Application of Geographical Information Science to the Northwest Pacific Albacore (Thunnus alalunga) Fishery: Biological and Physical Interactions. Michael Thompson- Marine Resource Management, College of Oceanography and Atmospheric Sciences, Oregon

1
Application of Geographical Information Science
to the Northwest Pacific Albacore (Thunnus
alalunga) Fishery Biological and Physical
Interactions.Michael Thompson- Marine Resource
Management, College of Oceanography and
Atmospheric Sciences, Oregon State University
Abstract Albacore tuna (Thunnus alalunga)
are a commercially important fish species which
migrate annually from the waters of the Western
Pacific to the waters off the U.S. Pacific
Northwest coast. Despite their economic
importance, little is known about the biological
and physical interactions that take place during
this migratory period. This project was designed
to investigate the physical interactions between
age class and sea-surface temperatures (SST) and
biological interactions, including age class and
lipid content relationships. Albacore that were
collected in the summer of 2003 were found to be
concentrated within a narrow range of SST,
between 16 and 19.94 Celsius, and age class did
not significantly correlate with SST (p0.574).
Some trends are present with regard to spatial
interactions between age classes, however, more
research needs to be done to determine if
albacore schools are age specific. Lipid content
does correlate with the date of capture (plt0.001)
but does not show any spatial trend or
relationship with size (p0.818).
Results The results of the age class SST
analysis show no correlation between age class
and ocean surface temperatures (p0.574) using a
linear regression model. The spatial
distribution of tuna of all age classes does not
show any recognizable trend and no relation to
strong ocean fronts was found to exist.
Discrepancies were found between the SST values
form the satellite imagery and those collected
with in-situ devices on board the fishing
vessels. Satellite derived SSTs were
approximately 4 degrees below those recorded on
the vessels, even though night-time SST data were
used (which tend to reduce the variability of
day-time SST due to backscattering and refraction
of light).
Figure 8 Age class point features converted to
raster with spatial analyst tools. From fish
caught on August 25, 2003. Temporal dimension
has not been taken into consideration so actual
trends may not be recognizable. Age classes go
from light (2 years) to dark (4 years).
Discussion Utilizing GIS tools to analyze
the interactions of marine fish with their
biological and physical environment can lead to a
greater understanding of their behaviors and
provide, both the fishing industry and fishery
managers with a valuable tool for building
sustainable fisheries around the world. The
results of this project, although inconclusive,
do provide some evidence that age classes tend to
school together and that fish caught later in the
season tend to have higher lipid concentrations.
It also provides evidence that GIS applications
can be used to investigate these and other
questions about the interactions that fish have
with their environments. Since many vessels
already collect spatial information about their
fishery operations, the use of this information
can be beneficial to, not only the fishermen, but
fishery managers by expanding the amount of data
they have access to when making critical
decisions about the health and composition of
fish stocks.
Figure 2 Albacore collected between August 24th
and 31st of 2003 and their relation to SST. SST
GeoTIFF was provided by MODIS Terra AVHRR
satellite imagery in 8 day intervals with a 4 km
resolution. Contours were added with ESRI
spatial analysis tools.
Materials and Methods The 371 albacore used
for this project were collected during the summer
of 2003 for two separate experiments, a quality
and handling experiment and a lipid
quantification study. Fish used for the quality
and handling study were collect by the researcher
on-board the vessel and those used for the lipid
study were collect by fishermen. ESRI ArcGIS 8.2
software was used to spatially analyze the data
and S-PLUS 6.1 was used to conduct the
statistical analysis. Geo-reference TIFFs were
obtained for SST through NASA (http//podaac-esip.
jpl.nasa.gov/poet/) from MODIS Terra AVHRR
satellite imagery in 8 day intervals with 4 km
resolution. SST was interpolated for
individual days using the spatial analyst tools
within ArcGIS in order to provide resolution
higher than the 4 km satellite imagery. Spatial
interpolation was also used for age class
relationships. Individual days were used due to
the highly migratory nature of albacore and the
unlikelihood that they would stay within the same
location on subsequent days. Radial Basis
Functions were used to provide prediction fields
for separate age classes.
Figure 4 Age class vs. SST fitted value plot
showing no correlation between the age of the
fish and the water temperature that they were
caught in (using on-board temperature
measurements).
Figure 5 Lipid percentage vs. Date of capture
fitted value plot showing a positive correlation
between higher fat content and later capture
dates.
Spatial interpolation of age class fields
did indicate the possibility that individual age
classes did school together. Trends were
observed on-board and can be seen in the trend
plots for the individual days. Spatial
interpolation and raster layers show that age
classes tend to be caught in the same locations
and linear regression analysis comparing age
class and location was significant (plt0.01).
However, the correlation was not high (R20.136).
Lipid percentage did not show a relationship
with location or age class, but was shown to be
positively correlated with the date of capture
(plt0.001).
Figure 1 The locations of all albacore
collected during the summer of 2003, including
fish collected for handling and quality project
and lipid analysis. Fish located north-west of
Hawaii were collected during June and those off
the Pacific coast were collected between July and
October.
Introduction The albacore fishery is
considered one of the last open-access fisheries
in the U.S. and due to its highly migratory
nature, little is actually known about the
biological and physical interactions that occur
within this species. On their migration across
the Pacific they have been found to follow the
transition zone between the warmer Central
Pacific gyre waters and the cooler waters of the
North Pacific as the warmer water moves north and
east during the spring and summer (Bartoo and
Foreman 1994, Polovina et. al 2001). During this
time the U.S. surface troll fleet has access to
the fishery and usually begin their annual season
during June and continue to follow the fish along
the West Coast through October, when they begin
their return migration (Laurs and Dotson 1992).
Beyond this limited information very little else
is known about their behaviors and how they
interact with the physical parameters of the
ocean. Observations during the summer seemed to
indicate that age classes tend to school
together, since fish of approximately the same
size were caught at the same times and locations.
Expanding our knowledge through the use of
Geographical Information Science (GIS) of this
commercially important tuna, which is coming
under increasing pressure due to the decline in
other tuna species populations, is vital if we
want to maintain healthy stocks and provide new
tools for the international regulation of
albacore (some work has been done with GIS on
blue fin tuna by Schick 2002). In addition to
providing new tools for fishery managers,
including the use of GIS analytical techniques,
GIS applications may provide new marketing
opportunities for the fishing industry and allow,
both fishery managers and the fishing industry,
to work together in providing a sustainable level
of harvest in the albacore fishery.
References Bartoo, N. and Foreman, T. 1994. A
synopsis of the biology and fisheries for North
Pacific albacore tuna. In Interactions of Pacific
Tuna Fisheries. R. Shomura, J. Majkowski and S.
Langi, eds. Proceedings of the first FAO expert
consultation on interactions pf Pacific tuna
fisheries. 3-11 December 1991. Noumea, New
Caledonia. FAO Fish. Tech. Paper. 336(2)
173-187. Laurs, M. and Dotson, R. 1992. Albacore.
In Californias living marine resources and their
utilization. W.S. Leet, C.M. Dewees and C.W.
Hauger, eds. Pp. 136-138. Polovina, J., Howell,
E., Kobayashi, D. and Seki, M. 2001. The
transition zone chlorophyll front, a dynamic
global feature defining migration and forage
habitat for marine resources. Progress in
Oceanography. 49 469-483. Schick, R. 2002. Using
GIS to track whales and bluefin tuna in the
Atlantic ocean. In Undersea with GIS. Dawn
Wright, ed. ESRI Press, Redlands, CA. 2002.
Figure 6 Trend analysis for age class and
location using the geo-statistical analyst in
ArcGIS 8.2. Fish of the same age tend to be
caught at similar times and location, however,
some variability does exist.
Figure 3 Spatially interpolated age class
layers, using mean age, showing some spatial
relationships between fish of the same ages. A)
fish caught on August 27, 2003 B) fish caught on
August 26, 2003 C) fish caught on August 25,
2003. Fish of the same age class tend to be
caught in similar locations but some variability
can be seen, especially between different days.
Figure 7 Radial Basis Function showing
prediction fields for age classes. However, due
to the variability, predictions of age class
specific harvest would be limited.