Northeast Fisheries Science Center Reference Document 03-15
Estimates
of the number of vessels and quantity
of gear deployed in the lobster
and gillnet fisheries
in 1999 off the northeast coast of the United
States
by Kathryn Bisack
National Marine Fisheries Serv., Woods Hole Lab.,
166 Water St., Woods Hole, MA 02543
Print
publication date September 2003;
web version posted October 6, 2003
Citation: Bisack, K.D. 2003. Estimates of the number of vessels and quantity of gear deployed in the lobster and
gillnet fisheries in 1999 off the northeast coast of the United States. Northeast Fish. Sci. Cent. Ref. Doc.
03-15; 26 p.
Download complete PDF/print version
ABSTRACT
This paper presents an estimate of the number of
vessels and the quantity of gear deployed in the lobster and gillnet
fisheries in 1999 off the northeast coast of the United States. The
analyses presented here were developed to support the Environmental
Assessment (EA) of the Atlantic Large Whale Take Reduction Plan (NMFS,
2000). Vessels actively fishing lobster gear were identified as well
as vessels which held a fishing permit but had no recorded fishing
activity in any database. In contrast, only vessels actively fishing
gillnet gear were estimated. In both fisheries the total quantity
of gear fished was estimated as the product of the number of vessels
and the amount of gear fished per vessel. Vessel and gear estimates
were stratified by management areas defined within the American Lobster
Fishery Management Plan and the Atlantic Large Whale Take Reduction
Plan. In 1999, there were 7,539 vessels potentially fishing lobster
traps of which 79% fished in the northern inshore area (or state
waters), 9% in northern nearshore, and 2% in the northern offshore
area. The quantity of lobster gear deployed was expressed as the
number of trawls, where a lobster trawl consists of several traps
tied together with one line. A lower bound estimate was developed
from recorded fishing trips, and an upper bound estimate was derived
based on the maximum amount of gear a lobster vessel could legally
fish. The lower and upper bound estimates of total lobster trawls
fished in 1999 are 436,620 and 1,436,979, respectively. In 1999,
546 vessels actively fished gillnet gear. The quantity of gillnet
gear deployed was expressed as the number of strings, where a string
consists of several 300 foot nets tied together. Approximately 38%
of the vessels operated in the northern nearshore area, 17% offshore,
and 42% in the southern nearshore area. These vessels fished an estimated
3,254 strings of gillnet gear in 1999.
1.0
Introduction
Estimates were developed to establish a single source of information
on the number of vessels and the quantity of gear in the lobster and
gillnet fisheries off the northeast coast of the United States in 1999.
The paper provides a methodological template for updating these estimates
in the future.
The estimates were developed to support the Environmental Assessment
(EA) of the Atlantic Large Whale Take Reduction Plan (NMFS, 2000) and
to evaluate the economic impacts of gear modifications in the lobster
and gillnet fisheries aimed at reducing right whale entanglements.
Gillnets and lobster traps are the primary source of vertical lines
that may entangle right whales and other marine mammals.
The lobster and gillnet fisheries off the northeast coast of the United
operate from Maine to Cape Hatteras. Traps are the principal fishing
gear used to catch lobsters. A lobster trawl consists of one to forty
lobster traps strung together, where the line between traps averages
120 feet in the inshore management area and 180 feet in the offshore
management area (NMFS, pers. comm.). Lobster traps are set on the bottom
of the ocean to soak for one or more days, hauled back and reset in
the water.
Gillnet gear as deployed in the USA Atlantic region, consists of nets
(typically monofilament mesh) suspended between a buoyed head rope
and a weighted ground line. A gillnet vessel fishes four to seven strings
per trip on average. A single gillnet string normally consists of five
to twelve nets strung together, in which the standard net is about
three hundred feet long and eleven feet high. Gear is set in the water
to soak for 24 to 72 hours, after which the gear is hauled and reset.
Target species include groundfish such as pollock, cod, flatfish, monkfish
and dogfish. Both gillnet gear and lobster trawls are retrieved by
hauling in one of the two buoy lines attached to the end of the lobster
trawl or gillnet string.
This paper estimates the number of vessels that fished lobster gear
in 1999 as well as those vessels who held a permit to fish lobster
gear but for which no fishing activity was recorded in any database.
In contrast, only vessels which actively fished gillnet gear in 1999
was estimated. In both fisheries, the total quantity of gear fished
was calculated as the product of the number of vessels and the amount
of gear fished per vessel. A lower bound estimate was developed from
recorded fishing trips, and an upper bound estimate was derived based
on the maximum amount of gear a lobster vessel could legally fish.
In the gillnet fishery, a point estimate is presented for the number
of vessels and quantity of gear. Vessel and gear estimates are stratified
by management areas defined within the American Lobster Fishery Management
Plan and the Atlantic Large Whale Take Reduction Plan. Data were provided
by the National Marine Fisheries Service and the Department of Marine
Resources from the states of Maine, New Hampshire, Massachusetts and
Rhode Island. Several databases are used because none are comprehensive
and each provides a different level of information.
2.0
Current Regulations
Fishery regulations have been imposed to protect the American lobster
and commercially landed groundfish such as cod. Under the Marine Mammal
Protection Act (MMPA) of 1972 and the Endangered Species Act (ESA),
regulations have been implemented to protect marine mammals. Typical
and common regulations used in fisheries include output controls (Total
Allowable Catch, Individual Quota, Vessel Catch Limits), input controls
(limited licenses, individual effort quota such as days at sea, gear
and vessel restrictions), and technical measures (size and sex selectivity,
time and area closures). Current regulations are presented and used
in the process of estimating the number of vessels and the amount of
gear fished.
Right Whale
Seasonal area closures have been implemented to protect right whales.
In the 1997 and 2000 Atlantic Large Whale Take Reduction Team (ALWTRT)
plan (NMFS,1997), there are two areas closed to fishing. The Cape Cod
Bay Critical Habitat (CCBCH) area is closed to vessels fishing lobster
traps and gillnet gear from January 1 to May 15 (when the presence
of right whales is highest). The Great South Channel Critical Habitat
(GSCCH) is also closed to both gear types from April 1 to June 30.
Areas have been redefined between ALWTRT plans. Specifically, the
1997 right whale management areas were modified to comply with the
American lobster fishery management plan (FMP) areas (NMFS, 1999) in
2000.
Lobster Fleet
New measures were implemented to the American lobster FMP on December
6, 1999, with the objective of ending over-fishing (NMFS, 1999). These
new measures include: an extension of the current moratorium of new
entrants into the EEZ fishery; designation of lobster management areas;
nearshore and offshore area trap limits; a 5-inch maximum carapace
size in the Gulf of Maine; trap size restrictions; a trap escape vent
size increase; trap tag requirements; and an annual specification of
additional management measures necessary to end over-fishing and rebuild
the American lobster.
In 2000, federal and state permit holders in ME,
NH, MA and RI were required under the American lobster FMP (NARA, 2000)
to purchase one tag for each lobster trap. Tags were sold by one authorized
vendor within each state. In addition, trap limits were imposed. The
maximum number of lobster traps that can be fished depends on the management
area. Vessels which fish exclusively in offshore waters (Management
Area 3) can fish up to 1,800 lobster traps (Figure
1). Vessels fishing in other areas (or in management area combinations)
can fish a maximum of 800 lobster traps.
Gillnet Fleet
Vessels fishing gillnet gear are affected by the New England multi-species
FMP, Harbor Porpoise Take Reduction Team Plan (HBPTRP), Dogfish FMP,
Cod FMP and the ALWTRT plan. Gillnet vessels are subject to days-at-sea
limits, mesh size restrictions, seasonal and year-round closures, minimum
fish sizes and trip limits. Under the New England multi-species FMP,
vessel trip report (VTR) logbooks are mandatory for the gillnet fleet.
3.0
DATA
3.1 Data
Data from the following databases were used to estimate the number
of vessels and the quantity of gear fished: 1) the Northeast Dealer
Report Weighout (WO) data base; 2) the Northeast Vessel Trip (VTR)
data base; 3) National Marine Fisheries Service Permit Data (NMFSPD);
4) State Tag data - lobster trap tag data provided by individual states
(Maine, New Hampshire, Massachusetts, Rhode Island) and the NMFS Northeast
Regional Office (Gloucester, MA) (1) and;
5) the Northeast Domestic Fisheries Observer Program (NDFOP).
Commercial landings information by species are collected in the NEFSC
WO and VTR data bases. In the VTR, 'kept' pounds are recorded and 'landed'
pounds are recorded in the WO. It is assumed that the NEFSC WO data
contains more complete records of total landings, and that the VTR
data is a subset. In addition to species landed (in live pounds), records
in the WO identify the vessel's permit number with the date landed,
information on the market category of each species, the type of gear
fished, the port landed and the value of the catch, as well as other
information. A shortcoming of the WO data is that it does not contain
information on the location of fishing trips.
The NEFSC VTR data is a self-reporting commercial logbook system.
VTR logbook data contains fishing trip information on area fished,
kept and discarded portions of the catch, and effort information, which
includes the hull number (and/or permit number) and the port landed.
It is uncertain whether 'kept' weight in the VTR data was recorded
in live or landed pounds; both types are likely contained in the VTR
database. Live weight includes the head and guts of the catch. For
details on the VTR and Dealer data see Wigley et al. (1998). VTR logbooks
are mandatory for vessels registered under the multi-species FMP, and
voluntary for those registered under the American lobster FMP.
The NMFSPD data identifies vessels that have acquired federal permits
under the American lobster or multi-species FMP. Data elements include
the name and address of the permit holder, vessel characteristics,
and the types of gear used under each permit type. Under the American
lobster FMP, permit holders must identify on the application the management
areas they intend to fish (Figure
1). The selected management area
then determines the maximum number of lobster traps a vessel can fish.
Although an individual holds a permit to fish, this does not imply
the permit holder is actively fishing. Therefore, estimates on the
number of vessels from this database represent the number of vessels "potentially" fishing.
State tag data were provided by three of
the four states. Maine and New Hampshire's trap tag data identified
vessels by their state hull and/or federal permit number and included
the number of lobster trap tags purchased. Massachusetts did not have
trap tag data available at the time of this analysis. Instead, the
state of Massachusetts provided monthly catch data. Each data record
recorded a vessel's monthly catch and included the state hull and/or
federal permit number, the number of lobster traps fished, total pounds
of lobster landed, total number of trips and the areas fished (Figure
2). Rhode Island trap tag data were available. However, RI data
were limited to those applicants who included their state hull or federal
permit number on their application, which was approximately 35% of
the permit holders. (2)
Similar to the NMFSPD, lobster trap tag data provided by ME, NH,
and RI represent the number of vessels "potentially" fishing with traps.
In contrast, catch data provided by MA represents vessels "actively" fishing.
Data provided by the NFDOP consists of biological and economic data
collected by on-board observers on commercial fishing trips. These
data are a sub-sample of total fishing effort. In addition to the on-board
observer trip data, the NFDOP provided a list of the gillnet vessels
sampled as well as those not sampled from Maine to North Carolina.
Vessels on the list that were not sampled were identified with gillnet
gear in their possession. The list of vessels included the hull number,
the primary fishing port, however, fishing grounds were not identified.
This list of gillnet vessels provided by the NFDOP also represents
vessels "potentially" fishing.
Five databases are used in these analyses because none are comprehensive.
To analyze the lobster fleet, all databases were used except the NFDOP.
To estimate the number of gillnet vessels and the quantity of gear,
data from NEFSC's VTR, WO, and the NFDOP were used. Data on "active" gillnet
vessels are more available than data for the lobster fleet, because
the majority of vessels fishing gillnet gear fish under the multi-species
FMP and therefore have mandatory reporting.
3.2 Stratification
Data from all databases were temporally and spatially stratified to
incorporate the American lobster FMP and ALWTRT plan (NMFS, 2000).
The temporal stratification identified time blocks in which areas were
are open and closed to fishing. The spatial stratification was based
on management areas defined within the American lobster FMP and the
ALWTRT plan.
The spatial stratification consisted of two
tiers. The first tier follows the American lobster FMP stratification
(Figure
1), and the second tier incorporates the ALWTRT plan. Under
the second tier, the northern nearshore areas (Lobster Management Areas
1 and 2 within the American lobster FMP) were subdivided into northern
inshore (state waters) and northern nearshore waters for Maine, New
Hampshire, Massachusetts and Rhode Island (Figure
3). The offshore area (Lobster Management Area 3) was subdivided
into the GOM and Mid-Atlantic area. The southern nearshore area corresponds
to Lobster Management Area 4 and 5 under the American lobster FMP.
Areas closed to fishing include: 1) Stellwagen Bank and Jeffreys Ledge;
2) Cape Cod Bay Critical Habitat (CCBCH) and; 3) the Great South Channel
Critical Habitat (GSCCH).
Minor changes are incorporated to the above
spatial stratification for the gillnet fleet. Specifically, the northern
inshore and nearshore areas as defined within the lobster fishery above,
are aggregated into the northern nearshore area. "Other Northeast Waters" as
shown in Figure 4 encompass the northern nearshore
and northern offshore waters.(3)
4.0
Lobster Fleet
To determine the total impact a management regulation will have on
a fishery, the number of active vessels is the most desirable number.
However, vessels fishing under the American lobster FMP do not have
mandatory reporting. Vessels referred to as "active" have recorded
fishing activity. Vessels referred to as "potentially active" include
vessels that hold permits, do not have any recorded fishing activity
in any database, but may or may not be fishing. Both types of data
are used. A lower and upper bound estimate of the number of vessels
and quantity of gear fished in the lobster fleet are presented. Methods
are in section 4.1 and the results are in section 4.2.
4.1 Methods
In section 4.1.1, first we evaluate the databases to determine the
number of active vessels. We then proceed to estimate the number of "active" and "potentially
active" vessels, which is considered a maximum number of vessels fishing
in the lobster fleet. Section 4.1.2 presents the method to estimate
a lower and upper bound estimate of total traps and total trawls fished
by the lobster fleet. Estimates are spatially and temporally stratified
as outlined above (Section 3.2).
4.1.1 Method to estimate
the number of lobster vessels
The number of "active" vessels is determined by tracking a vessel's
hull number across NEFSC's VTR, WO and the NMFSPD databases. Any vessel
that has recorded fishing effort using lobster gear or holds an American
lobster permit is included. The percent of active vessels recorded
in the VTR and WO is compared to the number of vessels permitted. That
is, a Venn diagram approach is used to measure the ratio between the
number of active vessels and permitted vessels.
The total number of vessels "potentially active" and "actively" fishing
with lobster traps is investigated next. There are two major steps.
First, all vessel information within each database needs to be aggregated
to one record per vessel. In the second step, databases are merged
by vessel. Information from each database for each vessel is then used
to estimate the total number of vessels and assign them to a stratum.
In the first step, three databases (VTR Logbook, NMFSPD, State Tag
Data) (4) are merged after some preliminary
processing within each database. Processing of the VTR data involved
totaling a vessel's landings for each stratum (or management area)
and then assigning the vessel to the stratum where the majority of
landings occurred. This avoids assigning a vessel to more than one
area, which eliminates over-counting issues. Preliminary processing
of the NMFSPD involved collapsing all management areas a vessel planned
to fish as identified on the permit application onto one data record.
In the lobster trap tag data, a single record existed for each vessel
showing the number of purchased lobster tags and their respective states.
Therefore, once the VTR and NMFSPD databases were processed and collapsed
into one record per vessel per database, all three databases were merged
by the vessel's hull number, resulting in documentation of a vessel's
activity according to all the databases on one record. This completes
the first step.
In the second step, vessels were assigned to a stratum (fishing area),
based on which databases a vessel was recorded in and the information
contained within each database. VTR logbook data was considered the
best source of fishing trip locations. If a vessel did not have a VTR
logbook the following was assumed: 1) vessels with a federal permit
only, fish in federal waters, and; 2) vessels with a state permit only,
fish in state waters. Each state was processed separately. The detailed
rules of assigning a vessel to a fishing area and the order of the
assignment follows:
Rules of Assignment
1. All vessels with a VTR logbook are processed first, since this
is considered the best and only source of data for determining where
a vessel is actively fishing. These vessels are assigned to the management
area where the majority of their fishing activity occurs. The following
rules apply to the remaining vessels without a VTR logbook.
2. Vessels with a state permit only (and no VTR logbook data) are
assigned only to state waters. In Massachusetts (MA) only, state waters
are further divided into the CCBCH area and other MA state waters.
Vessels in the Massachusetts's state database are assigned to an area
based on their monthly landings data. Vessels fishing MA areas 1 to
5 and MA area 9 to 14 are assigned to the northern inshore stratum,
MA areas 6 to 8 are assigned to CCBCH stratum, vessels in MA area 15
to 18 are assigned to the northern nearshore stratum, vessels in MA
area 19 to 20 are assigned to Jeffreys Ledge and Stellwagen Bank stratum,
and MA areas greater than 21 are assigned to the offshore stratum (Figure
2 and Figure
4). The following rules apply to all remaining vessels not identified
in these two steps.
3. Vessels with a federal permit only and vessels with both a federal
and state permit are subdivided into vessels that chose one fishing
area and vessels that chose multiple fishing areas within the American
lobster FMP areas. Vessels which choose one fishing area are assigned
to that area. Vessels with multiple areas are prorated according to
the next rule (#4).
4. Vessels in this category chose multiple fishing areas, and must
be assigned to one fishing area. Vessels in this step are processed
by state and prorated to a management area according to the distribution
of vessels in their state who have a VTR logbook (Step 1 above). That
is, if 40% of vessels from Maine with VTR logbooks fish in offshore
waters, then 40% of the Maine vessels which chose multiple fishing
areas are assigned to the offshore waters. (5)
Once all vessels are assigned to a stratum based on these rules, vessels
numbers are then summed across all strata to estimate the total number
of vessels potentially fishing in the lobster fleet.
4.1.2 Method to estimate
lobster gear
Total gear fished is the product of the number of vessels and the
quantity of gear fished per vessel. In the lobster fishery, gear is
counted by the number of individual lobster traps and lobster trawls,
where a trawl consists of several lobster traps tied together on a
single line. The lower and upper bound estimate of gear fished per
vessel is based on recorded fishing activity, and the maximum amount
of gear a vessel can legally fish, respectively. A point estimate of
number of vessels fishing lobster traps by management area has been
estimated as described in section 4.1.1. The lower and upper bound
estimates of total gear fished is defined as the product of the point
estimate of vessels and the lower and upper bound estimate of gear
fished per vessel. Estimates in total and by strata are presented.
Two sources of data were available to calculate a lower and upper
bound estimate of the number of traps fished per vessel. (6) First,
vessels fishing lobster gear report the number of traps they hauled
on a trip in the VTR database. I assumed a vessel hauls all their traps
within one fishing trip, and further assumed, the average number of
traps hauled per trip is equivalent to the average number of traps
fished per vessel. This estimate of "VTR traps per vessel" is acknowledged
as being downwardly biased. The lower bound estimate of total traps
fished is then the product of "traps fished per vessel" (as recorded
in the VTR) and the maximum number of vessels potentially fishing lobster
traps. Second, the American lobster FMP defines a maximum number of
traps that can be legally fished per permit holder per management area.
The upper bound is therefore an estimate of the "maximum" quantity
of gear fished since it is the product of the maximum number of vessels
and the maximum amount of gear that can be legally fished per vessel.
The total number of trawls fished is equal to total traps fished divided
by the number of traps per trawl. The number of "traps fished per vessel" and "traps
per trawl" varies by management area.
Following Wilson and Plantinga (1997), we assume 50% of the vessels
in the northern inshore tie two lobster traps per trawl and 50% of
the vessels tie fifteen traps per trawl. In offshore waters we assume
vessels fish forty traps per trawl, and fifteen traps per trawl in
all other areas. (7) Number of traps
fished per trawl could not be estimated from the 1999 VTR logbook or
any other source of data at the time of this analysis. Similar to the
estimates of number of traps fished, a lower and upper bound estimate
of the number of trawls fished is also presented. (8)
4.2 Results
4.2.1 Vessel Estimation
The percent of vessels that reported fishing effort within the lobster
fleet was evaluated first. Approximately 20% of the federal lobster
permit holders show fishing activity in NEFSC's 1999 VTR and Dealer
data. (9) Due to such a low percentage
of vessels reporting their fishing activity with lobster traps, an
estimate of "active" vessels was not pursued further. Presented next
is an estimate of the number of "potentially active" and "active" vessels.
Results indicate the total number of vessels potentially
fishing with lobster traps was 7,539 in 1999. Sixty four percent (or
4,792) of the vessels were in Maine, four percent (or 320) in New Hampshire,
twenty percent (or 1,507) in Massachusetts, eight percent (or 628)
in Rhode Island, and four percent (or 292)were in the Mid-Atlantic
states (Table 1). Of the 7,539 vessels, eight
percent (or 603) filed VTR logbooks, sixty six percent (or 4,996) were
recorded in state databases only, eleven percent (or 816) had only
a federal permit, and fifteen percent (or 1,124) of the vessels had
both a state and federal permit.
The
next step of the analysis was to assign vessels to a management stratum.
Vessels with a VTR logbook were examined first and each state was processed
separately. Vessels were assigned to the management area where the
majority of their fishing occurred. In Maine, VTR logbooks show 74%
of the vessels fishing in the northern inshore stratum, 18% in the
northern nearshore and roughly 6% fishing offshore (Table
2). Similar results are available for other states. Comparing states,
74%, 100%, 45% and 22% of the vessels in the 1999 VTR fished in northern
inshore waters for ME, NH, MA and RI, respectively (Table 2, Table 3, Table 4, and Table 5, second
column).
Under the second rule of assignment, vessels with only a state permit
were examined next. Sixty six percent of the total vessels (0.66 =
4,996/ 7,539) which purchased state lobster trap tags, did not hold
a federal lobster permit or hand in a VTR logbook (Table 1). These
vessels, with the exception of MA, were assigned to the northern inshore
stratum, since they were licensed to fish in state waters only (Table 2, Table 3, Table 4, and Table 5, third column). (10)
The third rule of assignment applies to 816 vessels with only a federal
lobster permit (Table 1). There were 67, 0, 200, and 14 vessels that
chose one area of fishing in ME, NH, MA and RI, respectively, and were
assigned accordingly. Vessels that chose more than one area were assigned
next, according to the VTR distribution of vessels fishing in federal
waters only. (11) For example, according
to Maine's VTR data, 51 vessels (51=35+11+1+4 from Table 2, column
2) were fishing in federal waters only. Of these 51 vessels, 68% of
these vessels (0.68 = 35/51) fished in the northern nearshore area
(Table 2).
Now, fifty five ME vessels have a federal lobster permit and chose
multiple management areas to fish, have no VTR logbook, and need to
be assigned to a management area. Using the fourth rule, thirty seven
vessels (37=55*0.68) were assigned to the northern nearshore area,
and twelve to the offshore area (12=55*(11/51)) based on Maine's VTR
logbooks (Table 2, column 4). This same method was applied to data
from NH, MA and RI (Table 3, Table 4, and Table 5). Of the four states, Massachusetts
had the most vessels (396) holding only a federal permit.
The third rule of assignment applies to the 1,124 vessels which hold
both a federal and state permit (Table 1). There were 564, 8, and 43
vessels holding both permits that were from ME, NH and RI, respectively,
that chose one area to fish and were assigned to that management area. (12)
In cases where a vessel chose the nearshore area on the permit application,
vessels were subdivided further into the northern inshore and northern
nearshore areas according to the distribution of VTR vessels for those
two areas (Section 4.1.1, Rule 4).
In the mid-Atlantic region, the estimated number
of vessels is 292 (Table 1). These states do not participate in the
lobster trap tag program. Thus, only vessels fishing in federal waters
were examined. Using the above rules of assignment, in the mid-Atlantic
region 76% (0.76=222/292, as seen in Table 6)
of fishing occurred in the nearshore area.
The allocation of vessels to a stratum is complete. Within each stratum,
vessels were now aggregated across states.
In summary, of the 7,539 vessels fishing lobster
traps during 1999, 79% fished primarily in the northern inshore area,
9% in the northern nearshore area, 2% in the northern offshore area,
2% in CCBCH when it was open to fishing, 3% in the Jeffrey's Ledge/Stellwagen
Bank management area, 3% in the southern nearshore area and 1% in the
southern offshore area (Table 7).
A possible bias was identified in the methods section in relation
to the fourth rule. This is the case where vessels only have a federal
permit or vessels have a federal and state permit, chose multiple fishing
areas on their federal permit, and did not have a VTR logbook. Specifically,
using the fourth rule to allocate these vessels to a stratum could
result in an upward bias toward offshore vessels, if more vessels fishing
offshore were turning in VTR logbooks compared to other areas. Approximately
10% (0.097=729/7539) of all vessels fell into this category and were
assigned to a stratum using the fourth rule. If the bias exists, it
is not considered to be major. By adding an additional screening layer
such as vessel length, the potential bias could be minimized further.
Finally, suppose a limited entry program was established within the
lobster fishery which was based on a vessel's historical participation.
That is, vessels who show recorded fishing activity can remain in the
fishery. When determining the total number of vessels, we could eliminate
those vessels who have a federal permit but did not purchase trap tags
and did not show any fishing activity in the VTR logbook database.
There were 815 vessels which fell into this category (Table 1). Then
the total number of participants that showed VTR activity and/or purchased
trap tags was 6,724 vessels. (13) Although
a limited entry program based on historical participation is being
pursued, no such regulation was in place when this paper was written.
Therefore, the number of vessels "potentially active" and "actively" fishing
with lobster traps remains at 7,539 vessels.
4.2.2 Gear Estimation
Estimates of the number of traps per vessel from
the 1999 VTR data range from a minimum of 240 traps (northern inshore
area) to a maximum of 854 traps (northern offshore area) (Table
8). A maximum of 1800 traps is allowed if the vessel registers
to fish offshore exclusively and a maximum of 800 traps is allowed
for other vessels, according to the American lobster FMP (Section 2).
Total lower and upper bound estimates of lobster trawls for the Gulf
of Maine and Mid-Atlantic waters were436,620 and 1,436,979, respectively
(Table
8). In the northern nearshore area the lower and upper bound
estimates of trawls is 12,002 (12,022 trawls = 677 vessels* 266 traps
per vessel/15 traps per trawl) and 36,097 trawls fished, respectively
(Table
8). The majority of lobster trap fishing occurs in the northern
inshore and northern nearshore area.
4.2.3 Comparison of other
estimates
Total lobster traps estimated here were consistent at the aggregate
level with those presented by NMFS (1997). Total lobster trap estimates
for Maine and Massachusetts were compared with those presented by the
Atlantic States Marine Fisheries Commission (ASMFC) Report of the American
lobster (1998). Estimates of total traps in Maine by the ASMFC based
on 1998 data were within the lower and upper bound of these 1999 estimates.
Massachusetts trap estimates by the ASMFC were12 % below the 1999 lower
bound estimate. This difference may be due to the point estimate of
vessels including the "potentially active" and "active" number of vessels
fishing, where ASMFC reports "active" vessels.
Finally, the total number of trap tags purchased
in Maine, New Hampshire, Massachusetts and Rhode Island were compared
to the lower and upper bound estimate of total lobster traps in these
states. Results show that the total lobster tags sold is between the
lower and upper bound estimate (Table 9).
For example in Maine, 1.75 million trap tags were sold in 2000. Maine's
lower and upper bound estimate is 1.2 and 3.87 million traps, respectively.
For each state, the number of tags purchased is between the 1999 lower
and upper bound estimate of total traps by state. If individuals purchase
more tags then they fish, the lower bound estimate may be closer to
the true estimate of the number of lobster traps fished.
5.0
Gillnet Fleet
The number of gillnet vessels and the quantity of gillnet gear deployed
were estimated here. Since the majority of vessels fishing gillnet
gear operate under the multi-species FMP for groundfish, the VTR logbook
and dealer data are mandatory, and data on active gillnet vessels are
considered reliable. Therefore, the number of vessels presented here
represents the number of vessels actively fishing gillnet gear in 1999.
A point estimate is presented for both the number of vessels and the
amount of gillnet gear fished.
Section 5.1 presents the methods to estimate the total number of vessels
and quantity of gillnet gear actively fished. Results are reported
in Section 5.2.
5.1 Methods
Sections 5.1.1 and 5.1.2 presents the methods to estimate the number
of vessels and quantity of gillnet gear actively fished, respectively.
Estimates are spatially and temporally stratified as outlined in Section
3.2.
5.1.1 Method to estimate
gillnet vessels
The number of active vessels in 1999 was determined by tracking a
vessel's hull number in the VTR logbook, Dealer and NFDOP database.
If a vessel fished at least one trip with gillnet gear it was included.
Methods used are similar to those described within the lobster fishery
(Section 4.1). There are two step in determining the number of vessels
within each stratum.
First the number of "active" vessels is determined. A vessel's records
for the year are aggregated to one record within each database and
then merged across three databases (VTR, Dealer, SS). In the VTR, a
vessel is assigned to the stratum where the majority of its fishing
occurred. Location of fishing trips are not recorded in the dealer
data. However, the vessel's port and state of landing are identified.
A vessel in the Dealer data was assigned to the state where the majority
of its fishing occurred. Once each database is processed separately,
vessels from the VTR, NFDOP and Dealer data were merged by the vessel's
hull and permit number. Vessel hull numbers provided by the NFDOP are
reported, however, they are not used to estimate the active number
of gillnet vessels. (14)
In the second step, vessels are assigned to a stratum based on the
databases in which a vessel was recorded and the information contained
within the database. Each state was processed separately. Detailed
rules of assignment to a stratum and the order of assignment follows:
Rules of Assignment
1. All vessels in the VTR logbook were processed first and assigned
to the management area where the majority of their reported fishing
occurred. The following rules apply to vessels without a VTR logbook.
2. Vessels in the Dealer database and not in the VTR logbook database
were sorted and assigned to the state in which they landed the majority
of their catch. Within each state, vessels were prorated to a stratum
(or management area) based on the spatial distribution of vessels with
a VTR logbook fishing within their state. For example, if 50% of ME
vessels with a VTR logbook fished in northern nearshore waters, then
50% of the ME vessels without a VTR logbook were assigned to the northern
nearshore waters. This assumes at the aggregate level, vessels in the
VTR database are representative of vessels without a logbook. The states
of Maine, New Hampshire, Massachusetts, and Rhode Island were processed
individually. Vessels from the mid-Atlantic region (south of Rhode
Island) were aggregated and processed as one group.
Vessels provided by the NFDOP that have no recorded fishing activity
in the VTR or Dealer data, are not assigned to a stratum. These vessels
are "potentially" fishing gillnet gear. Once all vessels are assigned
to a stratum, vessels are summed across the strata to estimate the
total number of active vessels in the fleet.
5.1.2 Method to estimate
gillnet gear
"Total gear fished" is the product of the number of vessels and the
quantity of gear fished per vessel. The quantity of gear fished per
vessel is defined as the number of strings fished per trip and the
total length of strings being fished on average. A string consists
of several 300 foot nets tied together. Trip data collected by on-board
observers from NFDOP were used to estimate the average net length,
number of nets per string, and strings per trip by stratum. Given one
net is typically 300 feet long, then one string with 10 nets would
be 3,000 feet long. If a vessel fished 5 strings per trip, the total
length is 15,000 feet of gillnet gear per vessel.
The estimate of total gear fished is potentially downwardly, since
gillnet vessels may not pull all their strings in one trip. For example,
if a vessel fishes every day and soaks the strings for 48 hours in
the water, then the vessel may have twice as much gear fishing compared
to the recorded gear on an observed trip. Soak time and vessel fishing
frequency were not accounted for in this analysis. Vessels are not
sampled on consecutive fishing trips and the recorded unit of effort
contained in VTR logbooks are inconsistent, which prevents this type
of analysis.
5.2 Gillnet Results
5.2.1 Vessel Estimation
Based on 1999 data from Maine to North Carolina,
there were 953 unique gillnet vessels "potentially active" and "active" according
to these three databases (Table 10). The
1999 vessel list provided by the NFDOP identified an additional 407
gillnet vessels not in the Dealer or VTR. (15) There
were 546 vessels "actively" fishing gillnet gear based on the VTR and
Dealer databases, of which 250 gillnet vessels were identified in the
Dealer data (without a VTR logbook), and an additional 296 gillnet
vessels were identified in the VTR logbook.
Vessels with a VTR logbook were
assigned to the stratum in which the majority of their fishing occurred
(Table 11). Vessels without VTR logbooks
were assigned to a stratum based on fishing distributional patterns
of other vessels from their state of landing who did have a VTR logbook.
For example, there were 27 vessels from Maine with a VTR logbook (Table 10), of which 48% fish primarily in the northern nearshore waters and
52% fish in the northern offshore waters (Table 11). The 11 vessels
from Maine in the WO database without a VTR logbook (Table 10) were
prorated to these two strata, resulting in 5 vessels being assigned
to the northern nearshore area and 6 vessels to the northern offshore
area (Table 12). This step was repeated by
state for the remaining 239 vessels in the Dealer database (239=250-11)
in New Hampshire, Massachusetts, Rhode Island and the Mid-Atlantic.
According to the 1999 Dealer and VTR logbooks, the total estimate
of active gillnet vessels was 546. Approximately 57% of the fleet (312
vessels) fished in northern waters and 43% (234 vessels) fished in
the Mid-Atlantic region (Table 12). The majority of fishing occurred
in the northern nearshore (38.3%) and southern nearshore (41.9%) waters
with a smaller percentage of the fleet fishing primarily offshore (17.3%).
5.2.2 Gear Estimation
There were 947 observed gillnet trips in 1999 (Table
13). The average number of nets per string ranged between 5.2
(in the southern nearshore) to 18.4 nets (in the northern offshore).
Net length is typically 300 feet. The number of strings fished per
trip ranged between 3.9 (on the Stellwagen Bank/Jeffreys Ledge area)
and 10.7 strings (in the northern offshore area). Gillnet vessels
in the northern offshore area fished on average the most nets per
string (18.4 nets) and strings per trip (10.7 strings).
Total number of strings fished by the gillnet fleet
is 3,254 (Table 14). Total number of strings
fished by the industry in the northern nearshore stratum is 1,003 strings
(4.8 strings*209 vessels). The average length per string fished is
3,090 feet and total length of all strings fished per vessel on average
is 14,832 feet in the northern nearshore stratum.
6.0
Recommendations
Several databases were accessed to develop these estimates versus
individual estimates from each database. Key to improving our knowledge
of the number of lobster vessels was the requirement that vessels had
to purchase one tag for each lobster trap from Maine to Rhode Island
in 2000, and that the transactions were recorded. (16) The
use of purchased tag data for lobster traps, however, still only informs
us of the number of "potential" vessels fishing lobster traps. More
effort data are needed on vessels fishing lobster gear in state and
federal waters. Currently, the Massachusetts Division of Marine Fisheries
has the most detailed and comprehensive data records. Massachusetts
requires all vessels landing lobsters in their state to fill out a
monthly landings report which includes total pounds of lobster landed,
number of traps fished and trips by Massachusetts statistical area
designations. Other states may want to adopt these procedures.
Commercial landings data collected on vessels fishing gillnet gear
is considered superior to data on the lobster fleet. However, improvements
could also be made to the gillnet data. For example, VTR logbook data
needs to be more consistent across vessels. Specifically, data elements
such as the number of strings, nets, and amount of soak time are recorded
in different units, or not recorded at all. Landed catch is recorded,
but it is not clear whether the amount represents catch in terms of
live (i.e. with head and guts) or landed. As a result of these data
issues, other databases are used for these data elements. If these
data issues are corrected the VTR database has the potential of supplying
more refined units of effort which could be used in the estimation
process. For example, "average landings per string" is a more desirable
unit of effort than "total landings per trip".
Database quality typically improve over time as more analysts use
them and provide feedback.
Inconsistencies in recording data elements can be decreased by providing
fishers more examples on how to fill out the VTR reports, for example.
In some cases, data needs to be collected before its quality can be
improved.
Finally, the methods presented here may be improved. For example,
the algorithm of allocating vessels to a management area may be improved
by adding more data screening layers. Including the length of a vessel
will allow the analyst to determine whether a vessel is more likely
to fish inshore versus offshore.
7.0
References
[NARA] National Archives and Records Administration. 2000. Trap limits
and trap tag requirements for vessels fishing with traps. Code of Federal
Regulations:50(3):769, October 1, 2000
[NMFS] National Marine Fisheries Service. 1999. American Lobster Fishery;
Final Rule Federal Register 64: Page 68228, December 06, 1999
[NMFS] National Marine Fisheries Service. 2000. Environmental Assessment
of the Interim Final Rule amending the Atlantic Large Whale Take Reduction
Plan. National Marine Fisheries Service Regional Office. December 2000.
[NMFS] National Marine Fisheries Service. pers. comm. Personnel communication
with Gear Specialist. National Marine Fisheries Service Regional Office.
Gloucester, MA.
[NMFS] National Marine Fisheries Service. 1997. Environmental Assessment
and Regulatory Impact Review of the Atlantic Large whale Take Reduction
Plan and Implementing Regulations. National Marine Fisheries Service
Regional Office. July 1997.
Waring, GT., DL. Palka, PH. Clapham, S. Swartz, MC. Rossman, T. Cole,
L. Hansen, KD. Bisack, KD. Mullin, RS. Wells, DK. Odell and NG. Barros.
1999. U.S. Atlantic Marine Mammal Stock Assessments-1999. NOAA Tech.
Memo. NOAA/NMFS/NEFSC. NMFS-NE-153. 196p.
Wigley, S., M.Terceiro, A. DeLong, K. Sosebee. 1998. Proration of
1994-96 USA Commercial Landings of Atlantic Cod, Haddock, and Yellowtail
Flounder to Unit Stock Areas. NOAA, NMFS, NEFSC [Northeast Fisheries
Science Center] Ref. Doc. 98-02, Woods Hole, Massachusetts
Wilson, James and Plantinga, Andrew. 1997. The Economic Impact of
Proposed ALWTRT Regulations. University of Maine. Orono, Maine.
Footnotes
1. A "Memorandum of Understanding" exists between
NMFS and the states of Maine, New Hampshire, Massachusetts, and Rhode
Island to share fisheries data. Trap tag data were not available at
the time of this analysis from the state of Massachusetts. However,
Massachusetts did provide landings data by month and area for state
permitted vessels. States south of Rhode Island were not participants
in the lobster trap tag program.
2. On the application form to purchase tags, permit
holders were not required to identify their vessel's state hull or
federal permit number. To fully utilize the Rhode Island data, a separate
data file was needed (but not available) to match a vessel's hull number
to a person. As a result of not being able to cross the tag data to
hull numbers, the estimate of vessels fishing commercial lobster traps
in Rhode Island is considered "biased down".
3. The same stratification was used on the gillnet
fleet as on the lobster fleet above to estimate the economic impact
on both the lobster and gillnet fishery in an area where right whales
are present.
4. NEFSC's dealer data is not used as a data source
here since too few vessels reported.
5. An upward bias may exist within this algorithm
if more vessels fishing in offshore management areas (federal waters)
turn in VTR logbooks compared to the number of vessels fishing in northern
nearshore or inshore waters (state waters). Length of vessels could
be added as an additional screening layer to this algorithm to sort
out nearshore and offshore vessels. For example, typically vessels
over 50 feet are capable of traveling to offshore areas and vessels
of a smaller size are not capable of traveling long distances. The
cutoff length between vessels traveling offshore and nearshore can
be explored and used as an extension to this algorithm.
6. The number of traps fished per vessel varies
by management area, and may vary by season. However, seasonal variation
was not considered here and a constant average annual number of traps
fished per vessel is used.
7. Four members of the ALWTRT were asked what
was the average number of traps per trawl fished in nearshore waters.
They reported an average of 15 traps. The estimate of traps per trawls
offshore is based on earlier work in the 1997 ALWTRT plan.(62 FR 39157,
July 22, 1997).
8. Since we are concerned with reducing the entanglement
of right whales, the number of buoy lines in the water becomes a focal
point. On pair trawls there is one buoy line and for all other trawls
there are two buoy lines.
9. Specifically, 15%, 22%, 22%, and 42% of lobster
fishing activity is reported for Maine, New Hampshire, Massachusetts
and Rhode Island, respectively.
10. Although MA did not have trap tag data available,
their supply of catch data to the NEFSC was richer in detail compared
to the standard trap tag data, since fishing grounds were identified.
Ninety six percent of MA vessels were identified in the VTR and MA
catch data. For location information, MA catch data was chosen over
VTR data (Table 4).
11. In Massachusetts, the state landings data
were combined with the VTR logbook data to estimate the spatial and
temporal distribution of the other MA vessels which did not have fishing
location information (Table 4).
12. Since Massachusetts supplied state landings
data by area, these vessels did not need to be prorated.
13. However, the purchase of trap tags does not
imply fishing activity (1,125 vessels in Table 1).
14. The list of vessels provided by the NFDOP
were not based on observed gillnet trips The list represents a survey
of vessels that may potentially (ie. not confirmed) be fishing gillnet
gear.
15. Sixty four percent of these 407 vessels are
from the state of North Carolina and twenty one percent from Virginia.
16. However, the state of Rhode Island did not
require applicant's purchasing trap tags to identify their hull number.
Changing this requirement from voluntary to mandatory next year will
improve Rhode Island's lobster fleet data.
