2. MATERIALS AND METHODS

2.1 GIS APPROACH

The general approach was to:

1. Compile a database on species' physiology, culture technology and social and economic factors affecting the location of shrimp farming in ponds and fish culture in cages.

2. Establish siting criteria and tolerable and optimum values for shrimp and fish from syntheses of observations in the data base and from local experience.

3. Compile and organize information on the local environment relating to the siting criteria and assign ratings.

4. Enter, manipulate and analyse the data in a GIS.

5. Report the results in the form of maps showing locations and tables giving surface areas for various alternatives for development.

Criteria common to the development of shrimp and fish culture are water quality, salinity, and infrastructure.

2.2 WATER QUALITY

In order to provide the most synoptic view possible, water quality for aquaculture was examined in two ways: (1) water quality as indicated by time series of observations at specific locations, and (2) water quality as inferred from land uses in individual river drainages.

2.2.1 Water Quality Time Series

Brackishwater data on diskette were supplemented by raw data from the Brackishwater Fisheries Research Station, Galang Patah and by data in site selection reports (Anon., 1987a, 1987b, 1987c; Fakulti Perikanan Dan Sains Samudera, 1985). BOD, NH3-N, and pH were selected as indicative of water quality for aquaculture. In all, there were data in the form of means and ranges from 21 locations (Table 2, Figure 2). From 16 of the same locations raw data were available.

There were 543 observations of BOD and 485 of ammonia nitrogen, many of which began in the late 1970s and ended in September 1987. Thus, with the supplemental data, this collection of information is probably the most comprehensive and up-to-date that exists anywhere.

The suitability for aquaculture of each location was expressed in terms of the criteria laid out by Liong (1984):

2.2.2 Water Quality and Land Uses

Land uses in broad categories were calculated for each district from data tabulated by Wong (1980). The land uses were considered in three categories: (1) agriculture (all agriculture together with newly cleared land), (2) mining (tin mining, other mining and quarrying), and (3) urban (urban and associated areas together with estate buildings and associated areas) (Table 3). The interpretation of agriculture impacts on water quality includes erosion of cropland causing high turbidity, pesticides from crop treatments, eutrophication from fertilizer runoff and organic pollution from animal wastes and processing factories (e.g., piggeries, palm oil and rubber processing factories). Impacts of mining are heavy sediment loads and possible heavy metal contamination from tailings. Urban effects are high BOD from domestic wastes and high COD from industrial wastes.

The effect of each category of land use on water quality was estimated in relative terms (Table 3). For example, the district with the most agriculture development is Pontian with 74.8 percent of its area devoted to agriculture. Water quality in Pontian is thus more likely to be affected by agriculture than in other districts. With agriculture development in Pontian set to 100 percent, agriculture development in, say, Muar is only 7 percent of that of Pontian. Therefore the threat to water quality from agriculture is only 7 percent of that relative to Pontian.

The following rating system was developed to indicate the relative effects of agricultural, urban and mining development on each district's water quality:

This rating system reflects the present situation: that water quality problems have not yet seriously affected shrimp farming in Johor State.

Land use data are tabulated according to administrative boundaries, but in order to assess the effects of various land uses on aquaculture it is necessary to portray land uses in terms of drainage basins. Thus, land uses by category by district were transformed to land uses by river basin by overlaying. For this transformation it was necessary to assume that land uses were homogeneous in each district. Drainage basin boundaries were drawn on the Johor State map (Table 4) on the basis of evident drainage patterns. A small-scale map (Consultant Group on Water Quality, 1986) was also used as a general reference for basin boundaries.

2.3 SALINITY

According to Cook and Rabanal (1978), Penaeus monodon, the tiger shrimp, can tolerate salinities from 0.2 to 70 ppt. Salinities from 10 to 25 ppt will not affect growth. Chen (1985) indicates that 15–25 ppt is optimal for this species. The following rating scheme has been designated for siting P. monodon:

The upper limit on salinity has been increased over that of Chen (1985) to reflect the good yields being attained at the Fisheries Research Institute at Galang Patah in relatively high salinity waters. The upper limit, 33 ppt, is the highest salinity usually encountered in Peninsular Malaysia coastal waters.

Salinity data were obtained from the same sources as water quality data. In total, there were 619 observations among 21 stations (Table 2).

2.4 INFRASTRUCTURE

Infrastructure for aquaculture development was portrayed by primary and secondary roads and population centres. The former are important in determining capital and operation costs. Among the considerations are transport of heavy equipment for pond construction, the cost of local road building to reach an existing state road and the cost of land to purchase a right of way. The locations of aquaculture sites relative to district capitals illustrate the costs in terms of time and distance in order to obtain supplies and to ship cultured products to processors or to markets.

Secondary and primary roads and population centres were digitized from the Johor State map (Table 4).

Table 4

NAUTICAL CHARTS AND MAPS DIGITIZED

1. Nautical Charts

2. Soil Maps

3. State Map

2.5 CRITERIA FOR THE DEVELOPMENT OF SEMI-INTENSIVE SHRIMP FARMING IN PONDS

Criteria for semi-intensive shrimp culture in ponds include distance to a brackishwater source, rainfall, and soil characteristics in addition to the common criteria - salinity, water quality and infrastructure - mentioned above.

2.5.1 Distance to Brackishwater

The practical distance that a shrimp farm can be located from a brackishwater source is determined by the elevation of the land and the tidal range. Capital costs associated with distance are for the construction of intake and discharge canals. Operating costs are for pumping and canal maintenance. It was assumed that the maximum economically practical distance from a brackishwater source is 2 km.

The upstream limit of mangroves serves to indicate the landward limit of brackishwater. The inland limits of mangroves were shown on maps of 1:63 360-scale (Table 5). For each estuary, the inland limits of mangroves were located on the large-scale maps and the limits plotted on the Johor State map. When the shoreline of the Johor State map was digitized, the estuaries were digitized only to their brackishwater limits.

All maps 1:63 360-scale, published by the Director of National Mapping, Malaysia.

2.5.2 Soil Characteristics

Soil characteristics important for siting a shrimp farm are pH and texture. Soil boundaries were digitized from 3 maps (Table 4). Soil pH and texture characteristics were obtained from Paramananthan (1978) and from Sihibi Mohktar, Soil Division, Department of Agriculture (pers.comm.). Clay-sand-silt compositions and pH were averaged for all horizons through a depth of at least 1 m (Table 6), the maximum depth for excavation.

Briah through Yong Peng: characteristics from Paramananthan (1978); all others personally communicated by Satibi, Soil Division, Department of Agriculture.

According to local experience with acid soils the following rating system was developed for soil pH (dry):

Texture suitability for shrimp ponds was assessed using the guidelines of de la Cruz (1983), and interpreted herein mainly on clay content. The ratings were:

A listing of the soils and their individual ratings are in Table 7.

Some of the soils of the Johor coastal area are mapped as series and others are mapped as associations. In the case of associations, no data on the surface area of each series within each association were available. Therefore, it was assumed that the first, dominant soil in each series was characteristic of the entire association. In order to test this assumption, pH and texture characteristics of major and minor soils, in terms of suitability ratings for pond culture of shrimp, were tabulated side by side and compared. This comparison (Table 7) showed that there were some differences. For pH these differences amounted to only one rating unit, but texture differences were larger in three of fifteen instances.

2.5.3 Rainfall

High mortalities and slow growth of shrimp in ponds can be caused by the entry of acidic water containing high concentrations of dissolved iron. Rainfall runoff is the most important source of acid waters. The rain water becomes acidic after reacting with the pyrite in the soil (Cook, Pongsuwana and Wechassitt, 1984). Given this situation, areas with minimum rainfall are preferable to those with more rain if shrimp farms have to be sited on acid soils. Rainfall isohets in 250 mm increments were obtained from Wong (1980), transferred to the state map and digitized. The interpretation of total annual rainfall in relation to shrimp farming is to deem those areas with the least rainfall as most suitable in the Johor State context.

2.6 CRITERIA FOR FISH CULTURE IN CAGES

Criteria used to locate opportunities for the culture of fish in floating cages were shelter from winds and waves, current speed and bathymetry.

2.6.1 Bathymetry

Depth is an important criterion for floating cages to ensure that metabolic products can be flushed away by currents. Currents should not only pass through and around but also underneath the cage, and so that there is some distance between accumulated waste food and faeces on the seabed and the bottom of the cage. According to Ikenoue (1985), the allowable minimum depth between the bottom of a floating cage and the seabed is 1 m. Most cages are about 2 m deep. Thus, a minimum depth at low tide for floating cages is 3 m. Tiensongrusmee, Pontjoprawiro and Soedjarwo (1986) rated depths (below the cage bottom) as: >10 m, good; 4–10 m fair; <4 m, poor.

Chua, Hooi and Wang (1985) present data that show anchoring costs as a percentage of capital investment costs. These increase from 3 percent for cages moored in shallow water to 6 percent for those in deeper water.

Depth contours on metric nautical charts are at 2, 5, 10 and 20 m. These contours were digitized except for the northeast coast of Johor where it is evident that the 20 m depth contour extends well to seaward and beyond a convenient distance to service cages from land, except around islands. Taking into account water circulation needs and anchoring costs, the interpretation of bathymetry as to suitability for floating cages is thus:

The charts from which bathymetry was digitized are listed in Table 4. Chart datum is the lowest astronomical tide and thus portrays the minimum depth available for cages.

2.6.2 Shelter

Shelter from wind and waves is an important siting criterion because storm winds and waves damage installations. Constant motion due to the action of lesser waves will cause wear on the cage structures and associated raft.

Monthly wind data from Johor Baharu and Mersing for 1985 (Anon., 1987) showed that winds equal to or greater than Force 8 (62–74 kph) occurred on 6 days at the former and on 16 days at the latter. Winds equal to, or exceeding Force 6 (39–49 kph) occurred during 139 days at Johor Baharu and 184 days at Mersing. During the most windy months north and northeast winds were the most frequent. At both stations north winds predominated in speed and frequency.

Daily data for 1986 indicated that highest wind gusts at Johor during any one month ranged from 46 to 94 kph and at Mersing gusts ranged from 52 kph to 91 kph (Anon., 1986, 1987; Anon., 1987 a). Over the year gusts averaged 60 kph at Johor Baharu and 67 kph at Mersing.

Allowable wave heights for floating cages are given as less than 0.5 m by Ikenoue (1985) and 1 m by Chua (1978); however, it is believed that the latter limit is too high and a more conservative maximum wave height was adopted, 0.6 m.

Wave height is a function of wind speed and depth. Lesser wave heights are generated in shallower water than in deeper water for a given wind speed. Graphical solutions to estimate wave height have been prepared by the US Army Coastal Engineering Research Center (1977). Using wind conditions consistent with the observations mentioned above, with gale force winds (Force 8) of 70 kph, waves of 0.6 m height would be generated at a depth of 11 m with a fetch of 2 km. A 3 km fetch would give a 0.8 m wave height with the same wind speed and depth. Also considering the distance offshore that would be convenient to supply and maintain the floating cages as well as for the comings and goings of the live-aboard staff and their security, it was decided to adopt the 0.6 m wave height and maximum 2 km fetch as representative of shelter under Johor State wind conditions.

In order to represent shelter, a line 2 km south of the shoreline, or outer limit of the mud bank, was drawn on the same nautical charts as used for bathymetry (Table 4) and digitized.

2.6.3 Current Speed

Current is an important criterion for cage siting because water movement is needed to remove metabolites from the cages and to replenish dissolved oxygen. Too much current, however, can strain anchors and lines and cause fish to expend energy in maintaining their positions at the expense of growth. Chua (1978) gives the allowable current range of from 0.2 to 0.5 m/s; Ikenoue (1985) indicates a maximum of 0.3 m/s; however, if the area is narrow and shallow, then 0.5 m/s is allowable. Ratings of current velocities have been made by Tiensongrusmee, Pontjoprawiro and Soedjarwo (1986), and these have been adopted herein:

Current speeds were obtained from the nautical charts listed in Table 4.

2.7 REMOTE SENSING

Satellite data have a number of applications in GIS which are related to updating and expanding information available from other sources.

Three scenes of SPOT digital data were available, one covering a part of the west coast and the other two the east and southeast coasts. The path and row numbers and dates of acquisition are:

2.8 VERIFICATION OF GIS RESULTS

Five days were set aside for the verification of GIS results. The objective was to test the criteria and the rating systems assigned to them.

There are several ways to obtain verifications. On-the-ground, in-the-water verification is the most reliable and also the most time consuming. Such an approach is appropriate to verify individual sites after the GIS has been employed to identify the most promising areas. The approach was to compare the locations and site-related performances of existing aquaculture facilities with locations and location ratings provided by the GIS. Performances were evaluated by observations during site visits and by data from interviews. The information volunteered was treated as confidential and therefore shrimp farms are not named, but referred to by number to facilitate comparisons among performances.

2.8.1 Verification of Shrimp Farming Opportunities

The objective was to acquire information that could be used to evaluate the success of an enterprise dependent on its location separate from its management. The information sought was: soil type, texture or acidity problems and remedial measures, water quality and related problems, salinity ranges and related problems, indications of good location and/or management as manifest in yield performance, time and distance from a metalled road, and distance to the water supply.

A supplementary objective was to gather information to evaluate the economic consequences of siting on areas deemed less than suitable through GIS analyses.

2.8.2 Verification of Cage Culture Opportunities

As with shrimp farming opportunities, the objective was to identify the spatial factors contributing to the success (or lack of it) of an enterprise as well as the economic implications of spatial choices. The information sought at cage culture sites included depth of water, shelter, security, water quality and agglomeration.

2.9 SUMMARY OF CRITERIA

Table 8 summarizes the GIS main criteria and sub-criteria for shrimp farming and fishculture.

2.10 HARDWARE AND SOFTWARE

The ERDAS (Earth Resources Data Analysis System, Inc.) system version 7.2 was used for GIS and image processing. The ERDAS PC system was developed for the IBM family of computers with the DOS operating system. Basically it is a tool for generating, managing, displaying, and processing geographic and image data.

Figure 3 shows the modules and functions of ERDAS.

The hardware and software components used by the project are listed in Table 9.