeled separately, the model will substantially overestimate contact‐adjusted area swept. Unlike for trawl gear, additional dredge components do not add width to the area swept because they follow one behind the other as the gear is towed. Therefore, the dredge model shown below does not consider the potential impact of individual components of a dredge, but groups them together. While this does not allow for modification to the contact indices of various gear components, such modifications could be accommodated by varying the sensitivity index (discussed later in this document).

Given these simplifying assumptions, the scallop dredge SASI model is:

SASI (m2) = dt (wrb∙crb)

dt = distance towed in one tow (m) wrb = effective width of widest dredge component (m) c = contact index, all dredge components

Similar to the otter trawl model, the scallop dredge SASI assumes the following: • Fishing gear impact is constant within a tow • There is constant impact along the entire length of a gear component • A gear component has the same impact on the epibethos and infauna irrespective of its size, length, weight, design and rigging, unless it translates to reduced seabed impact (contact index) • Seabed topography and composition is consistent within a tow (although this could potentially be picked up in the sensitivity index) • The abundance of epibenthos and infauna within a tow is uniform • Seabed contact does not change within a tow • The effect of towing speed on seabed contact is accommodated by dt

The contact index is assumed to be 1.0.

5.1.1.3 Hydraulic clam dredge Similar to the scallop dredge model, the hydraulic clam dredge model shown below does not consider the potential impact of individual components of a dredge, but groups them together. While this does not allow for modification to the contact indices of various gear components, such modifications could be accommodated by varying the sensitivity index (discussed later in this document). Given these simplifying assumptions, the hydraulic clam dredge SASI model is:

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 78 DRAFT April 17, 2009

SASI (m2) = dt (wrb∙crb)

dt = distance towed in one tow (m) wrb = effective width of widest dredge component (m) c = contact index, all dredge components

Similar to the otter trawl model, the scallop dredge SASI assumes the following: • Fishing gear impact is constant within a tow • There is constant impact along the entire length of a gear component • A gear component has the same impact on the epibethos and infauna irrespective of its size, length, weight, design and rigging, unless it translates to reduced seabed impact (contact index) • Seabed topography and composition is consistent within a tow (although this could potentially be picked up in the sensitivity index) • The abundance of epibenthos and infauna within a tow is uniform • Seabed contact does not change within a tow • The effect of towing speed on seabed contact is accommodated by dt

5.1.1.4 Lobster traps and deep‐sea red crab traps The SASI model for a line or trawl of n lobster traps, accounting for each individual trap and ground line between traps is:

SASI (m2) = ∑[dtn∙w tn∙c tn] + ∑[drn∙wrn∙crn]

n = 1 ‐ ∞ dtn = distance nth trap moves over the seabed (m) wtn = effective contact patch (width x length) of nth trap (m2) ctn = contact index, nth trap drn = distance the nth ground line moves over the seabed (m) wrn = effective contact patch of nth ground line (m2) crn = contact index, nth rope

The distance that each gear component moves is a function of movements over the seabed both while the gear is fishing (soaking) and during the hauling process, although the extent of these movements is unknown.

5.1.1.5 Demersal longline and gill net A demersal longline or gillnet has two key components that potentially contribute to seabed impact: the weights and the line. Again, these components are identified based on their potentially differing impact on the seabed and benthic fauna.

The SASI model for a demersal longline or gillnet is:

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 79 DRAFT April 17, 2009

SASI (m2) = (d1∙w1∙c1)+(d2∙w2∙c2)+(dr∙wr∙cr)

d1 = distance end‐weight #1 moves over the seabed (m) ws = effective contact patch of weight #1 (m2) cs = contact index, weight #1 d2 = distance end‐weight #2 moves over the seabed (m) ws = effective contact patch of weight #2 (m2) cs = contact index, weight #2 dr = distance longline or leadline moves over the seabed (m) wr = effective contact patch of longline or leadline (m2) cr = contact index, longline or leadline

The distance that each gear component moves is a function of movements over the seabed both while the gear is fishing (soaking) and during the hauling process, although the extent of these movements is unknown.

5.1.2 Data and parameterization To parameterize the SASI model for trawl and scallop dredge gears, the individual gear component widths, tow durations, and contact indices must be estimated. Two primary data sets are available for this purpose: on‐board observations taken by fishery observers (observer data), and self‐reported observations from fishermen via the Vessel Trip Report (VTR data) program.

The observer program collects specific data on trawl net configurations and dimensions, as well as towing speeds. For these analyses, observer program data are used to: (1) estimate the speed at which gear is towed; and (2) estimate the size of the three primary gear components. Observer data, however, is a subset of the overall fishery and not representative of total fishing effort. VTR data are the only synoptic data source for vessel activity, area fished, and fishing effort for commercial fisheries. Therefore, gear size and towing speed estimates generated using observer‐based data are applied to VTR data to derive overall area swept estimates from the SASI model.

Estimating tow speeds Tow speed is reported via the observer program. For both trawl and scallop dredge fishing tows, the speed at which the gear is towed exhibits little variation. All trawl trips were assumed to tow at 3.0 knots, while scallop dredge trips were assumed to tow at 4.4 knots for all years prior to 2004, 4.5 knots for trips taken in 2004, 4.6 knots for trips taken from 2005 to 2007, and 4.7 knots for trips taken in 2008 (Table 29).

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 80 DRAFT April 17, 2009 Table 29 – Tow speeds by year and fishing gear type scallop dredge otter trawl YEAR N Mean Var N Mean Var 2003 5266 4.4 0.215 7152 3 0.145 2004 8276 4.5 0.152 10567 3 0.12 2005 6109 4.6 0.169 26449 3 0.107 2006 5998 4.6 0.202 13339 3 0.101 2007 7546 4.6 0.189 14733 3 0.1 2008 8103 4.7 0.108 8775 3 0.13

Estimating gear component sizes Within the SASI model, trawl gears are comprised of three gear components: otter boards, ground cables and sweep. The specific parameters estimated are: 1) wo , the effective width of an otter board (m), which equals otter board length (m)∙sin (αo), where αo = angle of attack (ranging from 30o to 50 o) 2) wc , the effective width of a ground cable (km), which equals ground cable length (m)∙sin(αc), where αc = angle of attack (ranging from 10o to 20 o) 3) ws, , the effective width of sweep (m)

For (1), αo is assumed to be 40 o. Otter board weight data is collected through the observer program, but dimensions are not. Using commercially available data on the size and weight of otter boards for two different door designs (Thyboron Type II and Bison, both distributed by Trawlworks, Inc of Narragansett RI), a linear relationship between otter board weight and otter board length was established (Figure 4). The type and brand of otter boards used in the fishery are not reported, and it is not known if this sample is representative of the gear used on observed trips, or in the fishery as a whole.

Figure 4 – Linear regression of otter board length on otter board weight Analysis of Variance

Sum of Mean Source DF Squares Square F Value Pr > F

Model 1 3573631 3573631 303.61 <.0001 Error 24 282493 11771 Corrected Total 25 3856124

Root MSE 108.49211 R-Square 0.9267 Dependent Mean 1995.19231 Adj R-Sq 0.9237 Coeff Var 5.43768

Parameter Estimates

Parameter Standard Variable DF Estimate Error t Value Pr > |t|

Intercept 1 1223.66251 49.12562 24.91 <.0001 AVG_WEIGHT_LBS 1 0.83332 0.04783 17.42 <.0001

This relationship provides an estimate of otter board length for each observed trip, which must be applied to the universe of fishing activity by constructing a relationship between reported door weight and a variable or variables common between both observer and VTR datasets. Several relationships were investigated. A significant and

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 81 DRAFT April 17, 2009 relatively strong linear relationship was observed between door weight and a combination of gross tonnage and horsepower (Figure 5).

Figure 5 – Linear regression of otter board weight on vessel gross tonnage and vessel horsepower, observer data 2003‐2008 Parameter Estimates

Parameter Standard Variance Variable Label DF Estimate Error t Value Pr > |t| Inflation

Intercept Intercept 1 70.84823 7.75592 9.13 <.0001 0 GTONS GTONS 1 1.84431 0.09525 19.36 <.0001 2.40978 VHP VHP 1 0.53446 0.02173 24.59 <.0001 2.40978

Collinearity Diagnostics

Condition ------Proportion of Variation------Number Eigenvalue Index Intercept GTONS VHP

1 2.81590 1.00000 0.02071 0.01190 0.00919 2 0.13667 4.53920 0.84689 0.22238 0.03041 3 0.04744 7.70439 0.13240 0.76572 0.96040

Applying this relationship to all VTR‐reported trips using otter trawls provides an estimate of door weights. Applying the modeled relationship between otter board weight and otter board length, and correcting for angle of attack provides an estimate of the effective linear width of otter boards used for each trip.

For (2) αo is assumed to be 15 o. Ground cable length data are collected directly through the observer program. Relationships between ground cable length and independent variable common between both observer and VTR datasets were investigated. A significant but weak linear relationship was observed between ground cable length and vessel length (Figure 6).

Figure 6 – Linear regression of ground cable length on vessel length, observer data 2003‐2008 Analysis of Variance

Sum of Mean Source DF Squares Square F Value Pr > F

Model 1 92928 92928 209.32 <.0001 Error 2960 1314125 443.96129 Corrected Total 2961 1407054

Root MSE 21.07039 R-Square 0.0660 Dependent Mean 47.63079 Adj R-Sq 0.0657 Coeff Var 44.23691

Parameter Estimates

Parameter Standard Variable Label DF Estimate Error t Value Pr > |t|

Intercept Intercept 1 23.34782 1.72249 13.55 <.0001 LEN LEN 1 0.37242 0.02574 14.47 <.0001

Applying this relationship to all VTR‐reported trips using otter trawls provides an estimate of ground cable length, and correcting for angle of attack provides an estimate of the effective linear width of ground cables used for each trip.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 82 DRAFT April 17, 2009 For (3), the effective linear width of the sweep is modeled as the diameter of a circle with a perimeter of 2 * sweep length. Sweep length data are collected directly through the observer program, and are reported as part of the VTR data as well. Due to extremely high variance in the VTR reporting for this parameter, stemming perhaps from inconsistent reporting units (meters, feet, inches, etc), observer data were once again used to estimate sweep length and relationships between sweep length and independent variable common between the two datasets were investigated. A significant and relatively strong relationship was observed between sweep length and both vessel gross tonnage and vessel horsepower. Some collinearity was observed between the two dependent variables, but nonetheless the model was deemed superior to any similar model using only one dependent variable based on the proportion of variance explained (Figure 7).

Figure 7 – Linear regression of sweep length on vessel gross tonnage and horsepower, observer data 2003‐ 2008 Analysis of Variance

Sum of Mean Source DF Squares Square F Value Pr > F

Model 2 3319562 1659781 1247.97 <.0001 Error 4154 5524735 1329.97954 Corrected Total 4156 8844297

Root MSE 36.46888 R-Square 0.3753 Dependent Mean 105.74498 Adj R-Sq 0.3750 Coeff Var 34.48758

Parameter Estimates

Parameter Standard Variance Variable Label DF Estimate Error t Value Pr > |t| Inflation

Intercept Intercept 1 46.35534 1.33193 34.80 <.0001 0 GTONS GTONS 1 0.22521 0.01636 13.77 <.0001 2.40978 VHP VHP 1 0.07616 0.00373 20.41 <.0001 2.40978

Collinearity Diagnostics

Condition ------Proportion of Variation------Number Eigenvalue Index Intercept GTONS VHP

1 2.81590 1.00000 0.02071 0.01190 0.00919 2 0.13667 4.53920 0.84689 0.22238 0.03041 3 0.04744 7.70439 0.13240 0.76572 0.96040

Applying this relationship to all VTR‐reported trips using otter trawls provides an estimate of sweep length, which is modeled as ½ the perimeter of a circle to calculate a corresponding diameter, providing an estimate of the effective linear width of the sweep used for each trip.

Importantly, the observer program does not sample all fisheries and gear types evenly. The distribution of trips in terms of size (horsepower, length) and fishing locations (latitude, longitude) for observer and VTR data are significantly different for trips made with trawl gears (Table 30).

Data used to generate parameter estimates for the SASI model may be biased upward with respect vessel size. The magnitude and direction of bias resulting from the fishing

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 83 DRAFT April 17, 2009 location differences between the two datasets is unclear, though any average depth and substrate type variation across latitudes and longitudes may play a role in the configuration of trawl gears and their dimensions. Year effects cannot be ruled out, as these analyses include the years 1996 – 2008 while observer data is only available from 2003 onward.

Table 30 –Independent group t‐test for observer‐reported trips made between 2003‐2008 with trawl gears, and VTR‐reported trips for the same years; paired records discarded from VTR group (Class 1 = VTR, Class 2 = OBS) Lower Upper Lower Upper CL CL CL CL Std Std Variable class N Mean Mean Mean Dev Std Dev Dev Std Err Minimum Maximum VHP 1 1.64E+05 403.73 404.70 405.67 199.68 200.36 201.05 0.495 25.0 2985.0 VHP 2 4664 489.77 496.39 503.02 226.32 230.91 235.70 3.381 54.0 2775.0 VHP Diff (1‐2) ‐97.55 ‐91.69 ‐85.84 200.59 201.27 201.95 2.989 LEN 1 1.64E+05 56.79 56.87 56.94 14.81 14.86 14.91 0.037 18.0 138.0 LEN 2 4664 64.82 65.25 65.68 14.68 14.98 15.29 0.219 32.0 138.0 LEN Diff (1‐2) ‐8.81 ‐8.38 ‐7.95 14.82 14.87 14.92 0.221 GTONS 1 1.64E+05 64.08 64.31 64.53 46.69 46.85 47.01 0.116 0.0 476.0 GTONS 2 4664 93.22 94.75 96.27 52.19 53.25 54.36 0.780 3.0 246.0 GTONS Diff (1‐2) ‐31.81 ‐30.44 ‐29.07 46.88 47.04 47.20 0.699 lat 1 1.17E+05 41.06 41.07 41.08 1.65 1.65 1.66 0.005 35.0 44.6 lat 2 4658 41.09 41.13 41.17 1.35 1.38 1.41 0.020 35.2 43.9 lat Diff (1‐2) ‐0.11 ‐0.07 ‐0.02 1.64 1.64 1.65 0.025 lon 1 1.17E+05 71.52 71.53 71.54 1.80 1.81 1.81 0.005 65.6 77.3 lon 2 4658 70.43 70.49 70.55 2.10 2.14 2.19 0.031 66.5 76.5 lon Diff (1‐2) 0.99 1.04 1.10 1.81 1.82 1.83 0.027

T‐Tests Variable Method Variances DF t Value Pr > |t| VHP Pooled Equal 1.70E+05 ‐30.68 <.0001 LEN Satterthwaite Unequal 4927 ‐37.68 <.0001 GTONS Pooled Equal 1.70E+05 ‐43.58 <.0001 lat Pooled Equal 1.20E+05 ‐2.69 0.0071 lon Pooled Equal 1.20E+05 38.32 <.0001 Equality of Variances Variable Method Num DF Den DF F Value Pr > F VHP Folded F 4663 1.64E+05 1.33 <.0001 LEN Folded F 4663 1.64E+05 1.02 0.4485 GTONS Folded F 4663 1.64E+05 1.29 <.0001 lat Folded F 1.17E+05 4657 1.43 <.0001 lon Folded F 4657 1.17E+05 1.41 <.0001

For New Bedford‐style scallop dredge gears, the SASI model requires estimates for the size of dredges and the number of dredges. Both data fields are included in the VTR data.

Contact indices were specified for trawl gears (Table 31).

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 84 DRAFT April 17, 2009

An index of 1.0 was assumed for scallop dredge gear.

Table 31 ‐ Trawl contact indices Component Groundfish Raised footrope Shrimp Squid Doors 1.00 1.00 1.00 1.00 Ground cable 0.95 0.95 0.90 0.95 Sweep 0.90 0.05 0.95 0.50

5.1.3 Results ‐‐To be completed‐‐

5.1.4 Discussion ‐‐To be completed‐‐

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 85 DRAFT April 17, 2009 6.0 Estimating spatially‐defined habitat and energy (Spatial model) A spatially‐specified approach is employed to define habitats based on dominant substrates and environmental energy as determined by water flow and naturally induced particle movement along the seabed.

6.1.1 Base grid The spatial domain includes US Federal waters from Cape Hatteras to the US‐Canada border (Figure 8). The X‐Y locations of the geological data (details below) are tessellated to create a Voronoi diagram or grid composed of polygons. Each polygon is convex, and defined by the perpendicular bisectors of lines drawn between geological data points such that each polygon bounds the region closest to that data point relative to all others (Thiessen and Alter 1911, Gold 1991, Okabe et al. 1992, Legendre and Legendre 1998). Voronoi diagrams have been used in terrestrial and aquatic ecological studies and are particularly useful when data are spatially clustered (Isaaks and Srivastava 1989, Fortin and Dale 2005).

The type, quantity and quality of data available for habitat analyses vary a great deal throughout the model domain. The advantage of this base grid is that the resulting unsmoothed interpolated surface consists of cells maintain the spatial characteristics of the source data. For example, where geological sampling is spare the polygons are large. This is in contrast to mathematical interpolations which result in a standardized grid despite the spatial resolution of the source data (e.g. spline, inverse distance weighting or kriging). Further this approach allows the SASI model to be implemented on a cell‐ by‐cell basis.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 86 DRAFT April 17, 2009 Figure 8 – Voronoi diagram of geological data points used as base grid in spatial model

6.1.2 Data and parameterization

6.1.3 Substrate classification and data sources The substrate classes and subclasses used in the analysis are defined by particle size according to the Wentworth particle scale (Table 32, Wentworth 1922). The subclasses ‘clay‐silt’ and ‘muddy‐sand’ are collapsed into the class ‘mud’.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 87 DRAFT April 17, 2009 Table 32 – Substrate subclasses by particle size range Substrate class Substrate subclass Particle size range (mm) Corresponding Wentworth class < 0.0039 Clay Clay‐silt Mud 0.0039 – 0.0625 Silt Muddy‐sand < 0.0039 ‐ 2 Clay to sand Sand Sand/sand ripple 0.0625 – 2 Sand 2‐4 Gravel Granule‐pebble 4 – 64 Pebble Cobble 64 – 256 Cobble Boulder > 256 Boulder

To generate the Voronoi diagram, geological substrates are first organized into the following five classes according to the above scale: mud, sand/sand ripple, granule‐ pebble, cobble, and boulder. Substrate data are assembled from three primary sources:

1) The SMAST video survey (Stokesbury 2002, Stokesbury et al. 2004, Stokesbury and Harris 2006, Stokesbury et al 2007); 2) The usSEABED extracted and parsed datasets from the U.S. Geological Survey; and, 3) To address the paucity of boulder data outside the domain of the SMAST survey, NOAA survey trawl hangs (codes 5 and 9) are included as indicators of boulder‐ equivalent habitats (Link and Demarest 2003).

Only substrate data with positive location, time and sampling device metadata are used. Not all data sources provide information based on sampling capable of detecting all five dominate substrate classes; for example, much of the substrate data compiled in the usSEABED database was collected using grab and core samplers that are typically not capable of sampling grain sizes larger than those coded in Table 32 as Granule‐pebble. The NOAA survey trawl hangs dataset is the most extreme example: no substrate samples are actually made and hangs are assumed to be boulders, and the sampling device is not capable of detecting the other four substrate classifications. These sampling limitations are coded into the geological datasets R_sub value which is a ratio of detectable substrate types to total types (5). For example, the SMAST optical survey technique R‐sub = 5/5 because it detects all 5 substrate classes, while the usSEABED R_sub = 0.6 datasets 3/5 because cobbles and boulders are not detected.

To achieve a logistically practical grid a minimum cell area of 3 km2 was imposed. In cases where more than one geological sample occurs in a cell (indicated by the N_sub column in the data) the mean dominant substrate value was used and the N=sub and SD_sub field indicate the number of coded substrate values per cell and standard deviation of the mean respectively.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 88 DRAFT April 17, 2009 6.1.3.1 SMAST video survey The SMAST Video Survey uses a multi‐stage quadrat‐based sampling design and a dual‐ view video quadrat for all surveys. Survey stations are arranged as grids based on random starting points, with the resolution (distance between stations) calculated to obtain estimates of the dominant macrobenthic species density (sea scallops m‐2) with a precision of 5 to 15% for the normal and negative binomial distributions respectively (Stokesbury 2002). At each station four replicate video‐quadrats are sampled haphazardly with a steel pyramid lander equipped with underwater cameras and lighting (for details see Stokesbury 2002, Stokesbury et al. 2004).

The SMAST database presently includes 143,327 quadrat samples from 76,084 stations covering 65,675 km2 of USA continental shelf including Stellwagen Bank, Georges Bank from the Northern Edge to the Great South Channel and the Mid‐Atlantic Bight. The SMAST survey uses three live‐feed S‐VHS underwater video cameras, two in plan‐view and one in parallel‐view. The two plan‐view cameras sample 3.235 m2 (large camera) and 0.8 m2 (small camera) quadrats respectively, with the small camera view nested within the large camera view. The parallel‐view camera (side camera) provided a cross‐ quadrat view of both large and small camera sample areas and is used to validate the quadrat observations.

Each quadrat is characterized as containing silt, sand, sand ripple, granule‐pebble, cobble, and/or boulder substrates based on particle diameters from the Wentworth scale (Wentworth 1922, Lincoln et al. 1992). Sand and sand ripple are mutually exclusive classes. Substrates are visually identified in real time using texture, color, relief and structure observed in all three camera views. All video footage is reviewed in the laboratory where analysts digitize and catalog a still frame from the large and small camera footage at each quadrat and verify substrate identification.

The SMAST database spatial sampling scales are grain (quadrat) = 3.24 m2, mean sub‐ interval (between quadrats) = 22 m (SD 20.2 m), mean interval (between stations) = 3.6 km (SD 2.01 km) and the extent of the surveys ranged from 502.8 km2 (individual survey) ‐ 65,675 km2 (all surveys). The temporal sampling scales are grain = 0.25 – 0.5 minutes (quadrats), sub‐interval 5 – 15 minutes (between quadrats), interval 0.5 – 1 hours / 5 – 10 days (between stations / between surveys) and extent years (all surveys).

There are strengths and limitations to the dataset for mapping purposes. Strengths include: 1. Formal sampling design with replication. 2. Multiview optic sample of sand to boulder substrates 3. High spatial sampling frequency 4. Annual sampling of Georges Bank and the Mid‐Atlantic since 2003

Limitations include:

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 89 DRAFT April 17, 2009 1. Database includes only surficial geology and does not include particles finer than silt. 2. Surveys do not include the GOM or depths greater than 200m.

6.1.3.2 usSEABED database The usSEABED database contains a compilation of published and unpublished sediment texture and other geologic data about the seafloor from numerous projects (Reid et al 2005). The USGS DS 118 Atlantic Coast data extend from the U.S./Canada border (northern Maine) to Key West Florida, including some Great Lakes, other lakes, and some rivers, beaches, and estuaries. The database is built using more than 150 data sources containing more than 200,000 data points distributed across the five output data files.

Extracted (numeric, lab‐based) and Parsed (word‐based) data are used in the current analysis. Extracted data (_EXT) are from strictly performed, lab‐based, numeric analyses. Most data in this file are listed as reported by the source data report; only minor unit changes are performed or assumptions made about the thickness of the sediment analyzed based on the sampler type. Typical data themes include textural classes and statistics (TXR: gravel, sand, silt, clay, mud, and various statistics), phi grain‐ size classes (GRZ), chemical composition (CMP), acoustic measurements (ACU), color (COL), and geotechnical parameters (GTC). The _EXT file is based on rigorous lab‐ determined values and forms the most reliable data sets. Limitations, however, exist due to the uncertainty of the sample tested. For example, are the analyses performed on whole samples or only on the matrix, possibly with larger particles ignored? Parsed data (_PRS) are numeric data obtained from verbal logs from core descriptions, shipboard notes, and (or) photographic descriptions are held in the parsed data set. The input data are maintained using the terms employed by the original researchers and are coded using phonetically sensible terms for easier processing by dbSEABED.

Using the scaling terminology where grain is the elementary sampling unit area, interval is the distance between samples, and extent is the area covered by the surveys (Legendre and Legendre 1998), sampling temporal and spatial scales are variable dependent on study. Samples consist largely of cores and grabs so grain is typically less than 1 m2.

This database is a compilation and sampling scales represented are highly variable dependent on the specific study. Reid et al (2005) provide the following caveats for use of the usSEABED database. • As many reports are decades old, users of usSEABED should use their own criteria to determine the appropriateness of data from each source report for their particular purpose and scale of interest. • In cases where no original metadata are available, metadata are created based on existing available information accompanying the data. Of particular importance,

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 90 DRAFT April 17, 2009 site locations are as given in the original sources, with uncertainties due to navigational techniques and datums ignored in the usSEABED compilation. • As a caution in using the usSEABED database in depicting seabed sedimentary character or creating seafloor geologic maps, users should aware that all seafloor regions are by their nature dynamic environments and subject to a variety of physical processes such as erosion, winnowing, reworking, and sedimentation or accretion that vary on different spatial and temporal scales. In addition, as with any such database, usSEABED is comprised of samples collected and described and analyzed by many different organizations and individuals over a span of many years, providing inherent uncertainties between data points. • Plotting the data can also introduce uncertainties that are largely unknown at this time. • There are uncertainties in data quality associated with both the extracted data (numeric/ analytical analyses) and parsed data (word‐based descriptions). • On occasion grain‐size analyses are done solely on the sand fraction, excluding coarse fractions such as shell fragments and gravel, while word descriptions of sediment samples can emphasize or de‐emphasize the proportion of fine or coarse sediment fraction, or disregard other important textural or biological components. There are strengths and limitations to the dataset for mapping purposes.

Strengths: • As a compilation the usSEABED database covers the model domain. • The extracted data are based on physical examination of substrates. Limitations: • The sampling design, device, and analytical methods used are temporally and spatially variable. • Few individual studies used a formal experimental design. • Most sampling devices used are not capable of sampling cobbles and boulders. Many devices used have sampling selectivity characteristics which may over or under represent small or large particles.

6.1.3.3 NMFS Trawl Survey Gear Hangs Northeast Center trawl survey hangs are compiled as specified in Link and Demarest (2003). The 3,446 trawl hangs recorded from 1948 – 2007 are assumed to indicated the presence of boulder‐equivalent substrates.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 91 DRAFT April 17, 2009 Figure 9 – Base grid cell coding of dominant substrate

6.1.4 Classifying natural disturbance – Critical Shear Stress Model As water flow increases over the seabed, the shear stress velocity increases and the hydrodynamic forces acting on the bottom will eventually dislodge and start to move substrate particles. This threshold for movement is termed critical shear stress. The recovery of fish habitat features to fishing disturbance is expected to be related to the typical shear stress velocities in a particular area. To allow for the use of separate sensitivity parameters based on shear stress velocities, each cell in the base grid is classified as either high or low energy. Habitat feature recovery values are then determined separately for high and low energy. High or low energy is determined based on a combination of depth and FVCOM‐derived benthic boundary flow conditions (Table 33).

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 92 DRAFT April 17, 2009 Table 33 ‐ Critical shear stress model components Condition Data source Parameterization High energy Low energy Shear stress The max shear stress magnitude on High = sheer stress ≥ 0.194 N∙m‐ Low = sheer stress < the bottom in N∙m‐2 derived from the 2 (critical sheer stress sufficient 0.194 N∙m‐2 M2 and S2 tidal components only to initiate motion in coarse sand) Depth Coastal Relief Model depth data High = depths ≤ 20m Low = depths > 20m

Depth is used as a proxy for wave‐driven annual flow events. A depth of 20 m is selected as the boundary for high‐energy flow levels based on the average depth where annual storm‐event wave height conditions occur (G. Cowles personal communication).

Shear stress is calculated using the Gulf of Maine module of the Finite Volume Coastal Ocean Model (FVCOM‐GoM) (Chen et al., 2003, 2006, Cowles, 2008). The bottom stress in the model is calculated where the drag coefficient is depth‐based and critical shear stress is log10 (shear). Maximum shear stress magnitudes are derived from the M2 and S2 tidal components; these would thus represent the mean spring tides and would not include the effects of perigee/apogee.

FVCOM is an open source Fortran90 software package for the simulation of ocean processes in coastal regions (Chen et al. 2003, 2006, Cowles, 2008). The kernel of the code computes a solution of the hydrostatic primitive equations on unstructured grids using a finite‐volume flux formulation. The unstructured grid modeling approach is highly advantageous for resolving dynamics in regions with complex shorelines and bathymetry, such as estuaries, embayments, and archipelagos. The horizontal spatial fluxes are discretized using a second‐order accurate scheme. For the vertical discretization, a generalized terrain‐following coordinate is employed. FVCOM is interfaced to the General Ocean Turbulence Model (GOTM) libraries to provide an array of turbulence closure schemes including the standard Mellor‐Yamada 2.5 and k‐epsilon approaches. An explicit mode‐splitting approach is used to advance the governing equations in time. To take advantage of modern multiprocessor computers, FVCOM is parallelized using an efficient SPMD approach.

The FVCOM‐GoM domain includes the entire Gulf of Maine, the Scotian Shelf to 45.2° N, and the New England Shelf to the northern edge of the Mid‐Atlantic at 39.1° N. The model mesh contains 30K elements in the horizontal and 30 layers in the vertical. Resolution ranges from approximately 3km on Georges Bank to 15km near the open boundary. The model is advanced at a time step of 120s. A high performance computer cluster (32 processors) is used to run FVCOM‐GoM, requiring about 8 hours of wall clock time for each month of simulated time. Boundary forcing in the FVCOM‐GoM system includes prescription of the five major tidal constituents at the open boundary, freshwater input from major rivers in the Gulf of Maine, and wind stress and heat flux derived from a high resolution configuration of the MM5 meteorological model. At the

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 93 DRAFT April 17, 2009 open boundary, hydrography is set using monthly climatology fields derived from survey data using optimal interpolation techniques. Assimilation of daily mean satellite‐ derived sea surface temperature (SST) into the model SST is included to improve the model temperature state. The model has been validated using long‐term observations of tidal and subtidal currents and as well as hydrography (Cowles et al. 2008).

Figure 10 – FVCOM domain and nodes

The circulation in the Gulf of Maine, Georges Bank and the New England Shelf regions is simulated from 1995‐present. Hourly model hydrographic and velocity data fields are computed at each cell in the domain. Shear stress is computed from the model velocity fields using the “law of the wall” with a depth‐based estimate of bottom roughness (Bradshaw and Huang 1995).

High or low energy values were inferred from the shear stress surface to the SASI model grid based on spatial overlap. Where more than one shear stress estimate occurred per SASI model grid the mean of the values was used. Outside the FVCOM model domain energy values were assumed to be low. This is reasonable given regions outside the domain include the deep water GOM and the southern Mid‐Atlantic were tidal flows are slow or are diminished by depth.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 94 DRAFT April 17, 2009 Figure 11 – Base grid cell coding of energy (CSS model plus depth)

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 95 DRAFT April 17, 2009 7.0 The FiGSI model: combining the Vulnerability Assessment, the SASI model and the Spatial model

7.1 Methods The Fishing Gear Seabed Impact model calculates the relative impact of specified fishing gears on substrate‐dominated habitats by combining the susceptibility and recovery values estimated in the Vulnerability Assessment with the contact‐adjusted area swept values estimated in the SASI model. To do this, a sensitivity coefficient is estimated as a function of the susceptibility and recovery value, and this coefficient is applied to the each fishing gear’s contact‐adjusted area swept estimate. This coefficient is applied spatially based on the location (binned by grid cell) of the fishing effort, and the dominant substrate and energy environment (high/low) coded at that location.

7.1.1 Estimating sensitivity (Se) coefficients The susceptibility and recovery values for each feature are combined to produce a single sensitivity value. This sensitivity value, ranging from zero to one, is used to further modify the contact‐adjusted area swept calculated for each gear type from the SASI model. Because multiple features occur in each substrate subclass (with substrate subclasses constituting the base grid for the spatial model), the features are weighted according to their expected abundance to generate a single sensitivity value for the substrate subclass.

Susceptibility and recovery values resulting from the Vulnerability Assessment are combined into a single sensitivity value used to scale the area‐swept results for each gear type. Across rows or down columns in Table 34 (below) the percentage increase in sensitivity between two adjacent S or R values is constant.

Table 34 ‐ Sensitivity values RECOVERY Se=f(S,R) fast moderate slow very slow 0 1 2 3 none 0 0 0 0 0 low 1 0.13 0.20 0.30 0.44 SUSCEPTIBILITY medium 2 0.20 0.30 0.44 0.67

high 3 0.30 0.44 0.67 1.00

Sensitivity is set to zero if susceptibility was valued as zero, as recovery does not apply if there is no loss of functional value.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 96 DRAFT April 17, 2009 7.2 Results

7.2.1 Geological habitat components Weighted sensitivity values for the trawl and scallop dredge matrices are given below. Generally, gravel features were more sensitive than mud or sand features. Features were generally most sensitive to scallop dredge gear, followed by generic trawl gear, squid trawls, shrimp trawls, and raised footrope trawls.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 97 DRAFT April 17, 2009 Table 35 ‐ Weighted sensitivity values, groundfish trawl gear Substrate Feature Feature weight weight Dominant Subclass R high Se high high Weighted Se R low Se low low Weighted Se class Physical features S energy energy energy high energy energy energy energy low energy Mud Clay‐Silt featureless clay‐silt 2 1 0.3 1 0.3 1 0.3 1 0.3 Muddy Sand featureless muddy sand 2 0 0.2 1 0.2 1 0.3 1 0.3 Sand featureless sand 2 0 0.2 0 1 0.3 0.5 Sand 0.2 0.3 bedforms 2 0 0.2 1 1 0.3 0.5 granule‐pebble 2 1 0.3 0.9 2 0.44 0.9 Granule‐Pebble 0.3 0.44 pebble pavement 2 1 0.3 0.1 2 0.44 0.1 cobble 2 2 0.44 0.8 3 0.67 0.8 Cobble cobble piles 3 3 1 0.1 0.496 3 1 0.1 0.68 cobble pavement 2 2 0.44 0.1 2 0.44 0.1 boulders 2 3 0.67 0.9 3 0.67 0.9 Boulder 0.67 0.67 piled boulders 2 3 0.67 0.1 3 0.67 0.1 Shell shell deposits 2 1 0.3 1 0.3 2 0.44 1 0.44

Table 36 ‐ Weighted sensitivity values, raised footrope trawl gear Substrate Feature Feature weight weight Dominant Subclass R high Se high high Weighted Se R low Se low low Weighted Se class Physical features S energy energy energy high energy energy energy energy low energy Mud Clay‐Silt featureless clay‐silt 1 1 0.2 1 0.2 1 0.2 1 0.2 Muddy Sand featureless muddy sand 1 0 0.13 1 0.13 1 0.2 1 0.2 Sand featureless sand 1 0 0.13 0 1 0.2 0.5 Sand 0.2 0.25 bedforms 2 0 0.2 1 1 0.3 0.5 granule‐pebble 1 1 0.2 0.9 2 0.3 0.9 Granule‐Pebble 0.2 0.3 pebble pavement 1 1 0.2 0.1 2 0.3 0.1 cobble 1 2 0.3 0.8 2 0.3 0.8 Cobble cobble piles 2 3 0.67 0.1 0.337 3 0.67 0.1 0.337 cobble pavement 1 2 0.3 0.1 2 0.3 0.1 boulders 1 3 0.44 0.9 3 0.44 0.9 Boulder 0.463 0.463 piled boulders 2 3 0.67 0.1 3 0.67 0.1 Shell shell deposits 1 1 0.2 1 0.2 2 0.3 1 0.3

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 98 DRAFT April 17, 2009 Table 37 ‐ Weighted sensitivity values, shrimp trawl gear Substrate Feature Feature weight weight Dominant Subclass R high Se high high Weighted Se R low Se low low Weighted Se class Physical features S energy energy energy high energy energy energy energy low energy Mud Clay‐Silt featureless clay‐silt 1 1 0.2 1 0.2 1 0.2 1 0.2 Muddy Sand featureless muddy sand 1 0 0.13 1 0.13 1 0.2 1 0.2 Sand featureless sand 1 0 0.13 0 1 0.2 0.5 Sand 0.2 0.25 bedforms 2 0 0.2 1 1 0.3 0.5 granule‐pebble 1 1 0.2 0.9 2 0.3 0.9 Granule‐Pebble 0.2 0.3 pebble pavement 1 1 0.2 0.1 2 0.3 0.1 cobble 1 2 0.3 0.8 3 0.44 0.8 Cobble cobble piles 2 3 0.67 0.1 0.337 3 0.67 0.1 0.449 cobble pavement 1 2 0.3 0.1 2 0.3 0.1 boulders 1 3 0.44 0.9 3 0.44 0.9 Boulder 0.463 0.463 piled boulders 2 3 0.67 0.1 3 0.67 0.1 Shell shell deposits 2 1 0.3 1 0.3 2 0.44 1 0.44

Table 38 ‐ Weighted sensitivity values, squid trawl gear Substrate Feature Feature weight weight Dominant Subclass R high Se high high Weighted Se R low Se low low Weighted Se class Physical features S energy energy energy high energy energy energy energy low energy Mud Clay‐Silt featureless clay‐silt 2 1 0.3 1 0.3 1 0.3 1 0.3 Muddy Sand featureless muddy sand 2 0 0.2 1 0.2 1 0.3 1 0.3 Sand featureless sand 2 0 0.2 0 1 0.3 0.5 Sand 0.13 0.25 bedforms 1 0 0.13 1 1 0.2 0.5 granule‐pebble 1 1 0.2 0.9 2 0.3 0.9 Granule‐Pebble 0.2 0.3 pebble pavement 1 1 0.2 0.1 2 0.3 0.1 cobble 1 2 0.3 0.8 3 0.44 0.8 Cobble cobble piles 2 3 0.67 0.1 0.337 3 0.67 0.1 0.449 cobble pavement 1 2 0.3 0.1 2 0.3 0.1 boulders 1 3 0.44 0.9 3 0.44 0.9 Boulder 0.463 0.463 piled boulders 2 3 0.67 0.1 3 0.67 0.1 Shell shell deposits 2 1 0.3 1 0.3 2 0.44 1 0.44

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 99 DRAFT April 17, 2009 Table 39 ‐ Weighted sensitivity values, New Bedford‐style scallop dredge gear Substrate Feature Feature weight Weighted weight Dominant Subclass R high Se high high Se high R low Se low low Weighted Se class Physical features S energy energy energy energy energy energy energy low energy Mud Clay‐Silt featureless clay‐silt 2 1 0.3 1 0.3 1 0.3 1 0.3 Muddy Sand featureless muddy sand 2 0 0.2 1 0.2 1 0.3 1 0.3 Sand featureless sand 2 0 0.2 0 1 0.3 0.5 Sand 0.3 0.37 bedforms 3 0 0.3 1 1 0.44 0.5 granule‐pebble 2 1 0.3 0.9 2 0.44 0.9 Granule‐Pebble 0.3 0.44 pebble pavement 2 1 0.3 0.1 2 0.44 0.1 cobble 2 2 0.44 0.8 3 0.67 0.8 Cobble cobble piles 3 3 1 0.1 0.496 3 1 0.1 0.68 cobble pavement 2 2 0.44 0.1 2 0.44 0.1 boulders 2 3 0.67 0.9 3 0.67 0.9 Boulder 0.703 0.703 piled boulders 3 3 1 0.1 3 1 0.1 Shell shell deposits 2 1 0.3 1 0.3 2 0.44 1 0.44

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 100 DRAFT April 17, 2009 Figure 12 – Spatially defined sensitivity (Se) values for groundfish and raised footrope trawl gears.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 101 DRAFT April 17, 2009 Figure 13 – Spatially defined sensitivity (Se) values for shrimp and squid trawl gears.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 102 DRAFT April 17, 2009 Figure 14 ‐ Spatially‐defined sensitivity (Se) values for scallop dredge gear.

‐‐Additional results, including contact‐ and sensitivity‐adjusted area swept estimates for each of the five gear types to be completed‐‐

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 103 DRAFT April 17, 2009

7.2.2 Biological habitat components ‐‐To be completed‐‐

7.2.3 Prey habitat components ‐‐To be completed‐‐

7.3 Future directions The next steps include: • continue to refine the scallop dredge and trawl gear SASI models and refine how trips are classified • continue work on gear descriptions to ensure they comport with regulations • complete a final review of the literature, and develop and annotated bibliography of gear impacts studies not considered here (from the larger list of 400+ studies initially reviewed, and elsewhere) • reconsider contact indices during a joint Committee and Advisory Panel workshop • determine how to weight biological and prey features according to dominant substrate so they can be modeled spatially • fill out biological and prey matrices • determine how to proceed given the lack of data for fixed gear matrices • document results and prepare discussion sections • prepare alternative based on model results for Committee and Council consideration

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 104 DRAFT April 17, 2009 8.0 References Asch, R. and J. S. Collie (2007). ʺChanges in a benthic megafaunal community due to disturbance from bottom fishing and the establishment of a fishery closure.ʺ Fish. Bull. 106(4): 438‐456. (Ref#404) Auster, P. J., R. J. Malatesta, R. W. Langton, et al. (1996). ʺThe impacts of mobile fishing gear on seafloor habitats in the Gulf of Maine (northwest Atlantic): Implications for conservation of fish populations.ʺ Rev. Fish. Sci. 4(2): 185‐202. (Ref#11) Ball, B., B. Munday and I. Tuck (2000). Effects of otter trawling on the benthos and environment in muddy sediments. Effects of Fishing on Non‐target Species and Habitats: Biological, Conservation and Socio‐economic Issues. M. J. Kaiser and S. J. de Groot. Oxford (UK), Blackwell Science: 69‐82. (Ref#17) Bergman, M. J. N. and J. W. Van Santbrink (2000). Fishing mortality of populations of megafauna in sandy sediments. Effects of Fishing on Non‐target Species and Habitats: Biological, Conservation and Socio‐economic Issues. M. J. Kaiser and S. J. de Groot. Oxford (UK), Blackwell Science: 49‐68. (Ref#21) Blanchard, F., F. LeLocʹh, C. Hily, et al. (2004). ʺFishing effects on diversity, size and community structure of the benthic invertebrate and fish megafauna on the Bay of Biscay coast of France.ʺ Mar. Ecol. Prog. Ser. 280: 249‐260. (Ref#24) Boat Kathleen A. Mirarchi Inc. and CR Environmental Inc. (2003). Near term observations of the effectsof smooth bottom net trawl fishing gear on the seabed. NMFS Cooperative Research Partners Program Northeast Region. (Ref#408) Boat Kathleen A. Mirarchi Inc. and CR Environmental Inc. (2005). Smooth bottom net trawl fishing gear effect on the seabed: investigattion of temporal and cumulative effects. NOAA/NMFS Unallied Science Project, Cooperative Agreement NA16FL2264: 187p. (Ref#409) Bradshaw, P. and G. P. Huang (1995). ʺThe Law of the Wall in Turbulent Flow.ʺ Proc. Math. Phys. Sci. 451(1941): 165‐188. Brown, B. E., B. Finney, S. Hills, et al. (2005). Effects of commercial otter trawling on benthic communities in the southeastern Bering Sea. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 439‐460. (Ref#34) Brown, E. J., B. Finney, M. Dommisse, et al. (2005). ʺEffects of commercial otter trawling on the physical environment of the southeastern Bering Sea.ʺ Cont. Shelf Res. 25(10): 1281‐1301. (Ref#35) Caddy, J. F. (1968). ʺUnderwater observations on scallop (Placopecten magellanicus) behaviour and drag efficiency.ʺ J. Fish. Res. Board Can. 25(10): 2123‐2141. (Ref#42) Caddy, J. F. (1973). ʺUnderwater observations on tracks of dredges and trawls and some effects of dredging on a scallop ground.ʺ J. Fish. Res. Board Can. 30(2): 173‐180. (Ref#43) Carr, H. A. and H. Milliken (1998). Conservation engineering: options to minimize fishing’s impacts to the sea floor. Pp 100‐103 in: Effects of Fishing Gear on the Sea Floor of New England (E.L. Dorsey and J. Pederson eds). Boston, MA, Conservation Law Foundation.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 105 DRAFT April 17, 2009 Chen, C., R. C. Beardsley and G. Cowles (2006). ʺAn unstructured grid, finite‐volume coastal ocean model (FVCOM) system. Special Issue entitled “Advances in Computational Oceanography”.Oceanography 19(1): 78‐89. Chen, C., H. Liu and R. C. Beardsley (2003). ʺAn unstructured, finite‐volume, three‐dimensional, primitive equation ocean model: application to coastal ocean and estuaries.ʺ J. Atmos. Oceanic Technol. 20: 159‐186. Chen, C., Q. Xu, R. Houghton, et al. (2008). ʺA model‐dye comparison experiment in the tidal mixing front zone on the southern flank of Georges Bank.ʺ J. Geophys. Res. Oceans 113(c2): C02005. Coggan, R. A., C. J. Smith, R. J. A. Atkinson, et al. (2001). Comparison of rapid methodologies for quantifying environmental impacts of otter trawls. Study proposal in support of the Common Fisheries Policy. Call for proposals XIV ‐ C1 D (98): 254. (Ref#414) Collie, J. S., G. A. Escanero and P. C. Valentine (1997). ʺEffects of bottom fishing on the benthic megafauna of Georges Bank.ʺ Mar. Ecol. Prog. Ser. 155: 159‐172. (Ref#69) Collie, J. S., S. J. Hall, M. J. Kaiser, et al. (2000). ʺA Quantitative Analysis of Fishing Impacts on Shelf‐Sea Benthos.ʺ J. Anim. Ecol. 69(5): 785‐798. (Ref#70) Collie, J. S., J. M. Hermsen, P. C. Valentine, et al. (2005). Effects of Fishing on Gravel Habitats: Assessment and Recovery of Benthic Megafauna on Georges Bank. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 325‐343. (Ref#71) Cowles, G. W., S. J. Lentz, C. Chen, et al. (2008). ʺComparison of observed and model‐computed low frequency circulation and hydrography on the New England Shelf.ʺ J. Geophys. Res. Oceans 113: C09015. De Biasi, A. M. (2004). ʺImpact of experimental trawling on the benthic assemblage along the Tuscany coast (North Tyrrhenian Sea, Italy).ʺ ICES J. Mar. Sci. 61(8): 1260‐1266. (Ref#88) de Juan, S., J. Cartes and M. Demestre (2007). ʺEffects of commerical trawling activities in the diet of the flat fish Citharus linguatula and the starfish Astropecten irregularis.ʺ J. Exp. Mar. Bio. Ecol. 349: 152‐ 169. (Ref#90) de Juan, S., S. F. Thrush and M. Demestre (2007). ʺFunctional changes as indicators of trawling disturbance on a benthic community located in a fishing ground (NW Mediterranean Sea).ʺ Mar. Ecol. Prog. Ser. 334: 117‐129. (Ref#89) DeAlteris, J. (1998). Training Manual: Fisheries Science and Technology, prepared for the NOAA Corps OfficerProgram. Unpublished Manuscript. Kingston, RI: 34. DeAlteris, J., L. Skrobe, C. Lipsky, et al. (1999). ʺThe significance of seabed disturbance by mobile fishing gear relative to natural processes: a case study in Narragansett Bay, Rhode Island.ʺ American Fisheries Society Symposium 22: 224‐237. (Ref#92) Dellapenna, T. M., M. A. Allison, G. A. Gill, et al. (2006). ʺThe impact of shrimp trawling and associated sediment resuspension in mud dominated, shallow estuaries.ʺ Estuarine Coastal Shelf Sci. 69(3‐4): 519‐530. (Ref#406) Drabsch, S. L., J. E. Tanner and S. D. Connell (2001). ʺLimited infaunal response to experimental trawling in previously untrawled areas.ʺ ICES J. Mar. Sci. 58(6): 1261‐1271. (Ref#97)

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 106 DRAFT April 17, 2009 Engel, J. and R. Kvitek (1998). ʺEffects of Otter Trawling on a Benthic Community in Monterey Bay National Marine Sanctuary.ʺ Conserv. Biol. 12(6): 1204‐1214. (Ref#101) Eno, N. C., D. S. MacDonald, J. A. M. Kinnear, et al. (2001). ʺEffects of crustacean traps on benthic fauna.ʺ ICES J. Mar. Sci. 58(1): 11‐20. (Ref#102) Everhart, W. H. and W. D. Youngs (1981). Principles of Fishery Science. Ithaca, NY, Cornell University Press. Fortin, M.‐J. and M. R. T. Dale (2005). Spatial Analysis: A Guide for Ecologists. Cambridge, U.K., Cambridge University Press. Freese, J. L. (2001). ʺTrawl‐induced Damage to Sponges Observed From a Research Submersible.ʺ Mar. Fish. Rev. 63(3): 7‐13. (Ref#110) Freese, L., P. J. Auster, J. Heifetz, et al. (1999). ʺEffects of trawling on seafloor habitat and associated invertebrate taxa in the Gulf of Alaska.ʺ Mar. Ecol. Prog. Ser.: Vol. 182, p. (Ref#111) Frid, C. L. J., R. A. Clark and J. A. Hall (1999). ʺLong‐term changes in the benthos on a heavily fished ground off the NE coast of England.ʺ Mar. Ecol. Prog. Ser. 188: 13‐20. (Ref#113) Gibbs, P. J., A. J. Collins and L. C. Collett (1980). ʺEffect of otter prawn trawling on the macrobenthos of a sandy substratum in a New South Wales Estuary.ʺ Australian Journal of Marine and Freshwater Research 31(4): 509‐516. (Ref#119) Gilkinson, K., M. Paulin, S. Hurley, et al. (1998). ʺImpacts of trawl door scouring on infaunal bivalves: results of a physical trawl door model/dense sand interaction.ʺ J. Exp. Mar. Biol. Ecol. 224(2): 291‐ 312. (Ref#120) Gilkinson, K. D., G. B. J. Fader, D. C. Gordon Jr, et al. (2003). ʺImmediate and longer‐term impacts of hydraulic clam dredging on an offshore sandy seabed: effects on physical habitat and processes of recovery.ʺ Cont. Shelf Res. 23(14‐15): 1315‐1336. (Ref#121) Gilkinson, K. D., D. C. Gordon, K. G. MacIsaac, et al. (2005). ʺImmediate impacts and recovery trajectories of macrofaunal communities following hydraulic clam dredging on Banquereau, eastern Canada.ʺ ICES J. Mar. Sci. 62(5): 925‐947. (Ref#122) Gilkinson, K. D., D. C. Gordon, Jr., D. McKeown, et al. (2005). Susceptibility of the Soft Coral Gersemia rubiformis to Capture by Hydraulic Clam Dredges off Eastern Canada: The Significance of Soft Coral‐Shell Associations. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 383‐ 390. (Ref#123) Gold, C. M. (1991). ʺProblems with handling spatial data ‐ the Voronoi approach.ʺ Canadian Institute of Surveying and Mapping Journal 45: 65‐80. Gordon, D. C., Jr., K. D. Gilkinson, E. L. Kenchington, et al. (2005). Summary of the Grand Banks Otter Trawling Experiment (1993‐1995): Effects on Benthic Habitat and Macrobenthic Communities. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 411‐424. (Ref#128) Hall, S. J., D. J. Basford and M. R. Robertson (1990). ʺThe impact of hydraulic dredging for razor clams sp. on an infaunal community.ʺ Neth. J. Sea Res. 27(1): 119‐125. (Ref#140)

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 107 DRAFT April 17, 2009 Hall, S. J., M. R. Robertson, D. J. Basford, et al. (1993). ʺThe possible effects of fishing disturbance in the northern North Sea: An analysis of spatial patterns in community structure around a wreck.ʺ Neth. J. Sea Res. 31(2): 201‐208. (Ref#141) Hansson, M., M. Lindegarth, D. Valentinsson, et al. (2000). ʺEffects of shrimp‐trawling on abundance of benthic macrofauna in Gullmarsfjorden, Sweden.ʺ Mar. Ecol. Prog. Ser. 198: 191‐201. (Ref#149) Henry, L.‐A., E. L. R. Kenchington, T. J. Kenchington, et al. (2006). ʺImpacts of otter trawling on colonial epifaunal assemblages on a cobble bottom ecosystem on Western Bank (northwest Atlantic).ʺ Mar. Ecol. Prog. Ser. 306: 63‐78. (Ref#157) Hermsen, J. M., J. S. Collie and P. C. Valentine (2003). ʺMobile fishing gear reduces benthic megafaunal production on Georges Bank.ʺ Mar. Ecol. Prog. Ser. 260: 97‐108. (Ref#158) Hixon, M. A. and B. N. Tissot (2007). ʺComparison of trawled vs untrawled mud seafloor assemblages of fishes and macroinvertebrates at Coquille Bank, Oregon.ʺ J. Exp. Mar. Biol. Ecol. 344(1): 23‐34. (Ref#164) Isaaks, E. H. and R. M. Srivastava (1989). An Introduction to Applied Geostatistics. New York, NY, Oxford University Press. Kaiser, M. J., F. E. Spence and P. J. Hart (2000). ʺFishing‐Gear Restrictions and Conservation of Benthic Habitat Complexity.ʺ Conserv. Biol. 14(5): 1512‐1525. (Ref#184) Kenchington, E. L., D. C. Gordon, Jr., C. Bourbonnais‐Boyce, et al. (2005). Effects of Experimental Otter Trawling on the Feeding of Demersal Fish on Western Bank, Nova Scotia. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 391‐409. (Ref#193) Kenchington, E. L. R., K. D. Gilkinson, K. G. MacIsaac, et al. (2006). ʺEffects of experimental otter trawling on benthic assemblages on Western Bank, northwest Atlantic Ocean.ʺ J. Sea Res. 56(3): 249‐270. (Ref#194) Kenchington, E. L. R., J. Prena, K. D. Gilkinson, et al. (2001). ʺEffects of experimental otter trawling on the macrofauna of a sandy bottom ecosystem on the Grand Banks of Newfoundland.ʺ Can. J. Fish. Aquat. Sci. 58(6): 1043‐1057. (Ref#192) Knight, E. P. (2005). The effects of trawling on benthic habitats: an analysis of recovery in the western Gulf of Maine closure, University of Maine. Master of Science (in Oceanography): 82p. (Ref#203) Koulouri, P. T., C. G. Dounas and A. Eleftheriou (2005). Preliminary Results on the Effect of Otter Trawling on Hyperbenthic Communities in Heraklion Bay, Cretan Sea, Eastern Mediterranean. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 529‐538. (Ref#211) Kutti, T., T. Hoeisaeter, H. T. Rapp, et al. (2005). Immediate Effects of Experimental Otter Trawling on a Sub‐Arctic Benthic Assemblage inside Bear Island Fishery Protection Zone in the Barents Sea. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 519‐528. (Ref#214) Langton, R. W. and W. E. Robinson (1990). ʺFaunal associations on scallop grounds in the western Gulf of Maine.ʺ J. Exp. Mar. Biol. Ecol. 144(2‐3): 157‐171. (Ref#217) Legendre, P. and L. Legendre (1998). Numerical ecology. 2nd English edition. Amsterdam, Elsevier Science, BV.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 108 DRAFT April 17, 2009 Lindholm, J., P. Auster and P. Valentine (2004). ʺRole of a large marine protected area for conserving landscape attributes of sand habitats on Georges Bank (NW Atlantic).ʺ Mar. Ecol. Prog. Ser. 269: 61‐68. (Ref#225) Link, J., F. Almeida, P. Valentine, et al. (2005). The Effects of Area Closures on Georges Bank. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 345‐368. (Ref#228) Link, J. S. and C. Demarest (2003). ʺTrawl hangs, baby fish, and closed areas: a win‐win scenario.ʺ ICES J. Mar. Sci. 60(5): 930‐938. MacKenzie, C. L., Jr. (1982). ʺCompatibility of invertebrate populations and commercial fishing for ocean quahogs.ʺ N. Amer. J. Fish. Mgmt. 2(3): 270‐275. (Ref#232) Mayer, L. M., D. F. Schick, R. H. Findlay, et al. (1991). ʺEffects of commercial dragging on sedimentary organic matter.ʺ Mar. Env. Res. 31(4): 249‐261. (Ref#236) McConnaughey, R. A., K. L. Mier and C. B. Dew (2000). ʺAn examination of chronic trawling effects on soft‐bottom benthos of the eastern Bering Sea.ʺ ICES J. Mar. Sci. 57(5): 1377‐1388. (Ref#238) McConnaughey, R. A., S. E. Syrjala and C. B. Dew (2005). Effects of Chronic Bottom Trawling on the Size Structure of Soft‐Bottom Benthic Invertebrates. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 425‐437. (Ref#239) Medcof, J. C. and J. F. Caddy (1971). Underwater observations of the performance of clam dredges of three types. ICES Contributed Manuscript 1971/B:10: 7. (Ref#244) Meyer, T. L., R. A. Cooper and K. J. Pecci (1981). ʺThe Performance and Environmental Effects of a Hydraulic Clam Dredge.ʺ Mar. Fish. Rev. 43(9): 14‐22. (Ref#245) Mirarchi, F. (1998). Bottom trawling on soft substrates. Pp. 80‐84 in: Effects of Fishing Gear on the Sea Floor of New England, E.L. Dorsey and J. Pederson (eds.). Boston, MA, Conservation Law Foundation. Morais, P., T. C. Borges, V. Carnall, et al. (2007). ʺTrawl‐induced bottom disturbances off the south coast of Portugal: direct observations by the ʹDeltaʹ manned‐submersible on the Submarine Canyon of Portimao.ʺ Mar. Ecol. 28(s1): 112‐122. (Ref#247) Moran, M. J. and P. C. Stephenson (2000). ʺEffects of otter trawling on macrobenthos and management of demersal scalefish fisheries on the continental shelf of north‐western Australia.ʺ ICES J. Mar. Sci. 57(3): 510‐516. (Ref#248) Morello, E. B., C. Froglia, R. J. A. Atkinson, et al. (2005). ʺImpacts of hydraulic dredging on a macrobenthic community of the Adriatic Sea, Italy.ʺ Can. J. Fish. Aquat. Sci. 62(9): 2076‐2087. (Ref#249) Murawski, S. A. and F. M. Serchuk (1989). Environmental effects of offshore dredge fisheries for bivalves. ICES Contributed Manuscript 1989/K:27: 12. (Ref#256) New England Fishery Management Council (2002). Fishery Management Plan for Deep‐Sea Red Crab (Chaceon quinquedens), Including an Environmental Impact Statement, an Initial Regulatory Flexibility Act Analysis, and a Regulatory Impact Review. Newburyport, MA: 446. Nilsson, H. C. and R. Rosenberg (2003). ʺEffects on marine sedimentary habitats of experimental trawling analysed by sediment profile imagery.ʺ J. Exp. Mar. Biol. Ecol. 285‐286: 453‐463. (Ref#407)

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 109 DRAFT April 17, 2009 Northeast Region Essential Fish Habitat Steering Committee (2002). Workshop on the Effects of Fishing Gear on Marine Habitats Off the Northeastern United States, October 23‐25, 2001, Boston, Massachusetts. NEFSC Reference Document 02‐01: 86. Okabe, A., B. Boots, K. Sugihara, et al. (1992). Spatial Tessellations: Concepts and Applications of Voronoi Diagrams (Wiley Series in Probability and Statistics) West Sussex, England, John Wiley & Sons. Palanques, A., J. Guillen and P. Puig (2001). ʺImpact of Bottom Trawling on Water Turbidity and Muddy Sediment of an Unfished Continental Shelf.ʺ Limnology and Oceanography 46(5): 1100‐1110. (Ref#277) Pilskaln, C. H., J. H. Churchill and L. M. Mayer (1998). ʺResuspension of Sediment by Bottom Trawling in the Gulf of Maine and Potential Geochemical Consequences.ʺ Conserv. Biol. 12(6): 1223‐1229. (Ref#283) Pranovi, F. and O. Giovanardi (1994). ʺThe impact of hydraulic dredging for short‐necked clams, Tapes spp., on an infaunal community in the lagoon of Venice.ʺ Scientia Marina 58(4): 345‐353. (Ref#287) Prena, J., P. Schwinghamer, T. W. Rowell, et al. (1999). ʺExperimental otter trawling on a sandy bottom ecosystem of the Grand Banks of Newfoundland: Analysis of trawl bycatch and effects on epifauna.ʺ Mar. Ecol. Prog. Ser. 181: 107‐124. (Ref#291) Reid, J. M., J. A. Reid, C. J. Jenkins, et al. (2005). usSEABED: Atlantic Coast offshore surficial sediment data release, U.S. Geological Survey Data Series 118, Version 1.0. Online at http://pubs.usgs.gov/ds/2005/118/. Rieman, B. and E. Hoffmann (1991). ʺEcological consequences of dredging and bottom trawling in the Limfjord, Denmark.ʺ Mar. Ecol. Prog. Ser. 69(1‐2): 171‐178. (Ref#304) Rosenberg, R., H. C. Nilsson, A. Gramare, et al. (2003). ʺEffects of demersal trawling on marine sedimentary habitats analysed by sediment profile imagery.ʺ J. Exp. Mar. Biol. Ecol. 285‐286: 465‐ 477. (Ref#313) Sainsbury, J. C. (1996). Commercial Fishing Methods: An Introduction to Vessels and Gears. Oxford, U.K., Fishing News (Books) Ltd. Sanchez, P., M. Demestre, M. Ramon, et al. (2000). ʺThe impact of otter trawling on mud communities in the northwestern Mediterranean.ʺ ICES J. Mar. Sci. 57(5): 1352‐1358. (Ref#320) Schwinghamer, P., D. C. Gordon, Jr., T. W. Rowell, et al. (1998). ʺEffects of Experimental Otter Trawling on Surficial Sediment Properties of a Sandy‐Bottom Ecosystem on the Grand Banks of Newfoundland.ʺ Conserv. Biol. 12(6): 1215‐1222. (Ref#325) Sheridan, P. and J. Doerr (2005). Short‐Term Effects of the Cessation of Shrimp Trawling on Texas Benthic Habitats. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 571‐578. (Ref#330) Simpson, A. W. and L. Watling (2006). ʺAn investigation of the cumulative impacts of shrimp trawling on mud‐bottom fishing grounds in the Gulf of Maine: effects on habitat and macrofaunal community structure.ʺ ICES J. Mar. Sci. 63(9): 1616‐1630. (Ref#333) Smith, C. J., K. N. Papadopoulou and S. Diliberto (2000). ʺImpact of otter trawling on an eastern Mediterranean commercial trawl fishing ground.ʺ ICES J. Mar. Sci. 57(5): 1340‐1351. (Ref#335)

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 110 DRAFT April 17, 2009 Smith, C. J., H. Rumohr, I. Karakassis, et al. (2003). ʺAnalysing the impact of bottom trawls on sedimentary seabeds with sediment profile imagery.ʺ J. Exp. Mar. Biol. Ecol. 285‐286: 479‐496. (Ref#336) Smith, E., M. A. Alexander, M. M. Blake, et al. (1985). A study of lobster fisheries in Connecticut water of Long Island Sound with special reference to the effects of trawling on lobsters. Unpublished report, Connecticut Department of Environmental Protection Marine Fisheries Program. Hartford, CT. (Ref#334) Smolowitz, R. (1998). Bottom Tending Gear Used in New England. Pp 46‐52 in: Effects of Fishing Gear on the Sea Floor of New England, E.L. Dorsey and J. Pederson (eds.). Boston, MA, Conservation Law Foundation. Sparks‐McConkey, P. J. and L. Watling (2001). ʺEffects on the ecological integrity of a soft‐bottom habitat from a trawling disturbance.ʺ Hydrobiologia 456(1‐3): 73‐85. (Ref#338) Stevenson, D., L. Chiarella, D. Stephan, et al. (2004). Characterization of fishing practices and marine benthic ecosystems of the northeast US shelf, and and evaluation of potential effects of fishing on Essential Fish Habitat. National Marine Fisheries Service NOAA Technical Memorandum. Gloucester, MA. NMFS‐NEFSC 181: 179. (Ref#348) Stokesbury, K. D. E. (2002). ʺEstimation of Sea Scallop Abundance in Closed Areas of Georges Bank, USA.ʺ Transactions of the American Fisheries Society 131(6): 1081‐1092. Stokesbury, K. D. E. and B. P. Harris (2006). ʺImpact of limited short‐term sea scallop fishery on epibenthic community of Georges Bank closed areas.ʺ Mar. Ecol. Prog. Ser. 307: 85‐100. (Ref#352) Stokesbury, K. D. E. and B. P. Harris (2006). ʺImpact of limited short‐term sea scallop fishery on epibenthic community of Georges Bank closed areas.ʺ Mar. Ecol. Prog. Ser. 307: 85‐100. Stokesbury, K. D. E., B. P. Harris, M. C. Marino, II, et al. (2004). ʺEstimation of sea scallop abundance using a video survey in off‐shore US waters.ʺ J. Shellfish Res. 23(1): 33‐40. Stokesbury, K. D. E., B. P. Harris, M. C. Marino, et al. (2007). ʺSea scallop mass mortality in a Marine Protected Area.ʺ Mar. Ecol. Prog. Ser. 349: 151‐158. Stone, R. P., M. M. Masuda and P. W. Malecha (2005). Effects of Bottom Trawling on Soft‐Sediment Epibenthic Communities in the Gulf of Alaska. Benthic Habitats and the Effects of Fishing: American Fisheries Society Symposium 41. P. W. Barnes and J. P. Thomas. Bethesda, MD, American Fisheries Society: 461‐475. (Ref#355) Sullivan, M. C., R. K. Cowen, K. W. Able, et al. (2003). ʺEffects of anthropogenic and natural disturbance on a recently settled continental shelf flatfish.ʺ Mar. Ecol. Prog. Ser. 260: 237‐253. (Ref#359) Tanner, J. E. (2003). ʺThe influence of prawn trawling on sessile benthic assemblages in Gulf St. Vincent, South Australia.ʺ Can. J. Fish. Aquat. Sci. 60(5): 517‐526. (Ref#360) Thiessen, A. H. and J. C. Alter (1911). Climatological Data for July, 1911: District No. 10, Great Basin. Monthly Weather Review July 1911:1082‐1089. Tuck, I. D., N. Bailey, M. Harding, et al. (2000). ʺThe impact of water jet dredging for razor clams, Ensis spp., in a shallow sandy subtidal environment.ʺ J. Sea Res. 43(1): 65‐81. (Ref#373) Tuck, I. D., S. J. Hall, M. R. Robertson, et al. (1998). ʺEffects of physical trawling disturbance in a previously unfished sheltered Scottish sea loch.ʺ Mar. Ecol. Prog. Ser. 162: 227‐242. (Ref#372)

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 111 DRAFT April 17, 2009 Van Dolah, R. F., P. H. Wendt and N. Nicholson (1987). ʺEffects of a research trawl on a hard‐bottom assemblage of sponges and corals.ʺ Fish. Res. 5(1): 39‐54. (Ref#382) Wassenberg, T. J., G. Dews and S. D. Cook (2002). ʺThe impact of fish trawls on megabenthos (sponges) on the north‐west shelf of Australia.ʺ Fish. Res. 58(2): 141‐151. (Ref#387) Watling, L., R. H. Findlay, L. M. Mayer, et al. (2001). ʺImpact of a scallop drag on the sediment chemistry, microbiota, and faunal assemblages of a shallow subtidal marine benthic community.ʺ J. Sea Res. 46(3‐4): 309‐324. (Ref#391) Wentworth, C. K. (1922). ʺA grade scale and class terms for clastic sediments.ʺ J. Geol. 30: 377‐392. Williamson, J. (1998). Gillnet Fishing. Pp 87‐89 in: Effects of Fishing Gear on the Sea Floor of New England, E.L. Dorsey and J. Pederson (eds.). Boston, MA, Conservation Law Foundation.

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 112 DRAFT April 17, 2009 9.0 Appendices

9.1.1 Definitions Adverse effect ‐ any impact that reduces quality and/or quantity of essential fish habitat

Bedform ‐ shapes in the sediment on the sea bed produced by tides and currents

Before‐after control‐impact – a type of experimental design where a control site and experimental site are investigated both before and after a disturbance (i.e. fishing) in order to determine the effects of the disturbance, typically for comparison with natural disturbance

Benthic – associated with the seabed

Biogenic – originated by living organisms

Biological habitat component – an organism that provides shelter for fish

Contact‐adjusted area swept – an estimate, in m2, of the area of the seabed contacted by a particular fishing gear, adjusted for the amount of physical contact between the various gear components and the seabed

Effect – the physical interaction between a particular fishing gear and a habitat feature; e.g. slicing, compression, crushing, smoothing

Epifauna – any organism living on or attached to the seabed

Essential fish habitat (EFH) – those waters and substrate necessary to fish for spawning, breeding, feeding, or growth to maturity

Experimental – in terms of study design, any study allowing for the testing of hypotheses and including the use of control or reference sites

Observational – in terms of study design, any study where the effects of fishing were investigated without the use of control or reference sites; might include multiple evaluations over time

Comparative – in terms of study design, a study examining two or more sites subject to different disturbance regimes

Geological habitat component – that non‐living seafloor features that constitute a portion of a fish’s habitat

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 113 DRAFT April 17, 2009 Gravel – a substrate class including granule/pebble, cobble, boulder

Habitat components – a portion of essential fish habitat, including geological, biological, and prey

High energy – any portion of the seabed with a depth of less than or equal to 20 m or that experiences average sheer stress thresholds equal to or greater than 0.194 N∙m‐2

Infauna – organisms living with the sediment

K‐selected – organisms that are slow growing, long‐lived, and have low fecundity

Low energy ‐ any portion of the seabed with a depth greater than 20 m or that experiences average sheer stress thresholds less than 0.194 N∙m‐2

Macrofauna – infauna or epifauna visible to the naked eye

Mud – substrate class that includes silt‐clay and muddy‐sand

Natural disturbance – any tidally or storm induced forces reaching the seabed

Northeast region – the Gulf of Maine, Georges Bank, and Mid‐Atlantic Bight, including the continental shelf and the continental slope out to 200 nm offshore

Prey habitat component – taxonomic groups of organisms that are a food source for managed species

Recovery – an estimate of the time required for the functional value of a habitat feature to be restored

R‐selected – an organism that is short‐lived, fast growing, and highly fecund

Sand – a sediment subclass with particle sizes ranging from 0.0625 – 2 mm

Sediment profile image – a photograph of the upper layer of seabed used to investigate physical, chemical, or biological process

Sensitivity – a 0‐1 scaler applied to adjust area swept, derived by combining a feature’s susceptibility and recovery values

Sensitivity‐adjusted area swept ‐ an estimate, in m2, of the area of the seabed contacted by a particular fishing gear, adjusted for the amount of physical contact between the various gear components and the seabed and for the sensitivity of that particular portion of seabed

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 114 DRAFT April 17, 2009

Susceptibility – an estimate of the proportion of functional value of a habitat feature lost due to a single theoretical passage of the gear over that portion of seabed in the path of the tow

Swept Area Seabed Impact (SASI) – a model that estimates the contact‐adjusted area of seabed touched by various types of gear as they are fished. The model requires gear component widths, gear component contact indices ranging from 0‐1, and linear area swept

Voronoi diagram (Voronoi tessellation) – a special kind of decomposition of a metric space determined by distances to a specified discrete set of objects in the space (in this case, X,Y substrate data). In the simplest case, a set of objects, S, is arranged in space. Each object, S, has a Voronoi cell surrounding it. This cell consists of all points closer to that object S than to any other S. The line segments constituting the edges of the Voronoi cells are all the points in the plane that are equidistant to the two objects.

Vulnerability – see sensitivity

Wentworth particle grade scale – a size based classification for sediments, ranging from clay (<0.0039 mm) to boulder (>256 mm). The traditional clay and silt categories were combined for this analysis, as were gravel and pebble

Omnibus Essential Fish Habitat Amendment Phase II – Fishing Gear Seabed Impact Model 115 DRAFT April 17, 2009