FSWP Volume 3 - applicable excerpt

March 1988

FISHERY STATISTICS
OF THE WESTER~ PACIFIC,
VOllJME~111
-

.~-

.~.

-

Territory of Al'lwlcan Samoa (1985-86)
Commonwealth of the
Northern IVIarlana Islands (1985-86)
Territory of Guam (1985-86)
State of HawaII (1985-86)
COMPILED BY
David C. Hamm and Michael M. C. Quach
Southwest Fisheries Center Honolulu Laboratory
National Marine Fisheries Service, NOAA
Honolulu, Hawaii 96822-2396

NOT FOR PUBLICATION

ADMINISTRATIVE REPORT

H-88-4

)

Southwest Fisheries Center Administrative Report H-88-4

)

)

FISHERY STATISTICS OF THE WESTERN PACIFIC
VOLUME III

)

Territory of American Samoa (1985-86)
)

Commonwealth of the Northern Mariana Islands (1985-86)
Territory of Guam (1985-86)
State of Hawaii (1985-86)

)

)

)

Compiled By
David C. Hamm and Michael M. C. Quach
Southwest Fisheries Center Honolulu Laboratory
National Marine Fisheries Service, NOAA
Honolulu, Hawaii 96822-2396

)

March 1988
)

NOT FOR PUBLICATION

)

•

GUAM FISHERY STATISTICS, 1985 AND 1986
INTRODUCTION
The Territory of Guam (lat. 13.4°N and long. 144.4°E) is the
southernmost, largest, and most populous island in the Mariana
Archipelago. All of the islands in the chain north of Guam
belong to the Commonwealth of the Northern Mariana Islands.
Guam is located about 6,000 km (3,700 mi) west-southwest of
Honolulu, 2,500 km (1,550 mi) south-southeast of Tokyo, and
2,600 km (1,600 mi) east of Manila. Guam is about 48 km (30 mi)
long, varies from 6 to 14 km £4 to 9 m~) wide, and has an
estimated land area of 554 km (214 mi ) and a population of
about 120,000.

•

Fishing activities on Guam can be divided into two basic
categories: offshore and inshore fishing. Offshore fishing
typically involves small boat (12 to 48 feet), 1 to 2-day
trolling and bottom fishing trips that usually originate from
one of the three principal harbors located on the west coast and
southern tip of the island. Inshore fishing is typically
conducted without the use of a boat and consists mostly of
nearshore casting, netting, and spearfishing. The Guam
Department of Agriculture's Division of Aquatic and wildlife
Resources (DAWR) has been conducting offshore and inshore creel
surveys since the early 1970's. Beginning in 1982, DAWR began
modifying its data collecting and processing systems to improve
estimates of catch and effort by improving sampling techniques
and by incorporating the use of microcomputers to expand the
survey data. The WPACFIN provided microcomputers and training
and worked with DAWR staff and a contractor to redesign the
sampling program. In 1982, WPACFIN also began working with
local fish wholesalers to obtain information on the commercial
landings of Guam. It is from these two sources, DAWR and
wholesalers, that the original data for the statistics presented
in this report have come.
DATA COLLECTING SYSTEMS
The Guam data collecting systems are divided into two
distinctly different systems, one for collecting commercial
landings information and one for collecting total landings
information through creel surveys.
Commercial Landings
Fish entering the commercial market in Guam come from three
sources, full-time commercial fishermen, part-time commercial
fishermen, and subsistence or recreational fishermen who
frequently sell portions of their catch. No licenses are
required to sell fish in Guam, nor are there any reporting

(
IV.2

requirements for those selling fish.
Before 1979, there was no
central place to sell fish, so fishermen had to develop their
own markets and peddle their own fish after each trip. The Guam
Fishermen's Coop was established, via some government funding,
in Agana in July 1979.
The Coop sUbsequently became the
central distribution center for fresh local fish.
In 1982,
WPACFIN began working with the Coop to improve their invoicing
system and obtain data on all fish purchases. A cooperative
system was established whereby the Coop would use the forms and
coding schemes designed by WPACFIN and would supply copies of
all invoices to WPACFIN for entering into computer format.
In
return, WPACFIN would provide the Coop with document quality
control and computer generated summary statistics to help the
Coop improve its business. All purchase data back to JUly 1979
also were coded and computerized.
Two other fish wholesalers began operating in Guam in late
1983, and WPACFIN established similar data collecting and
processing arrangements with them. Although these two fish
wholesalers eventually left the business, one in 1984 and one in
early 1987, it is through the voluntary cooperation of all three
wholesalers that reporting on the commercial fisheries of Guam
is possible. All tables and figures of commercial landings
information included in this report are provided with the
consent of these wholesalers. The majority of fresh fish
entering the commercial market in Guam during 1985-86 were
purchased by one of the main wholesalers, although a few
fishermen still peddled their catches themselves.
Data collected on commercial forms include
Date
Fisherman code
Number of fishermen
Hours fished
Area fished
Species caught
Number of pieces caught
Pounds caught
Price per pound
Creel Surveys
The DAWR has the responsibility to monitor and protect the
wildlife and marine resources of Guam. To this end, it began
conducting creel surveys in the early 1970's. By systematic,
random interviewing of fishermen, DAWR developed a means of
estimating total catch and effort by fishing method for the
inshore and offshore fisheries.
Sampling methodologies were
frequently modified in the early years to incorporate new
information and insights gained during the surveys. Aerial
surveys were conducted for several years to help improve
estimates of percent coverage. The basic survey methodology was

(

(

(

(

(

IV.3

fairly well established by 1979.
by hand.

All data processing was done

_
In the 1970's, an annual fishing derby was organized on
Guam by groups of local fishermen. This 3-day tournament soon
became a highly successful event, with much participation by
local recreational and commercial fishermen.
The DAWR began
collecting census information on the Annual Mariana's Fishing
Derby activities as a means of obtaining additional catch and
effort information. Although the significance of these data is
minor compared to the creel surveys, summaries of derby results
are included in this document as a point of interest.

)

)

)

In 1982, WPACFIN hired a contractor to work with DAWR staff
to improve the statistical validity of the creel surveys and to
establish mathematical algorithms to expand the sample data to
estimate total catch and effort with confidence intervals.
Consequently, DAWR further improved its sampling methodologies
based on the contractor's recommendations, such as adding
surveys to better estimate total participation. The WPACFIN
developed computer processing systems to automate the data
handling and expansion activities. The system design is
flexible enough to allow for continued improvements as
additional information, insight, and funding are gained.
It is
essential for the user to understand the basic sampling design
and some of the assumptions made for the offshore and inshore
surveys to facilitate proper interpretation of the resultant
statistics.
The DAWR's fishermen interviews, also called creel surveys,
are divided into two separate, major surveys, offshore and
inshore~
Both are based on a systematic, random sampling of the
fisheries: field sampling and interviews are done on a specific
number of randomly selected weekdays and weekend-holidays each
month.
Both surveys are stratified by weekday and weekendholiday sampling and, during 1985-86, were conducted on 4-6 days
per month. Both include two subsurveys, one for counting and
estimating total participation and one for actually interviewing
fishermen for catch and effort information. Both are based on
the assumptions that the information given by the fishermen is
accurate and the fishermen interviewed are representative of the
entire fishing population.

)

Offshore Creel Survey

)

Most offshore fishing trips originate from one of three
harbors on Guam. Apra Harbor is the largest of these harbors,
serves military and commercial shipping activities, and is
considered one of the best natural harbors in the western
Pacific.
It ranks third among the harbors as points of
origination for offshore fishing trips.
Cocos Lagoon on Guam's
southern tip is the second largest protected harbor and ranks
second as a launching area for offshore fishing trips. The

IV.4

Agana Boat Basin, centrally located on the west coast of Guam in
the capitol of Agana, is the smallest of the three harbors but
is the busiest launching area for offshore fishing trips.
Therefore, DAWR selected the boat basin as the site for
interviewing offshore fishermen.
Concurrent with interviewing fishermen returning from trips
at the boat basin, a participation survey is conducted to obtain
counts of boating activity for the entire island. For
estimating total participation for a survey day, unless contrary
information is available, a boat is assumed to be fishing if it
is "out," as evidenced by its trailer at a boat ramp or being
missing from its normal berthing area. A further assumption is
made that the fishing activity and success rate of fishermen
originating at the Agana Boat Basin are not statistically
different from those of fishermen leaving from other areas on
the island. The basic premise of the offshore sampling program
is that the combined interviews collected on each survey day are
sufficient to estimate the average catch and effort for each
fishing method used during that day. Therefore, each survey day
represents a measurement of the offshore fisheries.
Data
collected during the participation portion of the offshore creel
survey are limited to boat count by launching area, whereas data
collected during interviews include the following:

*

Date (year, month, day)
Type day (weekday or weekend-holiday)
* Fishing method
* Interview time
Area fished
Boat number
* Number of fishermen
* Number of gear units
* Hours fished per gear
Total count for all species combined
Type total count
* Total weight for all species combined
Type total weight
Total number of species
Type total number of species
# Total count for each species
Type count for each species
# Total weight for each species
Type total weight for each species
# Species name (or species group)
Length for an individual fish
Type individual length
Weight for an individual fish
Type individual weight
Bait used (up to three different types)
Wind direction and speed
Weather conditions
Cloud cover

*

)

IV.5

Lunar day
Percent of catch kept
Percent of catch sold to the Coop
Percent of catch sold elsewhere
It is not always possible for the interviewer to obtain
information on all items listed. However, those marked with an
asterisk (*) are essential to the data expansion process for
estimating total catch and effort. Those marked with a pound or
number sign (#) are essential to estimating the percent species
composition of the catch. The IItype" elements (e.g., type
individual length) identify the kind of measurements, i.e.,
either actual, estimated, or calculated.
Inshore Creel Survey

)

)

)

)

Fielding the inshore creel survey is considerably more
complex and troublesome than the offshore survey for several
reasons. For instance, fishing activities originate from and
occur over a large portion of the coastline, making
participation counts and fishermen interviews much more
difficult to obtain. Additionally, it is more difficult to
obtain interviews for completed fishing trips because the
interviewer must survey many miles of coastline where fishermen
may quickly terminate their activities at any time. The
turnover rate of fishermen during the sampling period is a
difficult factor for which to adjust. Tidal stage and moon
phase also influence inshore fishing much more than offshore
fishing. Nighttime and seasonal pulse fishing are also major
considerations for thelnshore fisheries. In October 1984, DAWR
began additional survey efforts to help quantify the nighttime
and seasonal inshore fisheries.
Notwithstanding these complexities and problems, the basic
designs of the offshore and inshore surveys are very similar in
that they both have participation counts and creel interviews.
Two of the significant differences between the offshore and
inshore surveys are that the inshore participation counts are
made by fishing method as well as by location, and that
interview information is combined to form averages of catch and
effort for a much larger time period (month, quarter, year) than
a single day as in the offshore survey. Therefore, daily
measurements of the inshore fisheries are based on island-wide
participation counts for a survey day by using averages for the
catch information based on user-specified, flexible time
periods, typically quarterly and annual averages. This
modification of the expansion algorithm was required for DAWR to
physically complete an inshore survey with limited manpower.
Participation counts for essentially the entire island can be
obtained during a single sample day, but adequate creel
interviews for all methods for the entire island cannot be
obtained with the manpower available. Additionally, the
surveyable portions of the coastline are divided into three

IV.6

regions to facilitate statistically sound sampling of fishermen.
Data for the day and night surveys are processed and expanded
separately.
Data on the seasonal fisheries for juvenile
rabbitfish and bigeye scad are collected at irregular intervals
when the fisheries are active. Information collected during the
inshore participation surveys includes

*
*
*
*
*

*

Date (year, month, day)
Type day (weekday or weekend-holiday)
Location fished
Time sighted
Method used
Number of persons
Number of gear units
Reef zone fished
Weather and water conditions
Tidal stage

Information collected during the inshore interviews includes

*
*
*
*
*
*

Date (year, month, day)
Type day (weekday or weekend-holiday)
Fishing method
Interview time
Location
Reef zone fished
Number of fishermen
* Number of gear units
* Actual hours fished per gear
* Estimated trip time
Total count for all species combined
Type total count
* Total weight for all species combined
Type total weight
Total number of species
# Total count for each species
Type count for each species
# Total weight for each species
Type total weight for each species
# Species name (or species group)
Length for an individual fish
Type individual length
weight for an individual fish
Type individual weight
Bait
Wind direction
Wind speed
Weather conditions
CloUd cover
Surf
Tidal stage
Swell direction

IV.?

•

As in the offshore survey, the interviewer cannot always
obtain information on all items listed. Those marked with an
asterisk are essential to the data expansion process for
estimating total catch and effort. Those marked with a pound or
number sign are essential to estimating the percent species
composition of the catch. The "type" elements (e.g., type
individual length) identify the kind of measurements, i.e.,
either actual, estimated, or calculated.
DATA PROCESSING SYSTEMS
The Guam data processing systems are divided into two
separate and distinctly different systems, one for processing
the commercial landings data and one for processing the DAWR
creel survey data.
Commercial Landings

)

)

)

)

The processing system for the commercial landings data
collected from the wholesalers is fairly straightforward. A
purchase form is completed by the wholesaler each time fish are
purchased from a fisherman. Catches are divided into categories
for weighing by species or species group, and where practicable,
number of pieces is recorded. Preferably, coding and initial
quality control of the forms are done by Coop or DAWR personnel
be£ore they are shipped to WPACFIN for computer processing;
however, these activities must sometimes be done by WPACFIN
staff.
Data are entered into a computer and loaded into central
WPACFIN data bases, where edit reports are generated and used to
locate and correct any errors in the data base. Once all edits,
verifications, and corrections are made, summary reports are
generated. Standard reports available include total monthly and
annual landings by species, total landings by fisherman, and
landings by fisherman by species. Purchase forms are returned
to the wholesalers along with summary reports and graphs for
their use.
Creel Surveys

)

The processing systems for the creel surveys are much more
complex than those for the commercial landings data. The basic
data handling and processing systems for the inshore and
offshore surveys are the same. Data forms completed in the
field during the participation and creel surveys are returned to
the office and edited for completeness and legibility before the
data are entered into structured computer data bases by using
commercially available data base management software. Edit and
summary reports are produced to verify the quality of the data,
and any errors are corrected in the data bases. Data bases are
then translated into standard record formats, which are readable
by the data processing and expansion systems programmed by
WPACFIN specifically for the offshore and inshore surveys. As

IV.8

data are converted into the Guam Offshore Expansion System
(GOES) and the Guam Inshore Expansion System (GIES), additional
error checks are performed by the computer to ensure only valid
information enters the expansion systems. Errors are flagged
and printed to facilitate correction. The GOES and GIES are
user-friendly, menu-driven systems that step the user through a
series of processes that summarize creel survey and
participation data to produce catch and effort expansion and
species composition files and reports. Although the GOES and
GIES allow processing data for whatever time increment the user
specifies, typically 1 month of data is processed at a time for
the offshore surveys, and 3-month or annual data are combined
for the inshore surveys.
Generally speaking, the expansion algorithms for the
offshore and inshore surveys are very similar. Estimates of
total catch, effort, and participation for each fishing method
are generated from information collected during the
participation and creel surveys. The GOES uses same-day catch
and effort averages to expand the participation counts, whereas
the GIES uses user-specified, time period catch and effort
averages to expand the daily participation counts.
Inshore day
and night surveys are treated identically but separately. The
daily estimates are considered measurements of the fisheries for
that day. Average weekday and weekend-holiday estimates and
their associated variances or confidence intervals are created
from individual daily measurements. These are weighted by the
number of each type of day in the month, or other timespan, and
multiplied by proportionality constants to adjust for percent
coverage to produce estimates of total catch, effort, and
participation along with their confidence intervals. All steps
in the expansion process are stratified by fishing method. The
expansion systems produce several detailed summary reports and a
summary expansion data file containing the final totals for all
important catch and effort statistics. This summary expansion
file is later used to produce the types of reports contained in
this document.
Estimates of species composition of the expanded catch are
obtained for each method by multiplying the calculated percent
species composition of the surveyed catch by the expanded total
catch.
Percent species composition by fishing method is
obtained from the sampled catch based on the average individual
weight and the total number of individuals recorded for that
species. The average size of each species is obtained by one of
three methods, depending on the availability of data in the data
base.
If total weight and count information are available, the
average size per individual is calculated by dividing the total
weight by the total count.
If total weight and count
information are not available but individual weight measurements
for a species are available, the average size per individual is
calculated by dividing the sum of all individual weights by the
total number of individuals weighed.
If neither of these

IV.9

methods can be used because no size information is available in
the data base, the user is asked to input the species' average
size, which is then multiplied by its total count to estimate
total sampled catch of that species. Therefore, percent species
composition is calculated by dividing the estimated sampled
weight of the species by the estimated total sampled weight of
all species combined. The species composition programs produce
summary reports for immediate reference and summary data files
for· later use by reporting and summarizing software for
generating the types of reports contained in this document.

)

»

Catch, effort, and participation data collected during the
seasonal fisheries for bigeye scad and juvenile rabbitfish are
processed by hand.
Interview records are scarce, so hand
tabulations and expansions are made to produce ballpark
estimates of catch.
DATA REPORTING SYSTEMS

)

The Guam data reporting systems are divided into two
separate systems, one for reporting on the commercial landings
data and one for reporting the results of the creel survey.
Commercial Landings

)

)

)

)

)

After completing all editing and quality control activities
for the commercial landings data, monthly and annual summary
reports by species are generated. The commercial landings
reports section of this document includes monthly and annual
reports for 1985 and 1986. Volume II of this report series
contained these reports for July 1979 through December 1984.
Each report contains information on the pounds, value, average
price per pound, and number of recorded landings for each
species or species group. The number of recorded landings
("RECORDS" in the tables) is a measurement of how many times
each species was purchased, regardless of its number or weight
in the landing. This statistic is provided to give an
indication of the frequency each species is reported. The
POUNDS can be divided by the RECORDS to calculate the average
weight of each landing. Each monthly report contains a subtotal
for the sum of all species combined for that month, and the
December report also includes the annual total. Annual reports
contain the total landings for each species and the total
recorded landings for all species for the calendar year.
Included with the commercial landings summary reports are
graphs of some of the important statistics. The following
groupings of species, species categories, and abbreviations are
used in the tables and graphs for Guam's commercial landings: