History,effort distribution and landings in an artisanal bottomlongline fishery:

Anempirical study from the North Atlantic Ocean

Abstract

Commercial fishing data were used to reconstruct historical spatio-temporal patterns of

fishing effort and landings in the bottom longline fishery of the Azores. Key events during an
important 15-year period were charted, through fisher interviews together with GIS analysis.
While effort distribution varied over time, the prevailing pattern was a shift in focus from
eastern to western areas and from shelf/slope to offshore banks and seamounts in response
to policy measures (e.g. banning bottom longline inside 3 NM, public aids for
modernization of the fleet) and reduced fishing yields. Areas 12–50 NM from shore
represented the most vital fishing grounds in terms of fishing effort and production.
Declining landings of the major demersal species, in recent years, indicate that present catch
levels are not sustainable and further measures need to be taken in order to sustain fisheries
resources. Knowledge of fleet behavior and the distribution of fishing effort, particularly in
patchy environments, provide valuable insights into the impacts of past management
decisions and help predict the outcome of new policies both in this region and elsewhere.

Keywords: Spatial analysis Fishing effort Marine policy Deep Sea fishing Metapopulations

1. Introduction

A fishery's impact on its own resources is determined in large part by the distribution of
fishing effort and the type of habitat where the effort occurs. Monitoring and quantifying
changes in fishing effort is crucial for effective fisheries management [11]. The distribution of
fishing activities depends on each fishing unit's response to management measures and market
conditions, as well as fish abundance and distribution [2]. Effort information is needed to
understand changes in fisheries production, and to regulate fishing efficiency to maximize profit
and minimize over- fishing [4]. One of the more common methods used by researchers in small-
scale and industrial fisheries is the use of fisher interview data to quantify effort and gather
information on catch composi- tion per fishing trip, number of trips, vessels and effort spatial
distribution [8,18,19,11]. The use of GIS applications to allocate fishing effort on the basis of
interviews constitutes a powerful tool that helps decision makers to evaluate decisions taken
and to assess the effectiveness of policy changes [5]. Such tools can be particularly useful in
systems where fish populations are distrib- uted heterogeneously in space and fishing vessels
tend to move back and forth between different patches in order to maximize profit opportunities
[23].

The spatial distribution of fishing effort of demersal fisheries in the Azores region was, until
now, scarcely studied. Until the early 1980s, the fishing fleet was predominantly involved in
small-scale fishing, which operated mainly on the shelf and slope of the islands [13,24]. In
1985, with the adhesion of Portugal to the European Economic Community (EEC), the
archipelago of the Azores, as an outermost region, received a large amount of public subsidies,
which permitted the modernization of the fleet and the introduction of larger vessels built from
modern materials and equipped with highly technological equipment such as sonar and radar,
with enlarged storage capacity and autonomy [7,22]. At the same time, the use of gears such as
the bottom longline were widespread, enabling the expansion of the fleet to the deepest
ocean layers (400–700 m) and led to an increase in demersal fishing effort [12].

Presently, fishing activities are of great importance to the Azores’ economy with emphasis on
demersal species, which represent more than 50% of the total annual value landed in the region
[20,6]. Azorean fisheries are characteristically small-scale and artisanal with 89.7% of the
fishing vessels being small, open or closed-deck boats (length o12 m), mainly operating nearby
the islands [7,14]. The remaining fleet is composed of larger vessels (12–31 m) that operate
mostly between 3 and 200 nautical miles (NM) from shore [20,7,14]. The largest part of the
fishing fleet, approximately 87%, is multispecies and multigear [20], mostly equipped with
handlines and bottom longlines.

The blackspot seabream (Pagellus bogaraveo) is the main species targeted by the Azorean
bottom longline fishery; however, the fishery also directs its effort to a great number of other
demersal species such as wreckfish (Polyprion americanus), bluemouth rockfish (Helicolenus
dactylopterus), forkbeard (Phycis phycis), conger eel (Conger conger), and alfonsinos (Beryx
splendens and Beryx decadac- tylus), species that typically occur down to depths of 700 m on
island slopes and seamounts [13].

The Azores Exclusive Economic Zone (EEZ) covers an area of approximately 1000,000 km2,
and average depth is approximately 3000 m. Since only about 1% of the EEZ has depths of less
than 600 m [13], the area available for demersal fishing very limited and dispersed. As in other
volcanic archipelagos, the fish popula- tions are distributed across a set of spatially discrete
habitats or patches, constituting metapopulations [23]. Archipelagos with this type of habitat
discontinuity present different levels of connectiv- ity between patches, which increases the
complexity of biological interactions and processes [23]. Particularly, the Azorean offshore
patches are dependent on annual replacement of recruits from: (i) Azorean inshore nurseries
(e.g. blackspot seabream; bluejack mackerel, Trachurus picturatus [13,21]), (ii) distant regions
such as the American east coast (e.g. wreckfish; [14]) and (iii) from each seamount itself or
adjacent seamounts and slopes (e.g. bluemouth rockfish). Demersal fish populations in remote
and isolated off- shore patches are thought to receive less biomass and fewer recruits from
other patches [23], making them more vulnerable to fishing, especially when the species’ life
histories present characteristics of late maturation, slower growth, and little capacity for egg
and larval dispersion [15].
Some of the traditional Azorean fishing grounds are presently subject to intense exploitation
rates that are unlikely to be sustainable in the medium and long term [20]. The current
differences found in catch rate and biodiversity in different areas of Azores are, in many cases,
a consequence of past exploitation [14]. Therefore, the objective of this study was to reconstruct
the history of the Azorean bottom longline fishery from 1998 to 2012, in terms of distribution of
fishing effort, derived through inter- polation of information provided by fishermen, together
with landings data acquired through in situ sampling. By examining and understanding the
consequences of past management actions, it is hoped that this study will provide valuable
information on the nature of artisanal fisheries and reveal their dynamics in a patchy
environment, and in this way inform and/or support future management strategies and policies
both in this region and elsewhere.

2. Materials and methods

The Azores Archipelago consists of nine isolated islands located in the Northeastern Atlantic
(Fig. 1). These volcanic islands are distributed in three groups over an area of 600 km2 with an
orientation WNW–ESE and a coastline of 790 km [13]. In recent years, a total of 63 large and
398 small seamount-like features were identified in the region [16], however, only a few of
these are relevant to bottom longline fisheries since many are too deep or too small.

The bottom longline gear is locally known as “stone-buoy longline”. It is characterized by

a series of buoys spaced 74 m apart that permit the gear to float above the seabed, alternating
with stones that anchor the gear. The bottom longline is composed of several units and each
unit is approximately 37 m long, known as a ‘quarter-skate’ and contains 25 to 30 J-type, no. 9
hooks, usually baited with “chopped salted sardine” [13].

Data collection

Fishing effort data were collected during the period of 1998– 2012 as part of the mandate of
the Data Collection Framework (DCF). Sampling was designed to cover the main ports of the
archipelago and was performed by clerks who carried out stan- dardized interviews (n ¼ 6253)
with the captains of the bottom longline vessels on a daily basis during the landing period. The
interviews provided information on fishing effort and fishing operation, namely: the number of
hauls, fishing gear, fishing hours, bait type, no. of hooks, fishing depths, crew number, and
fishing locations on a pre-defined spatial grid of 10 ~ 10 NM [14]. Sampling focused on the four
main islands (Faial, Pico, Terceira and São Miguel Island), which represented around 98% of
the Azorean bottom longline landings. Data were pooled by island/ port, since the importance of
bottom longlines differed among the islands/ports, in terms of landings and effort, and since the
spatial distribution of the fleet among those ports was heterogeneous. The DCF data covered
22% of the landings corresponding to about 25% in weight of the demersal fish landed in the
monitored ports. Official data on landings weight per species, date, landing ports, boat name and
fishing gears were attained from the database kept by the Department of Oceanography and
Fisheries, University of the Azores (DOP).
In terms of catch composition, six principal taxa were considered: bluemouth rockfish,
conger eel, forkbeard, blackspot seab- ream, wreckfish, alfonsinos, and “others”. The category
“others” included all species for which overall landings were r3% of the total. For the
group “others” landings were proportioned over all statistical squares in which fishing was
reported to have taken place, according to the estimated associated effort.

Spatial distribution of fishing effort and landings

To calculate the spatial distribution of landings ~ and fishing effort from the bottom longline

fleet several steps were taken. Effort per 10 10 NM grid square was estimated from information
given by fishermen during the interviews. The number of hooks per trip was divided by the
number of areas fished per trip to obtain a typical effort index for each square of the mapped grid,
since absolute effort per longline set was unknown. Bottom longline landings were pooled by
island/port and year, and weighted by the per-square effort index to obtain an estimate of the
landings and effort per square. Landings per taxa per square were calculated by first identifying
the squares in which each taxa could potentially have been captured, considering knowledge of
the typical depth ranges for each species group provided by Menezes et al. [13], and applying this
ratio to the total official landings by bottom longlines from the Lotaçor S.A./DOP databases, for
the given island/port and year. Total outputs for each island were then summed to obtain the spatial
distribution of annual fishing effort and landings. The landings (total and per species) and effort
spatial distribution were plotted and calculated using GIS (ArcView 9.3) in relation to (a)
distance from the island shore (0–12 NM, 12–50 NM, 50–100 NM and 100–200 NM) and (b) per
area (coast and island slope, eastern banks, central banks, Mid-Atlantic Ridge (M.A.R.) banks, and
Great south banks (see Fig. 1). Further, the area in km2 (see Fig. 2) of fishing grounds at a depth
of r600 m, in relation to distance from the islands shore (0–12 NM, 12–50 NM, 50–100 NM
and 100–200 NM), was estimated in order to calculate the catch rates per area (kg/km2). These
calculations were performed using GIS (ArcView 9.3) tools together with bathymetric data from
Smith and Sandwell [25].

Fig. 1. Location of the study area of Azores region in the North Atlantic Ocean: (a) map
with lines indicating distance to shore [black dashed line (12 NM from shore), grey line
(50 NM from shore), grey dashed line (100 NM from shore) and black line (200 NM from
shore)]; (b) map defining different areas of Azores (Dashed line—shelf and slope insular
area, A—eastern banks, B—central banks, C—middle-Atlantic Ridge banks, D—great
southern banks. Bathymetric grey areas point out depths of ~ 1000 m.

Statistical analysis

The spatial distribution data of fishing effort and landings was submitted to arcsine
transformation to be further analyzed through Statistica 6.0 [26]. Repeated measures ANOVA
was used in order to determine if the fishing effort and landings from each zone (distance to
shore) changed over time (year). Moreover, a factorial ANOVA was used to determine if the
fishing effort and landings tended to differ due to some combined effect of distance from shore
and year. The level for statistical significance was set at α 0.05. Before calculating each
ANOVA a Levene0s test was computed to ensure homogeneity of variances.

3. Results

Bottom longline fleet production and fishing effort

On average, 169 vessels were registered each year, during the study period, as using bottom
longline gear. Numbers of boats fell slightly in the first half of the time series, reaching a
minimum of 142 vessels in 2004, but the trend turned around from that point onwards, rising to
246 vessels in 2012; driven in large part by activity amongst the smallest boats in the fleet (
o10 m) (Fig. 3). Production by the fleet followed a similar trend until 2008, when landings
began to decline despite increased numbers of boats registered (Fig. 4). From its peak, the
production of the bottom longline fleet, fell from approximately 3000 t to 2000 t, a decline that
was mainly due to decreasing landings from the larger vessels ( 412 m), which dropped by 46%
(1553 t to 843 t). On the other hand, smaller vessels ( o12 m) showed two distinct periods, the
first with a sharp decrease in production, from 1300 t to 687 t, between 1998 and 2004, and
the second period characterized by an increase of production reaching 1273 t in 2012. During
this peak, which corresponds to the highest point in the time series for vessels of o12 m,
landings from these smaller boats amounted to 60% of all bottom longline landings (Fig. 4).
The analysis of the main landed species showed some hetero- geneity (Fig. 5). Species such
as bluemouth rockfish and conger eel showed a clear decreasing trend in landings; both
species recorded maximum of landings in the year 2000 and decreased in the following years
by 43.7% and 49.8%, respectively. Landings of forkbeard decreased by 32.2% from 1998 to
2006, a tendency which later inverted and in 2012 achieved a maximum of 292 t. Blackspot
seabream showed two distinct periods, the first with rising production, achieving a maximum
in 2005 with 767 t, and a second period, after 2009, with a sharp decrease of 42.9%. Wreckfish
showed high variation in terms of landings, with a small peak in 2002 of 241 t and a maximum
in 2007 of 428 t, decreasing 24.3% in the final period. The landings of alfonsinos decreased
63.8% between 1998 and 2004; however, production began to improve in 2005 and reached a
maximum in 2009 with 295 t. “other species” showed a strong decrease between 1998 and 2002
from 1043 t to 560 t and maintained similar values in the following years with a small
increase in 2012 reaching 677 t (Fig. 5).
In general, the areas o50 NM from shore represented the most important grounds in terms of
fishing effort (Figs. 6–8) and production of the bottom longline fleet (Figs. 9–11). However,
effort, and consequently landings, from the areas adjacent to the islands ( o12 NM) fell such that
the fishing effort decreased from 66.7% to 45.5% in these areas, and landings decreased from
68.3% to 49.7% (Figs. 7 and 10). During the study period, a strong intensification of fishing
effort and production was noted in areas 50–200 NM from shore (Figs. 7 and 10), mainly on the
seamounts situated on the M.A.R. (Figs. 8 and 11). Fishing effort and landings from each zone
(distance to shore) changed significantly over time (F ¼ 19.5, d.f. ¼ 8, p o0.05). Specifically in
2002, a large proportion of effort was focused on these offshore areas (50–200 NM);
constituting 31.0% of the fishing effort and producing 27.3% of the landings (see Figs. 7 and
10). After 2003, the areas 450 NM from shore maintained their importance despite some
oscillations; for example, in 2005 a momentary shift of fishing effort and catches in the areas
closer to shore ( o12 NM) was observed. This shift was linked to an increase in landings of
blackspot seabream in nearshore waters (Fig. 5), accompanied by a temporary shift in effort
concentration in the fleet (Fig. 7).

Fig. 2. Fishing area at different distances to shore available for bottom longline (ranged
from 0 to 600 m depth, area in km 2). Restrictions on longlining within 3 NM of
shore were introduced in 2002.

Fig. 3. Annual number of vessels that used bottom longline gear (at least one landing),
categorized by size.
Fig. 4. Annual landings of bottom longline fleet, categorized by vessel size.

Fig. 5. Annual landings of the main six species from the bottom longline fleet.
 Fig. 6. Spatial distribution of fishing effort (number of hooks) of bottom longline fleet
(pooled) for 1998–2000 (A), 2001–2003 (B), 2004–2006 (C), 2007–2009 (D), and 2010–
2012 (E).
 Fig. 7. Percentage of bottom longline fishing effort in relation to distance to shore (0–12
NM, 12–50 NM, 50–100 NM, 100–200 NM and 4200 NM).

Fig. 8. Percentage of bottom longline fishing effort in relation to different areas (central
group banks, eastern group banks, islands coast, great southern banks, southwest M.A.R.
and other banks).

However, in 2006 the bottom longline fleet continued the tendency to displace an important
part of the fishing effort and catches to offshore areas, namely to the southwest offshore sea-
mounts between 100 and 200 NM from shore (Figs. 7 and 10), reaching the maximum fishing
effort recorded in that area (17.5%). In the following years, 2007–2011, the fleet maintained a
considerable level of fishing effort in the offshore areas located at more than 50 NM from the
islands (i.e. M.A.R. banks; Fig. 11). Effort and landings fluctuated owing to a combined effect of
a temporal shift in spatial preference of the fleet (F¼ 1.7, d.f.¼ 112, po0.05).
At the beginning of the study period the yield per km was higher in fishing grounds
between 12 and 50 NM, reaching a maximum of 444 kg/km2 in the year 2000. In 2001 a
decrease in catches was seen from areas of o12 NM producing 212 kg/km2, and although
landings coming from this zone oscillated during the following years, one has to take into
account the reduction of available fishable area after 2002 (ban of bottom longline inside 3
NM). Landings per km2 at areas 50–100 NM from shore increased considerably
between 1998 and 2009 reaching a peak of 686 kg/km2; however, this was followed by a sharp
decline falling to 109 kg/km2 in the year 2012. Landings per km 2 from areas 100–200 NM
from shore increased greatly between 1998 and 2006, reaching a peak of 667 kg/km2;
however, this subsequently dropped off considerably, reaching 283 kg/km 2 in 2012 (Fig. 12).
The analysis of the species composition per area showed that landings of almost all species
peaked in the year 2000, a period during which effort was mainly concentrated in the areas o50
NM from shore. These areas yielded lower returns in sub- sequent years while the landings from
banks at 450 NM were on the rise (i.e. blackspot seabream, alfonsinos, wreckfish) (Fig. 13).
The blackspot seabream was generally caught inside 50 NM, in the initial period after 2002,
however, the catches tended to decrease inside this area while increasing between 50 and
100 NM from shore, reaching a maximum in the year 2010. The year 2005 broke with this
trend, with increased landings from the 0–12 NM provenances. Forkbeard was distinct from the
other main species as it was mainly captured o50 NM from shore, with negligible landings from
areas 450 NM from shore. Trends in landings of this species were mainly affected by the
captures inside the 12 NM zone, which decreased from 1998 to 2006 and increased after 2007,
culminating in very high values in 2012, similar to those of 1998 (Fig. 13).
Overall, the average minimum depth of the fishery tended to increase, rapidly plunging from
200 m to 300 m in the period 1998 to 2002. This trend was briefly reversed in 2003 (216 m) but
returned to a downward drift in subsequent years, reaching a typical minimum of 304 m in
2011. Average maximum depth also tended to increase during the period 1998 to 2002, going
from 366 m to 545 m during that period; however, in the following years 2003–2005, the
typical average fishing depth returned to 429 m. From 2006 to 2009, maximum operational
depth stabilized at 450 m, returning to slightly shallower depths, 440–350 m, from 2010 to 2012
(Fig. 14).

4. Discussion

The time series presented encapsulates the history of the Azorean bottom longline fishery
from the end of the 1990s through to the year 2012 (Fig. 15). Until the 1980s, the Azorean
fishing fleet operated primarily along island slopes ([13] and [24]), but this behavior changed
towards the end of the decade, when public aids led to the transformation of the fleet to larger,
better equipped vessels, with greater autonomy [7,22]. With the upgraded fleet came the
capability to expand fishing effort tonearby offshore seamounts, a trend that the data shows to
be strongest between 1998 and 2007, approaching or even exceeding the limits of the EEZ.
The expansion of fishing effort to offshore areas was also coupled with the fishing strategy
of maintaining returns by shifting from depleted areas to less-exploited patches. High fishing
pressure in the eastern island group during the 1980s resulted in the local depletion of several
areas off Terceira island, the island slope of São Miguel and Mar da Prata seamount — areas
that continue to afford the lowest catch rates in the archipelago [20,14] sparking a shift in
fishing effort towards less-depleted areas in the middle of the 1990s [20]. Although the data
series does not depict this period in its entirety, the patterns seen between 1998 and 2000
probably captured the final stages of this decline, in terms of fishing effort and catches, in
eastern islands and sea- mounts (Figs. 6 and 8).

Fig. 9. Spatial distribution of landings (tonnes) of bottom longline fleet (pooled) for 1998–
2000 (A), 2001–2003 (B), 2004–2006 (C), 2007–2009 (D), and 2010–2012 (E).
 Fig. 10. Percentage of bottom longline landings in relation to distance to shore (0–12
NM, 12–50 NM, 50–100 NM, 100–200 NM and 4200 NM).

Fig. 11. Percentage of landings from the bottom longline fishery by fishing area (central
group banks, eastern group banks, islands coast, great southern banks, southwest M.A.R.
and other banks).
 Fig. 12. Annual changes in the production of fished areas by the bottom longline
fleet (kg/km2).

Depletion indicators in coastal areas lead the local fisheries authorities, in 2002, to adopt
spatial management measures in order to protect the demersal fish stocks in the vicinity of the
islands. A precautionary approach was taken, leading to a “temporary” ban against bottom
longline within the 3 NM off the shore, a measure that is actually still in place to this day. The
results showed a general drop off in the proportion of fishing effort in coastal areas and
nearby seamounts ( o12 NM), while the seamounts located between 12 and 50 NM from
shore maintained an elevated importance in all periods of study, due to the high occurrence
of favourable fishing grounds (depth r700 m) in those zones (e.g. Princess Alice
seamount and the Azores Bank) (Figs. 6 and 7). In the period 2002 to 2007, the increasing
fishing effort and landings from the M.A.R, and some of the most distant offshore seamounts
(e.g. Sarda, Monte Alto, Gigante), suggests that the 3 NM ban was an important trigger to the
spatial expansion of bottom longline fishing effort in the offshore areas of the Azorean EEZ
(Figs. 6 and 7). It is worth noting that between 50 and 100 NM from shore, the sector with the
lowest area appropriate for bottom longlining (Fig. 2), effort decreased from 2010 onwards.
Although this might be due to remotion of the fleet to other areas, there is reason to believe that
stocks in those areas were becoming depleted, and yields no longer justified the additional costs
incurred to reach those fishing grounds.
Despite the increasing importance of remote offshore areas to bottom longline fishing they
have never become as important as those closest to shore ( o50 NM), either in terms of effort or
landings. This is most likely due to the limited area with optimal conditions for bottom longline
fishing in the most distant seamounts, where the fishable areas are more isolated and dispersed
(see Fig. 2), and the deeper waters do not support the main target species of this fishery,
consequently with lower catches. In addi- tion, the greater part of the Azorean fleet is composed
of smaller vessels, few of which have the autonomy to operate in more distant areas. Further,
patches that are more remote and isolated tend to shelter fish populations that are more
vulnerable to fishing and consequently, with lower recovery rates, are depleted more rapidly
[15].
The bottom longline fishery continues to be one of the most important fishing sectors in the Azores
in terms of number of vessels (mean¼ 169, 25.6% of the Azorean vessels), landings (mean¼ 2356
t, 26.6 % of the landings) and economic value (mean¼€1.17 M). However, the observed continuous
decrease in the landings of the main species is an indication of the depletion of the demersal
fish stocks. Total landings of demersal species decreased steadily between 1998 and 2012 from
approximately 3000 t to 2000 t. Lower landings have been mainly seen in the larger vessels
( 412 m) that operate mostly in offshore areas (Fig. 4). The decrease in the catches
could also explain the reduction in the numbers of vessel above 12 m (Fig. 4) employing
bottom longlines as, after 2010, some shifted operations to deep- water drift bottom longline
gear, targeting black scabbardfish, Aphanopus carbo. Lower landings can be an indication that
the current fishing pressure cannot be supported by demersal fish populations. This is in line
with reports by Pinho et al. [21] who gave indications that the current level of effort in the
Azorean demersal fishing sector is too high to be sustainable, and has likely contributed to the
depletion of the main target species, the blackspot seabream. Despite indications of depletion,
or over- exploitation, of some Azorean demersal fish populations, the underlying process is not
fully understood [14]. An evaluation of Azorean fishing data showed a stable trend in
commercial CPUE, which probably results from an adaptive strategy on the part of the
fishers in a complex and patchy environment of seamounts and islands [10]. In fact, current
variation in catch rates and biodiversity found at different seamounts and island coasts are, in
some situations, the consequence of cumulative fishing pressure over the years rather than
natural differences [14]. This stresses the importance of this study in reconstructing the history
of Azorean demersal fishing, effort and catch distributions.
Fig. 13. Annual changes in the landed catches of the main species by the longline fleet,
according to the distance to shore of the fishing areas.
 In addition to the area closures in 2002, the Azorean fishery resource management strategy,
in line with the EU Common Fisheries Policy, implemented total allowable catches (TACs) for
several species such as alfonsinos, blackspot seabream, great forkbeard (Phycis blennoides) and
deep-water sharks namely: Centrophorus spp., Centroscymnus spp., Deania spp., Etmopterus
spp., and Dalatias licha (EC Reg. 2340/2002; EC Reg. 2270/2004). Demersal fisheries in the
~
Azores target a great number of species with complex life histories that are highly vulnerable
to fishing and, therefore, the rate of removal may exceed the rate at which the exploited
resources can regenerate. As such, the Azorean demersal fishery is probably reaching its
threshold of profitability [21]. A relevant question raised by the results is, why are bottom
longline fisheries in Azores not following the global tendency of fishing at ever-greater depths?
Although the minimum fishing depths tended to progress to deeper levels, maximum depths
remained at 700 m. Because of this, the depth range of opera- tions was more constrained in
the latter years of the survey than in the earlier years. The deeper fishing threshold is probably
related to the occurrence of less economically valuable species below 700 m, coupled with the
occurrence of several species that are protected with low or zero TACs. Since 2007, the TAC
implemented for alfonsinos was attained in the third quarter of the year, which had obvious
consequences for subsequent land- ings. Therefore, there is little incentive for the fleet to fish
deeper waters even though it has the capacity to do so.
Notwithstanding several technical measures implemented by the Azorean Regional
—
Government on bottom— longline demersal fishing since 1998 minimum landing size/weight,
minimum hook sizes, TACs for various species it can be considered a multispecies, open-access
fishery that operates over a patchy environment of islands and seamounts [14]. As stated above,
the spatial distribution of fishing effort of the bottom logline fleet in Azores is, in part, a
dynamic strategy on the part of the fishermen to maintain levels of catch rate and returns. This
strategy is typical of fragmented ecosystems, as observed by Clark [9] for the orange roughy
fishery at seamounts in New Zealand, and history has shown that quota-based management does
not protect individual seamounts against overfishing. The Azorean strategy is partially
consistent with the typical model of sequence depletion described by Berkes et al. [3], “A
fishing operation locates a profitable resource patch, fishes it to unprofitability, then moves
on, repeat- ing this sequence until there are no more profitable patches to exploit.” In the Azores
the prevailing pattern has been a shift from east to west and shelf/slope to offshore; however,
the situation is not directly comparable to the model described by Berkes et al. [3] due to the
multispecies nature of the fishery as well as the complex mix of life history strategies among
the region's species. The fishery targets more resilient species than orange roughy and,
therefore, fishing effort may continue in areas that might be depleted for one species but not
another. As a result, the pattern of fishing a patch to depletion and moving on does not
necessarily occur. Evaluating the level of depletion can also be complicated owing to the
periodic input of recruits of some species [14], that can temporarily boost resources in otherwise
depleted areas. The Azorean model is, therefore, more abstruse, with difficult-to- measure
inputs that even extend to economic pressures as well as prevailing weather conditions at any
given time. These aspects, combined, play a role in the difficulty of assessing these local
depletions when evaluating integrated data that tends to raise a stable trend of commercial
CPUE (ICES [10]).
This study shows that, to a certain extent, large-scale temporal and spatial patterns of fishing
effort and catch of bottom longlines can be delineated from commercial data; however, it should
be borne in mind that much of the data used here were derived from information provided by
fishermen during interviews and extra- polated to the landings per island/port. Estimated effort
per grid square could be improved through an observer program where effort per fishing set was
directly recorded; however, such a program has only recently been implemented for this fishery.
Vessel monitoring system (VMS) data for the bottom longline fleet, although not available for
the time period considered here, would vastly improve these projections and should prove
valuable in the future. Further, in order to determine the impact on fish commu- nities and on
particular habitats, it would be necessary to intro- duce data from research bottom longline
surveys, to complement the information described here. For example, at Condor seamount
general agreement was confirmed between the LPUE (obtained by DCF) and CPUE [14],
which provides the impetus to pursue this guideline. Moreover the analyses of fishing effort
spatial patterns in relation to economic and biological variables and the integration of the
handline fishery would be crucial to developing effective management plans on Azorean
demersal fisheries as a whole.

Fig. 14. Averaged minimum and maximum depths of bottom longline fishing
operations obtained from interviews with fishers.

Fig. 15. Schematic representation of the key events in the history of the bottom longline
fishery of the Azores, 1998–2012. The upper portion shows events in the fishery while
the lower portion shows policy and management-related events.
 Marine protected areas (MPAs) or further closed areas are presently being studied and
debated as a potential technical measure to be implemented in demersal Azorean fisheries [14].
Such measures would have to take into account, the life history strategies of deepwater species,
the recovery of populations and stocks that are likely to be much more difficult and slower than
shallower and continental margins ecosystems [17]. Moreover, the implementation of MPAs
without resizing the demersal fishing effort could have adverse consequences since the
current level of exploitation are not supported by some of demersal species (ICES [10]; Pinho
et al., [21]) and further implications on effort re- distribution would emerge (e.g. Stelzenmuller
et al., [27]). The historical relocation of Azorean demersal fishing effort from islands shelf/slope
to offshore areas was in part triggered by the ban of bottom longline inside 3 NM. The impact of
this measure in coastal areas was positive on red porgy, P. pagrus [1] and similar evidence for
other, shallower species is to be expected; for example, forkbeard is mainly found on island
shelves, slopes and nearby shallow seamounts [13] and protection measures within 3 NM,
seem to have had positive effects on landings even from other areas, since that landings now are
on par with those of 1998. However the negative impact of the relocation and concentration of
the same fishing effort in a more limited fishable area char- acterized by offshore patches with
less connectivity, habitat of many deep-water and vulnerable species are yet to be fully
evaluated.

Acknowledgements

The authors wish to acknowledge all the fishers who so graciously cooperated in this study
without whom it this study would not have been possible. We also thank Ricardo Madeiros for
assistance with GIS. This study was funded by the Azorean Government under the Data Collection
Framework of European Commission. RMH benefited from funds provided by the Directo- rate
for Science, Technology and Communication of the Azores Regional Government
(M3.1.7/F/003/2008/0000016) through the initiative Pro-Emprego.

References

1. Afonso P, Tempera F, Menezes G. Population structure and habitat preferences of red

porgy (Pagrus pagrus) in the Azores, central North Atlantic. Fish Res 2008;93:338–45.
2. Babcock EA, Pikitch EK, McAllister MK, Apostolaki P, Santora C. A perspective
on the use of spatialized indicators for ecosystem-based fishery management through
spatial zoning. ICES J Mar Sci 2005;62:469–76.

3. Berkes F, Hughes TP, Steneck RS, Wilson JA, Bellwood DR, Crona B, et al. Ecology -
globalization, roving bandits, and marine resources. Science 2006; 311:1557–8.

4. Branch TA, Hilborn R, Haynie AC, Fay G, Flynn L, Griffiths J, et al. Fleet
dynamics and fishermen behavior: lessons for fisheries managers. Can J Fish Aquat Sci
2006;63:1647–68.

5. Caddy JF, Carocci F. The spatial allocation of fishing intensity by port-based inshore
fleets: a GIS application. ICES J Mar Sci 1999;56:388–403.

6. Carvalho N. Sea to shore: an economic evaluation of the Azorean Commercial

 Fisheries. (PhD thesis). University of the Azores; 2010; 283.

7. Carvalho N, Edwards-Jones G, Isidro E. Defining scale in fisheries: small versus large-

scale fishing operations in the Azores. Fish Res 2011;109:360–9.

8. Cheung WWL, Sadovy Y. Retrospective evaluation of data-limited fisheries: a case

from Hong Kong. Rev Fish Biol Fisher 2004;14:181–206.

9. Clark M. Fisheries for orange roughy (Hoplostethus atlanticus) on seamounts in New

Zealand. Oceanol Acta 1999;22:593–602.

10. ICES. Report of the working group on the Biology and Assessment of Deep-sea
Fisheries Resources (WGDEEP). ICES CM 2012/ACOM: 17; 2012. 929 pp.

11. McCluskey MS, Lewison LR. Quantifying fishing effort: a synthesis of current methods
and their applications. Fish Fish 2008;9:188–200.

12. Menezes, G. Interacções tecnológicas na pesca demersal dos Açores. “APCC” Thesis
(Master thesis equivalent), University of the Azores, Department of Oceanography and
Fisheries, Portugal. Arquivos do DOP, Série Estudos, No. 1/1996, Horta, Azores; 1996. 187
p.

13. Menezes GM, Sigler FG, Silva HM, Pinho RM. Structure and zonation of demersal
fish assemblages off the Azores Archipelago (mid-Atlantic). Mar Ecol Prog Ser
2006;324:241–60.

14. Menezes GM, Diogo H, Giacomelo E. Reconstruction of demersal fisheries history in

the Condor seamount, Azores archipelago (Northeast Atlantic). Deep Sea Res Part II
2013;98:190–203.

15. Morato T, Cheung WWL, Pitcher TJ. Vulnerability of seamount fish to fishing: fuzzy
analysis of life-history attributes. J Fish Biol 2006;68:209–21.

16. Morato T, Machete M, Kitchingman A, Tempera F, Lai S, Menezes G, et al.

Abundance and distribution of seamounts in the Azores. Mar Ecol Prog Ser
2008;357:17–21.

17. Norse EA, Brooke S, Cheung WWL, Clark MR, Ekeland I, Froese R, et al.
Sustainability of deep sea fisheries. Mar Policy 2012;36:307–20.

18. Okada EK, Agostinho AA, Gomes LC. Spatial and temporal gradients in artisanal
fisheries of a large Neotropical reservoir, the Itaipu Reservoir, Brazil. Can J Fish Aquat
Sci 2005;62:714–24.

19. Otero J, Rocha F, Gonzalez AF, Gracia J, Guerra A. Modelling artisanal coastal
fisheries of Galicia (NW Spain) based on data obtained from fishers: the case of
Octopus vulgaris. Sci. Mar. 2005;69:577–85.
20. Pinho, MR, Menezes, G. 2006. Azorean deepwater fishery: ecosystem, species,
fisheries and management approach aspects. In: Shotton R. (Ed.), Deep Sea 2003:
Conference on the governance and management of Deep-sea fisheries. Part II:
Conference poster papers and workshop papers, pp. 330–343. Queenstown, New
Zealand, 1–5 December 2003. Dunedin, New Zealand, 27–29 November. FAO
Fisheries Proceedings. No. 3/2. Rome. The Food and Agricul- ture Organization of the
United Nations 2005, 487 p.

21. Pinho M, Diogo H, Carvalho J, Pereira JG. Harvesting juveniles of blackspot seabream
(Pagellus bogaraveo) in the Azores (Northeast Atlantic): biological implications,
management and life cycle considerations. ICES J Mar Sci,
http://dx.doi.org/10.1093/icesjms/fsu089, in press.

22. Pham CK, Canha A, Diogo H, Pereira JG, Prieto R, Morato T. Total marine fishery catch
for the Azores (1950–2010). ICES J Mar Sci 2013;70:564–77.

23. Sanchirico JN, Wilen JE. Bioeconomics of spatial exploitation in a patchy

environment. J Environ Econ Manage 1999;37:129–50.

24. Sedberry GR, Andrade CAP, Carlin JL, Chapman RW, Luckhurst BE, Manooch III CS,
et al. Wreckfish Polyprion americanus in the North Atlantic: fisheries, biology, and
management of a widely distributed and longlived fish. Am Fish Soc Symp 1999;23:27–
50.

25. Smith WHF, Sandwell DT. Global seafloor topography from satellite altimetry and ship
depth soundings. Science 1997;277:1957–62.

26. StatSoft, Inc. STATISTICA (data analysis software system), version 6. 〈www.statso
ft.com〉; 2001.

27. Stelzenmuller V, Maynou F, Bernard G, Cadiou G, Camilleri M, Crec’hriou R, et al.

Spatial assessment of fishing effort around European marine reserves: implications
for successful fisheries management. Mar Pollut Bull 2008;56: 2018–26.