This provides an overview of procedures used to collect fishery statistics for the shrimp fisheries in the Gulf of Mexico.Information about the structure of the data in the ORACLE tables will assist users with accurate use of the Gulf shrimp statistics.
Procedures and Table Structure
The Southeast Fisheries Science Center's (SEFSC) data collection
program for shrimp statistics only includes the commercial
harvesting sector.The Gulf shrimp statistics do not include
shrimp that are caught by recreational shrimpers for personal or
family consumption.Similarly, the SEFSC program does not include
the catches by small, part-time commercial fishermen that sell
their catches along the roadside as a means of supplementing their
personal income. In addition, this program does not include data
on catches of illegal or small shrimp that are discarded at sea.
The Role of the Port Agent
Shrimp statistics are collected by port agents that are located in
coastal ports along the Gulf of Mexico.Currently,
there are 24 port agents employed by either state or Federal agencies that are
participating in the SEFSC Gulf shrimp program.
Data Sources
Data for the Gulf shrimp program are collected from two sources,
and as such can be generalized into two categories. Data on the
amount and value of the shrimp are collected from seafood dealers
and because these shrimp are the ones that are actually unloaded or
landed at the dealers, they are referred to as "landings" data.
The second type of data includes detailed information on fishing
effort and location for an individual trip and is collected by
interviewing either the captain or a member of the crew.
Consequently, data in this category are referred to as "interview"
data.
Because only one port agent is responsible for a specific
geographical area, they collect the landings statistics, as well as
interview the fishermen for effort and location information.
Consequently, it is the port agent's responsibility to assure that
the right effort and location information is associated with the
landings data from the same trip. This procedure also guards
against the possibility of double counting that could occur if more
than one individual were collecting the data in the same geographical area.
Because the fishing trip is the basic data collection or sampling
unit, the fundamental principle of the data collection procedures
is to collect both landings and interview data on a trip-by-trip
basis.However, because the number of fishing trips that occur in
the Gulf shrimp fishery is so large, it is impossible for a record
to be made of every single fishing trip. Consequently, the data
collection procedures include two modifications to this principle.
The first modification is that the port agents are only required to
record landings statistics for individual fishing trip that are
made by large vessels that fish offshore. In contrast, the port
agents combine the landings statistics and record only monthly
totals for the pounds, value and number of trips for trips that are
in inshore areas by smaller boats. This consolidation is also used
for trips that are made in offshore areas, but the vessel name or
number was not available from the dealer's records.
The second modification is that the port agents only conduct
interviews with a sample of the vessels that fish offshore.
The intent of the sampling protocol is to select a few individuals that
are representative of the total population and collect the needed
information from the sample rather than the entire population.
The logistics of fishing, however, make it impossible for the port
agents to perform interviews that are selected representatively.
Most of the time the port agents do not know where and when vesselsare going to land, so specific vessels cannot be targeted in
advance for selection. As a result,
the port agents are instructed
to regularly visit the docks in their assigned areas and interview
vessel captains as the opportunity arises.
If there are more
vessels in port than can be interviewed, the agents are instructed
to select the vessels by a "random" process, thus trying to avoid
as much systematic bias as they can, i.e., always interviewing the
same vessels, at the same port, etc.
In Summary
The port agents visit shrimp dealers in their assigned
areas and collect landings statistics for individual fishing trips
for all of the vessels that fish offshore, and can be identified.
For a sample of these trips, the port agents interview a crew
member to collect fishing effort and location information.
For
inshore fishing trips, the port agents combine the landings
statistics for all of the trips that were made by the smaller,
undocumented boats.
The Objective
The objective of the data collection program for Gulf shrimp
statistics is to provide catch, value, area caught and effort data
for individual commercial fishing trips. Although this objective
is never totally achieved, the SEFSC's data do provide the most
comprehensive data on offshore, commercial catches, the majority of
which are for individual trips.A
discussion of the important aspects of the structure of the computer files in which these data
are stored follows.
Types of Records
There are three different types of records in the Gulf shrimp
database - dealer, interview and consolidated records.
The code
that designates the type of record is located in the element
labelled, RECORD_TYPE. The types
are:
1. Dealer records
These records contain data collected
from seafood dealers and are identified by a "T" in the
record type
field (a "0" is used in the 1961-1977 shrimp
files).These data provide
the amount of landings, the
value paid to the fishermen (i.e., ex-vessel price), the
official number of the vessel that landed the catch (if
available), the type of grading, the species, size of
shrimp, and type of gear. Port
agents also include their
opinion or the dealer's opinion of the location and depth
where the shrimp were caught.
2. Interview records
These records are identified by an
"S"
in the record type field (a "1" is used in the 1961-
1977 shrimp files) and contain the same
information as
the dealer records, except that information on fishing
effort is
included. Port agents interview the
vessel captain or
someone on board who is knowledgeable about
the fishing trip to determine the fishing location or
locations and
the amount of fishing effort expended on
the trip. Fishing effort is measured as the hours that
the trawls
are in the water fishing. The hours
of reported
fishing time are divided by 24 and this quotient
to the tenth
is stored in the DAYS_FISHED field.
3. Consolidated records
Port agents sometimes combine
information from more than one fishing trip into a single
'consolidated' schedule. Consolidated
records are used
for inshore
fishing trips made by small boats that are
not documented (i.e., less than 5 net tons) or for trips
offshore
where the vessel identification number was not
recorded by the dealer.Consolidated
records are coded
as a Dealer record ("T" or "0") and are also
identified
by the following coding scheme in the Official Number
field. The first four
characters of the official number
are 9s, i.e., the VESSEL_NUMBER field.
The fifth
character is coded with an 8 if the trip was made by a
vessel or a 9 if the trip was made by a boat.
The last
character is used to identify the state in which the trip
was landed (i.e., Florida west coast, 5; Alabama, 6;
Mississippi, 7; Louisiana, 9; and Texas, 8).
Shrimp statistics, in computerized form, are available beginning
with 1956.
Data
from 1956 to the
present are kept online in an ORACLE database as part of the
Southeast Fisheries Information Network (SEFIN).
Historical Data Collection Procedures
The shrimp data collection procedures were changed in 1978 in
several ways. The vessel
identification number (i.e., the Coast
Guard documentation number) was not included in dealer records.
All dealer records were coded with 9's in the vessel identification
field. The vessel identification
number was recorded when an
interview was conducted, and it was also included on the interview records. Another change that was implemented at this time was to
record only the month in which the trip occurred, instead of the
month and day.
In 1981, these procedures were discontinued and the pre-1978
procedures were re-implemented, i.e., the former method of
recording the vessel identification number and the month and day
when the trip was unloaded for both dealer and interview records.
Pounds and value for the shrimp catches are collected by size,
although the data collection procedures have changed over the
years. Prior to 1984, all landings
were grouped into a standard
set of eight market or size categories (i.e., <15, 15-20, 21-25,
26-30, 31-40, 41-50, 51-67, and >68 shrimp count per pound).
In
1984, port agents began recording the landings data in the size
ranges in which the shrimp were purchased by dealers.
In addition to changes in size ranges, more detailed information on
fishing effort were collected beginning in 1984.
This change
included the addition of data on the number of trawls, the size of
trawls, the period of the day that the shrimping occurred (i.e.,
day or night), and the actual number of hours fished.
As a result of the changes that were made in the data collection
procedures over the years, the file structures had to be changed as
well. Fortunately, all of the
various file structures are
accounted for in the ORACLE database management system for SEFIN.
All of the data in SEFIN have the same format and are stored in a
single table starting with data from 1984.
Earlier data from 1961
through 1983 will be added at a later time.
Because the original shrimp files were maintained in a fixed record
format, multiple computer records for a single fishing trip usually
occur. For example, if fishing
occurred only in one area, only one
species of shrimp were caught, and the shrimp were all the same
size (i.e., in the same market category), then only one record
would be required. However, where
there are different species or
sizes of shrimp caught, in different areas, etc., multiple records
were required for each trip. Thus, if the shrimp data are used in
a fixed record format, it is essential that all records associated
with a single trip be identified and used properly.
There are several fields that should be used to identify the
records associated with a fishing trip.
A
number is recorded on
each shrimp schedule that port agents prepare; however, this number
is not unique for the entire Gulf of Mexico.
The schedule number
is only unique for a given port and month (the month in the date of
landing field. A suggested means of
associating all of the records
with a single fishing trip is to sort the data by port, month and
schedule number, and as an additional assurance the official number
can be used in the sort.
Because the data on pounds and value are disaggregated for each
record, totals for pounds and value for a specific species, port,
area of fishing, etc. can be accomplished by simply adding all of
the data in the respective fields. However,
determining the number
of fishing trips and the amount of fishing effort (i.e., days
fished) is more complicated.
Data on the number of trips are provided in the NUMBER_OF_TRIPS
field and is always 1 for unconsolidated records. As discussed
above, port agents may combine the information for several trips
into a consolidated record. In this
case, the value entered in the
Trip field would be the number of trips combined by the port agents
in the respective schedule. The number of trips is not provided on every computer record; it is only provided on one of the records for each trip. Thus, the user must be careful to associate all of the computer records for a specific trip in order to assure that the correct numbers of trips for the desired analysis are counted.
Avoid Double Counting
Care must be taken when trip information is needed that requires records for a single trip to be split. For example, if an analysis of white shrimp is being done and the number of fishing trips on which white shrimp were caught is needed, the analyst has to makesure that the trips on which fishermen caught both brown and white shrimp are counted. It is possible that the number of trips is recorded on a brown shrimp record and not a white shrimp record and unless special provisions are made, the number of trips on which white shrimp were caught could be under-counted.Furthermore, care must be taken, because double counting of trips can occur if the number of trips is calculated separately for several species and then added together without an accurate accounting of the number of trips on which more than one species were caught.
Calculating the number of days fished has a different type ofproblem associated with it. As described above, information on the number of days fished is only provided on "Interview" records.Because interviews are conducted for only a sample of the fishing trips, the number of days fished must be estimated by expanding the sample information to the total universe that is desired. For example, if the total number of days fished is needed for shrimp caught off the coast of Texas, this number would have to be estimated by expanding the days fished information on the interview records to the total records for all shrimp caught in areas off the coast of Texas.
Because different assumptions can be made to accomplish whatever expansions are needed, it is possible that different analysts could calculate different estimates of days fished for essentially the same research question. It is important that users understand the structure of this file and utilize well conceived assumptions when estimating days fished. A method of estimating directed fishing effort for individual species of shrimp is presented in Nichols,1984.
Literature Cited Nichols, S. 1984. Updated assessment of brown, white, and pink shrimp in the U.S. Gulf of Mexico. Southeast Fisheries Center, Miami, Florida. 21 p. |
