Dr. Nedzad Mehic, Tom Ross, Dr. Mansoor Al-A'ali and Dr. Ghulum Bakiri

A GIS FOR BAHRAIN FISHERIES MANAGEMENT

ABSTRACT

This paper presents a GIS designed to aid the fisheries management at the Fisheries Directorate in Bahrain. The main purpose of the system is to provide rational management, fisheries statistics, fishing vessels utilization, economical forecasts, information on the location of threatened and endangered species, and fast access and easy generation of different status reports for decision makers in the fisheries management process. The system is based on ArcView Version 2.1 as it provides analytical tools to create spatial data and define spatial data relations, in addition to standard database manipulation capabilities. The available information was analyzed and a pilot system was developed for the calculation of profitability and utilization of the Bahrain shrimp fleet. The GIS developed contains the cartographic features (islands, reefs, fishing grounds and fishing block boundaries) of the coast of Bahrain, statistical attributes, and latitude and longitude grid. The system allows the user to organize, update and display data according to a dynamically set criteria and provides flexible query and analysis functions on the database and visual representation of summarized tabular data.

INTRODUCTION

Fish and especially shrimp is a very important natural resource in Bahrain. In the 1980's royal decree, the Amir of Bahrain defined strict regulations governing the fisheries industry including fishing net mesh size and location of fishing grounds. In 1985/86 shrimping season, the Directorate of Fisheries of the Ministry Works and Agriculture introduced a reporting scheme for collecting shrimp logbook information to cover both traditional and industrial shrimp fleets. In 1994 the same Directorate introduced computerized processing of these logbooks. In 1995, technical cooperation between Japan International Cooperation Agency (JICA) and the Directorate. This cooperation resulted in the design of an information processing system called Shrimp Fisheries Information System [Abdul-96]. As fishing stocks have been observed to be diminishing, the government of Bahrain has been studying a variety of measures to monitor these fishing stocks and place the necessary controls as well as give the best advice on the seasons for fishing for every species, the best areas for fishing, the types of fishing techniques etc. In order to make educated decisions having the required impact, the Directorate of Fisheries considered the development of a GIS to achieve their objectives. Collaboration between the Directorate and researchers in the Computer Science Department of the University of Bahrain has resulted in an overall system based on GIS technology aimed at preserving this vital food resource for future generations. With this system the catch in different fishing areas can be monitored and analyzed so that steps can be taken to protect breeding stocks.
This paper describes features and capabilities of the Fisheries GIS.

SYSTEM FUNCTIONS

The main functions of the system are catch and effort estimation, effort standardization, and fishing boat economics. The system has three main modules: Database module, Processing module, and Analysis and Reporting module.

The database module handles the normal file maintenance functions of the various tables used as part of the database used by the system. The processing module is employed to perform necessary calculations to produce monthly fisheries statistics, standard effort and fishing vessel economics. The Analysis and Reporting module includes display and query tools to view the data, perform analysis and prepare various reports. The system is designed to be expandable and caters for the daily, monthly and seasonal working procedures of the Directorate through user friendly tools and interfaces. The system is developed for the PC platform and can be configured for a LAN environment.

SYSTEM DEVELOPMENT

The Fisheries GIS is based on an integrated database from which data is retrieved and used to present the monthly and seasonal geographical change in the distribution of catch and fishing effort. The first task in the development process was the definition of database tables necessary for the system. A total of 11 database tables were implemented as shown as shown in Table 1.

Table 1. Data Tables

Boat Registration
Fishing boat price
Fishing Logbook
Engine price
Species Codes
Winch price
Fishing Areas Names/Positions
Fish price
Fishing Areas/Fishing Blocks
Fishing cost
Shrimp Size

The fishing statistics derived from this data included standardized values such as catch and effort estimation, effort standardization, and fishing boat economy. Parameters for these calculations include an estimation of the total number of working boats in the set of active boats, the total catch for a boat of class (i) for each fishing area (j), the total effort spent by boat of class (i) for each fishing area (j), the catch per unit of effort for boat of class (i) for each fishing area (j), the total catch and total effort, and the catch per unit effort rate U, as shown in Table 2.

The Bahrain shrimping fleet consists of vessels of different size with different fishing power. Hence it is difficult to compare fishing efforts of these vessels directly. Instead the fishing effort is standardized and expressed in standard units [Chakr-66] such as the total standard effort spent by boat of class (i) class, the catch per standard unit of effort in (j) fishing area or the standard effort and the catch per unit of standard effort, Table 2.

Table 2 Parameters for calculations

Item	Description	Calculations
N_icw	Total number of working boats in the set of active boats	N_icw = ( n_icw / n_ic) x N_i
C_ij	Total catch for a boat of class (i) for each fishing area (j)	C_ij = (SUM( c_ijw / n_icw )) x N_iw
F_ij	Total effort spent by boat class of class (i) for each fishing area (j)	F_ij = (SUM (f_ijw / n_icw )) x N_iw
U_ij	Catch per unit of effort for boat of class (i) for each fishing area (j)	U_ij = C_ij / F_ij
C	Total catch	C = SUM C_ij
F	Total effort	F = SUM F_ij
U	Catch per unit effort	U = C / F
FS_ij	Total standard effort spent by boat of class (i) for each fishing area (j)	FS_ij = ((SUM (f_ijw * p)) / n_icw )) x N_iw
US_ij	Catch per standard unit of effort in fishing area (j)	US_ij = C_ij / FS_ij
FS	Standard effort	FS = SUM FS_ij
US	Catch per unit of standard effort	US = C_ij / FS

Definitions:

N_i = Set of active boats in the fleet
n_ic = Number of boats with correct reports (n_icw + n_icnw )
n_icw = Number of working boats with correct reports
n_icn= Number of non-working boats with correct reports
f_ijw = Effort spent by n_icw boats of class i working in the fishing area j.
c_ij = Catch of n_icw boats of class working in the fishing area j

This processing requires the following database tables: Fishing Logbook, Boat Registration, Species Code, Fishing Areas, Fishing Blocks and Shrimp Size. The Fishing Logbook table includes boat number, vessel size, working status (Working/Not working), quality of the submitted report (Correct/Not correct), date of fishing, fishing area/block code, number of hauls, haul average duration, shrimp and other fish catch. Fishermen are required to keep daily logbook records and submit them to the Directorate each month.

The Boat Registration table provides information on fishing boat licenses and it is updated once every year. It includes information on the boat number, owner, engine details, winch information, vessel size class (A to F), and type.

The Species Codes table provides unique codes for most species fished commercially in Bahrain. The species are identified by species code, scientific name, family or group name and local name.

. Figure 1: Base map with fishing area

After defining the structure of databases and methodology for necessary calculations, the next task was conversion of existing paper maps and GPS coordinates into GIS coverages. The Directorate of Fisheries compiled a list of the various names of each of 47 traditional fishing areas. They used a GPS with an accuracy of +- 50 m to determine the vertices (beacons) of each fishing area and entered this information in a data file. This data file was used to generate AutoCAD scripts to draw polygons of the fishing areas. Other data files containing longitude and latitude coordinates for points along the coastlines of the major islands in the study area were used to generate polygons for the islands. Various problems which arose about the accuracy of some of the GPS records were resolved in AutoCad before exporting it to ArcView. In ArcView they were converted into shape files with accompanying attribute tables . This display was enhanced with polyline representations of the coral reefs in the study area which were taken from Esri Digital Chart of the World CD-ROMs.

The attribute table for the fishing areas was modified to include the following fields as shown in Table 3. The Area field with a numeric code of 1 to 47 is the primary key for this table. An Avenue script was used to calculate the area in square km for each fishing ground. As the fishing areas vary greatly in size, we used the catch per square km instead of the total catch for analysis and comparisons.

Table 3 Additional fields for the attribute file

Field	Type	Description
Area	Numeric	Code 1-47 for the fishing area
Name	Character	Most common names for the area
SQ_Units<.td>	Numeric	Area of the fishing ground in square km
Perimeter	Numeric	Perimeter of the area in km

The next task was to create data tables containing summary statistics extracted from the database. These tables contained the fields shown in Table 4.

Table 4 The fields for summary statistics

Field	Type	Description
Query	Character	Record selection criteria (year, month, boat etc.)
Area	Numeric	Fishing area code.
Visits	Numeric	Number of visits made to the area
Hauls	Numeric	Number of hauls made in that area
Hours	Numeric	Number of hours spent making hauls.
Catch_Tot	Numeric	Total Catch in Kg
Catch_Unit	Numeric	Catch per Sq. Km
Catch_Visi	Numeric	Catch per fishing boat per trip
Catch_Haul	Numeric	Catch per haul
Catch_Hour	Numeric	Catch per hour

These tables were populated by summaries of statistics extracted from the database. A summary table can be generated for any combination of year/ month and boat in the study period. A summary table can then be joined to the fishing areas attribute table in ArcView to generate thematic maps. Figure 2 shows a query based on Catch_Unit. Figure 2: Catch per Sq Km

A customized interface was then developed to allow the fisheries staff to select a query and see a map display. When the user chooses a selection criteria (a year, month or boat) the appropriate summary file is joined to the attribute table. The user then has a choice which statistic (total catch, catch per haul, etc.) to use to generate the map display. When that choice is made, the script will set the field for the legend and load a legend format which has the colors and labels for that display. Once the user has chosen a criteria and a legend he also has the option of printing a map using predefined layouts.

SYSTEM FEATURES

This system provides a facility for generating monthly, seasonal and trend analysis reports [Abdul-96]. These reports include thematic maps, charts and summarized tabular data. The monthly reports contain preliminary estimates of the current statistics. The seasonal reports provide the statistics for a complete season for the fishing fleet broken down by boat size, engine horse power, boat make and other parameters. Trend analysis reports provide the management of the Directorate of Fisheries with information to help refine their fisheries management strategies.

The system provides the following information:

Shrimp catch (kg), fishing effort (hr.), standard catch per unit effort-CPUE (kg per hour), density (number of shrimp per m2 per fishing ground and/or fishing block) for selected month, season; for the entire fleet, an individual boat, or a class of fishing boat.
Other fish catch (kg), fishing effort (hr.), CPUE (kg per hour), for other fish groups per fishing area for selected months and selected seasons.
Information on individual fishing boat activities based on the registration and logbook database table.
Standardization for a month or a season
Summary of operation cost for the licensed vessels
Monthly whole sale price of fish or fish group
Monthly whole sale price of fish during the season
Cost, revenue and profit by vessel sizes

This system provides several benefits to the Directorate of Fisheries including:

Better monitoring and protection of natural resources
Information on movement or distribution of fish, especially shrimp
Information on present fish resources and their trends and identifying areas that are at ecological risk [Shear-94,
Better fish price control
Generation of various recommendations for fisheries management to achieve maximum economic yield and its fishing effort
Eliminating redundancy and inconsistency [Dan-89]
Improving overall quality of fisheries management
The ArcView point-click interface incorporates buttons, tools bars and menus provides powerful analysis tools with user friendly interface for beginners and advanced users.

CONCLUSION

The paper demonstrates the use of desktop GIS as an integrating technology for the whole process of fishing management in Bahrain. This technology provides powerful visualization and analysis tools offering horizontal and vertical integration of data for planning and assisting the decision making process. This makes GIS an ideal tool for governmental agencies responsible for collecting, storage, maintenance and dissemination of large volumes of data including public records, and documents [Dan-89]. It provides tools to integrate otherwise unrelated data through spatial relations.

To further improve the system, the positional reporting by the fishermen must be more accurate. Some of the traditional fishing areas are very large and should be subdivided into smaller units for analysis. This would involve teaching the fishermen to recognize subdivisions of the traditional fishing areas or to introduce simple GPS technology for indicating the fishing location. Also, we hope that more historical data will be entered as well as data for future years to analyze temporal changes in the catch in specific locations and to identify events like pollution and dredging that are damaging the shrimp and other fish stocks. The database should be further expanded to include more detailed information on other species of fish not currently covered. A comprehensive database will provide valuable information for further ecological and economic analysis.

REFERENCES

[Abdul-95] Abdulqader, Ebrahim, Fujiwara Shunji, Bahrain Shrimp Fisheries Information System, Ministry of Commerce and Agriculture, Directorate of Fisheries, 1987.

[Abdul-96] Abdulqader, Ebrahim, Fujiwara Shunji, Development of Shrimp Fisheries Information System, Ministry of Commerce and Agriculture, Directorate of Fisheries, 1996.

[Chakr-66] Chakraborty D., Banerji S.K., On the Relative Fishing Power of Trawlers Operated Off Cochin, Indian Journal of Fish., 13 (1&2): 380-390.

[Dan-89] Dangermond Jack, The Organizational Impact of GIS Technology, Esri, 1989.

[Omar-93] Omar Adel , Implementation of a Port Planner Tools, 13th Esri User Conference, Palm Springs, California, 1993.

[Shear-94] Shearer, John G. SNOAA Relies on Tracing Technology to Speed Ocean Mapping, Geoinfo Systems, February, 1994.

Authors:

Dr. Nedzad Mehic,
Tom Ross ( tross@navtech.com)
Dr. Mansoor Al-A'ali
Dr. Ghulum Bakiri ( microcen@batelco.com.bh)

Department of Computer Science
University of Bahrain
P.O. Box 32038
Bahrain