420 
Abstract— Aerial photographic assess- 
ment is a promising technique that 
could be structured to yield a fishery- 
independent index of abundance for 
Atlantic bluefin tuna, Thunnus thyn- 
nus thynnus ( ABTl.The accuracy of this 
approach may be increased by incor- 
porating the relationship between the 
surface characteristics of a school and 
the total number of individuals. Our 
objective was to develop models to facil- 
itate the estimat ion of number of fish in 
ABT schools from aerial photographs. 
Video cameras were used to observe 
74 incidences of schooling for 50 cap- 
tive ABT approximately one meter in 
length. Relationships between the sur- 
face characteristics of ABT schools and 
the number of fish in the school were 
explored by using least-squares regres- 
sion. The schools ranged in number from 
2 to 45 individuals. A weighted regres- 
sion model incorporating the number 
of fish in the school at the surface 
as the independent variable and the 
number of fish in the remaining por- 
tion of the school yielded an r 2 of 0.74. 
A second weighted multiple-regression 
model incorporating the number of 
fish in the school at the surface and 
in the second depth interval (0-25% 
school depth below surface layer) of the 
school as independent variables, and 
the number of fish in the remaining 
portion of the school as the dependent 
variable, with 1/variance as the weight, 
achieved an r 2 of 0.70. A third model 
using the length and width of the sur- 
face layer of the school as the indepen- 
dent variables and the number of fish 
in the school as the dependent vari- 
able had an r 2 of 0.86. One data point 
from a wild school is currently avail- 
able to verify model predictions. This 
school of 125 individuals is well outside 
the range of school sizes used to con- 
struct the model (2-45 individuals), yet 
differs from model predictions by only 
7%. 
We believe that these models have the 
potential to improve an abundance index 
based on aerial photographs by estimat- 
ing the number of individuals in wild 
ABT schools from surface characteris- 
tics observed in aerial photographs. 
Manuscript accepted 26 January 2001. 
Fish. Bull. 99:420-431 (2001). 
Estimating the number of fish in Atlantic 
bluefin tuna ( Thunnus thynnus thynnus) schools 
using models derived from captive 
school observations 
Brian Hanrahan 
Francis Juanes 
Department of Natural Resources Conseivation 
University of Massachusetts 
Amherst, Massachusetts 01003-4210 
E-mail address (for F. Juanes, contact author): |uanes@forwild. umass.edu 
The bluefin tuna ( Thunnus thynnus ) is 
distributed worldwide in temperate and 
subtropical seas. It has a limited dis- 
tribution in the southern hemisphere. 
Endothermy by means of vascular heat 
exchangers allows bluefin tuna to inhabit 
a wide thermal niche and therefore wide 
geographic and depth ranges (Carey and 
Teal, 1969; Carey and Lawson, 1973). In 
the western Atlantic Ocean, the Atlantic 
bluefin tuna (ABT), Thunnus thynnus 
thynnus, is distributed from Labrador 
to Brazil, including the Gulf of Mexico 
and Caribbean Sea. Adult ABT occur 
throughout the entire range, but smaller 
bluefin tuna (less than 45 kg) are not 
observed frequently above the latitude 
of Cape Cod, Massachusetts. The Atlan- 
tic bluefin tuna is epipelagic and usu- 
ally oceanic but appears near the coast 
seasonally (Squire, 1962; Collette and 
Nauen, 1983) to feed on concentrated 
assemblages of prey. Adult ABT may 
attain a length of four meters and a 
body mass of 680 kg. Large medium 
(178-195 cm FL) and giant (>195 cm 
FL) bluefin tuna are targeted by com- 
mercial purse-seine, long line, and hook- 
and-line fisheries (Mather, 1974; Figley, 
1984). A recreational hook-and-line fish- 
ery (Mather, 1974; Figley, 1984) targets 
all sizes of bluefin tuna as they appear 
along the east coast of the United 
States and Canada from June to Octo- 
ber (Mather, 1962). 
The combination of changes in the 
spatial distribution over time and as- 
sociated uncertainty regarding the in- 
dependence of eastern and western At- 
lantic stocks makes the estimation of 
ABT stock size particularly problemat- 
ic. The stock assessment for this species 
has been based upon landings data and 
abundance indices (Scott et al., 1993). 
Use of ABT landings data to generate 
an abundance index may lead to bias 
due to variability in effort, improve- 
ments in fishing technology (Lo et 
al., 1992), and variability in annual 
geographic distribution linked to prey 
distribution. 1 These characteristics of 
the northwest Atlantic bluefin tuna 
fishery, in conjunction with popula- 
tion-level behavioral characteristics ob- 
served for similar tuna species, suggest 
that the use of catch-per-unit-of-effort 
(CPUE) data to evaluate tuna popula- 
tion trends could lead to inaccurate es- 
timates (Clark and Mangel, 1979). The 
accuracy of CPUE-based assessments 
in estimating the abundance of bluefin 
tuna in the northwest Atlantic remains 
controversial (Clay, 1991; Suzuki and 
Ishizuka, 1991; Safina, 1993). An exten- 
sive discussion of the issues involved in 
Atlantic bluefin tuna assessment can be 
found in the National Research Council 
report by Magnuson et al. ( 1994). 
Recent investigations have focused 
on the feasibility of using aerial pho- 
tographic assessment of large medium 
and giant bluefin tuna in New England 
waters and in the Straits of Florida 
(Lutcavage and Kraus, 1995; Lutcav- 
age et al., 1997) as an alternative fish- 
ery-independent method of obtaining 
indices of abundance. The aerial survey 
'Chase, B. C. 1995. Preliminary report of 
the Massachusetts bluefin tuna investiga- 
tion: the diet of bluefin (Thunnus thynnus ) 
off the coast of Massachusetts. Massa- 
chusetts Division of Marine Fisheries. 
Salem, MA 01947, 39 p. 
