CHAPTER 2 - SPATIALLY VARIABLE PRODUCTION FUNCTIONS CONTROLLING AQUACULTURE AND INLAND FISHERIES

2.1 Production Functions

The pursuance of any human activity will be controlled by a varying number of factors. Those factors controlling economic activities have been called “production functions”, i.e. since what is produced is a function of various combinations of the controlling factors. The necessary combination of production functions will vary both between and within and enterprises in an almost unlimited way; usually more complex activities will involve more production functions. Production functions are most frequently economic factors though they may also be physical factors and to a lesser extent social factors. Most economic production functions will be synonymous with “fixed” or “capital” costs in that they involve “once only” consideration, usually when the activity is being initiated. Other functions will be variable, i.e. synonymous with “variable operating” costs and considerations.

All production functions will show spatial variability. Specific functions will vary greatly within a given area - others hardly at all. A particular production function may vary greatly within one country (or region) but hardly at all in another. Production functions, especially economic ones, will also show temporal variations. These temporal variations are mostly responses to cost fluctuations which themselves respond either to inflationary cycles or to more immediate supply/demand situations. It is increasingly possible to manipulate production functions. Obviously economic functions undergo almost continuous fluctuations as costs and markets change, but increasingly physical functions can be manipulated, e.g. by building a pond, by fertilizing soils or by aerating water. Social manipulations occur by such factors as altering food preferences, as exemplified by the increasing popularity of take-away foods, or by the marketing of more value added prepared foods.

The rationale for most economic (commercial or business) activities is profit maximization or cost reduction. Enterprises will seek to manipulate production functions in such a way as to best secure the maximum return for the minimum outlay of time, capital or effort. For the rest of this study we must take profit maximization as our production rationale since failure to aim for this objective will mean that there is no point in seeking suitable locations for aquaculture or inland fisheries, i.e. with enough capital it is possible, through the manipulation of production functions, to produce almost anything anywhere.

From the spatial perspective there are two ways in which the optimizing of production functions can be viewed. Firstly, we can take the view that most farmers must take, i.e. “Given this piece of land which I own (or rent), what is the best way of using it?” The farmer must then consider how he can best organize and manage his choice of activities (usually crops or livestock) so as to achieve this usual goal of profit maximization. The second viewpoint, which is often used by industrialists and is increasingly being recognized as being of importance to all economic activities, is - “To produce commodity x, where best should production take place?” This question has long formed the core of spatial optimizing theory (e.g. Weber, 1909; Hoover, 1948; Loesch, 1954; Isard, 1956), and it is the fundamental question which forms the core of this present study.

We should not underestimate the importance of getting the spatial decision right. For many fish farmers this may be a once in a lifetime decision for, unlike normal farming practices where crop or livestock mixes can be regularly manipulated in order to achieve optimal combinations, where expensive fixed assets are required the location decision, based on optimizing production functions, is seldom repeatable. Clearly, the study of production functions per se, their spatial variability and the mixes of same is absolutely vital if correct spatial decisions are to be made.

Since the reader may be familiar with much of the factual information on production functions, the rest of this chapter will attempt to draw attention to their spatial variability aspects.

2.2 The Production Function Mix Controlling Aquaculture or Inland Fisheries

It has been shown (in section 1.2) that both aquaculture and inland fisheries are complex production activities. The range of controlling production functions will therefore be large and will vary greatly between and within regions, with different production systems, with different species being produced, etc. Table 2.1 shows typical production functions controlling freshwater aquaculture (FAO, 1988 with some additions), and Table 2.2 shows those functions controlling the mariculture of oysters (Cordell and Nolte, 1988b); further, very detailed lists can be obtained from Coche (1985) or Duncan (1985). Most of the functions shown in Table 2.1 are fixed (once only) considerations - there would also be a range of more detailed economic factors to consider.

The production functions themselves will be of varying importance. Though it is possible to identify universally important functions, it is impossible to state which functions are of which relative importance since this will vary not only objectively, i.e. between types of production units, size of units, types of system, etc., but also subjectively in the sense that some fish farmers would be in a better position than others to overcome perceived difficulties or problems.

Consideration of those spatially variable production functions which cause variations in the efficiency of fish production from area to area demands that a number of general factors are reviewed. Firstly, it will be apparent that it is not always possible to clearly differentiate between one function and another. For example - should water quality and water temperature be considered as two separate functions, when the former may be partly a function of the latter. Similarly, many climatic variables will be related to water temperatures.

Secondly, it would be possible to subdivide some production functions into narrower categories, e.g. markets for fish could be subdivided by type - perhaps “Wholesale” and “Retail”. These in turn could be subdivided into “Processors”, “Smokers”, “Freezers”, “Supermarkets”, etc. Clearly, any final choice of relevant production functions must depend on the exact circumstances of the study though anyone who is intrinsically involved in activities such as aquaculture or inland fisheries will recognize fairly definite functional sub-categories.

Thirdly, if we examine spatial variations in production functions at very different spatial scales, then differing functions may dominate. For instance, at a small scale (say an area 10km x 10km) considerations of relief, bathymetry, shelter and soils might be very pertinent site specific factors, but at this same scale considerations of evaporation rates, distances to main markets or access to extension services would be of only minor relevance. Clearly, at larger scales distance functions would be of far greater significance as would considerations of volume supplies of goods, the marginal utility of land, etc.

Fourthly, it may often be the case that the spatial disposition of any individual production function will alter, sometimes significantly, at an internal or national boundary. This would especially apply to economic functions. Thus Lindquist and Mikkola (1989) have shown how spatial variations in production functions between the four contiguous riparian countries around Lake Tanganyika, lead to completely different production pressures on the one lake.

There has been little research aimed at identifying the extent of economic penalties liable from poor locational choice, though it has recently been shown (Muir and Kapetsky, 1988) that financial penalties do accrue in locations where production function mixes are sub-optimal. The degree of this penalty would vary from enterprise to enterprise and therefore it is difficult to apply generalized models of production function mixes to existing locations.

Though there are many production functions which control aquaculture and inland fisheries in varying degrees, it will only be possible to examine some major functions in detail, i.e. chosen to illustrate the breadth of spatial considerations necessary, and some minor functions very briefly. For each major function we will endeavour to review :

1. The ways in which the function might control fish production.

2. The causes of production function variability.

3. The degree of possible manipulation of the function.

4. The degree of variability of the function at the macro and/or micro-scales.

5. Spatially related considerations that the intending entrepreneur, fishery planner or fishery manager should be concerned with.

2.3 The More Important Production Functions Controlling Aquaculture and Inland Fisheries

In this section it is intended that sufficient information is given such that any prospective fish producer would be aware of the essential elements or pitfalls in seeking optimum locations. It would then be in the interests of all intending fish producers to seek out more precise production function parameters before investing in any potential or intended enterprise.

2.3.1 Water Quality

Since each aquatic species will have its own optimum and extreme range of water qualitative parameters, it is impractical to do other than review the application of some of the more obvious qualitative factors, i.e. essentially those which can be readily monitored and measured. Table 2.3 briefly states the manner in which the water qualitative parameters determine the effect on aquatic growth and suggests how the parameter is influenced. Further details can be obtained from Alabaster and Lloyd (1980), Pickering (1981) or Payne (1986). Many other qualitative parameters could be defined but most would apply only to specific locations, e.g. pollutant sources from farm wastes and chemical applications; from mining seepages or heavy metal contamination; from industrial effluents or from urban areas (especially sewage outfalls).

The reasons why water quality varies over space can be conveniently divided into a) natural causes and b) human influences. Natural causes relate primarily to three major factors: precipitation, rock weathering and evaporation (Payne, 1986). Water of higher temperature is more prone to qualitative degrading, e.g. reduced oxygen levels, excessive nutrient accumulations, etc., and heavy rainfall may significantly increase turbidity through excessive run-off. Geology affects water quality by determining the rate of rock disintegration, and the rock type through (or over) which water may flow. In so doing it can accumulate dissolved minerals, such as calcium or sodium, or more concentrated minerals seeping from specific natural sources. Total ionic concentrations will generally progressively get higher further downstream and be higher during the dry season when the influence of ground water is greater. More turbulent patterns of water flow will enhance the oxygen content. Human influence on water quality is largely through unregulated activities which allow leakages or discharges from a diverse range of human activities, i.e. related to industry, mining, agriculture and urban sprawl.

Generally there are few chances of controlling or manipulating the larger scale negative water qualitative factors. Though some methods are practicable, e.g. clearing excessive water vegetation to adjust oxygen levels or by adding lime to increase alkalinity levels, in general the cost of continuously carrying out these procedures is prohibitive. It is only some highly intensive culture practices which can afford to moderate water quality, e.g. by water aeration to increase oxygen content when water temperatures are excessive. At the smaller scale effective pollution controls are certainly feasible, though this may depend upon the goodwill and financial situation of a third party. Larger scale pollution is certainly one of the major limiting factors to fishery expansion: this problem would usually need to be tackled collectively, probably via the use of statutory measures.

There is little comprehensive information on which to assess the extent to which poor water quality limits fish production on a world scale. Certainly Palm (1989) notes that it is uncommon for African water bodies to have fish growth inhibited by poor water quality. However, in other more developed parts of the world, the overall situation would certainly be deteriorating, largely due to rapid urban and industrial developments, to major forest clearance and to the increased use of agricultural chemicals. Indeed, in some of the more advanced developing countries the situation is already quite alarming, e.g. Menasveta (1987) has shown that, although the situation is now improving, mariculture output in Thailand declined from 148 369 tons in 1977 to 23 902 tons in 1982 due almost entirely to coastal pollution. Additionally, Lindquist and Mikkola (1989) have shown how even in a very large body of water such as Lake Tanganyika, the growth of pollution is evident. In many developed countries massive water clean-up efforts have been implemented but even here headway has not always been achieved. At the localized scale, water quality can be extremely variable, especially in the developed world. Table 2.4 summarizes those spatial considerations concerning water that should be of concern.

2.3.2 Water Temperature

Food intake, genetic constitution and water temperatures are the most important determinants of fish growth rates (e.g. Cooper, 1961). Since it is within the power of the fish farmer to control food intake and to select for genetic advantages, it will be apparent that actual water temperature will be a prime consideration in the selection of locations for fish culture, e.g. several authors (Allen, 1941; Weatherly and Lake, 1967; Bulleid, 1974) have shown that the time taken for trout to achieve marketable size can be greatly reduced if the temperature could be maintained within the optimum range.

Although all species have a preferred water temperature range, an optimum temperature range for maximum growth and a range of maximum and minimum tolerance, most can be acclimated to slightly different temperature regimes. Many species are well adjusted to seasonal fluctuations in the range e.g. most salmonids, whereas others, mainly tropical species, prefer homothermic conditions. In the optimal temperature range, metabolic rates will be high and thus there is the greatest likelihood of the fish eating more food than is necessary for maintenance alone - growth is then rapid as extra food is converted to flesh. In areas having higher water temperatures phytoplankton production may be greatly increased - this will greatly affect the maximum fish biomass potential in a water(Payne, 1986).

Water temperatures are determined by three groups of factors :

1. Meteorological, e.g. air temperature, intensity and duration of solar radiation, incidence of snow cover and surface wind velocity. Various authors (Macan, 1958; Ward, 1963; Cluis, 1972; Smith, 1975) have shown that air temperatures are the dominant factor controlling water temperature and, apart from sites where thermal modification by effluent discharge has occurred or where the river flow is composed largely of groundwater, the agreement between air and water values is sufficiently close to enable air temperatures to be used as a statistically reliable predictor of water temperatures (Smith and Lavis, 1975). This is demonstrated for several sites in Britain (Figure 2.1) and Balarin (1987) shows temperatures for two sites in Malawi (Figure 2.2). Clearly solar radiation will only be effective on the epilimnion (nearest to surface zone) of larger standing water bodies. Though water circulation may occur in these bodies, on average deeper water in the hypolimnion is markedly cooler. Standing water bodies usually exhibit a wider range of both diurnal and seasonal temperature variations.

   Figure 2.1 Relationship Between Air and Water Temperatures for Several Sites in Britain

   

2. Geophysical, e.g. depth of water, volume of water discharge, degree of water mixing, height of banks and temperature of inflows. The temperature of stream inflow, or of the basal water flow, is the second most important influence on water temperature. Rivers which contain a large groundwater component have a small temperature range. Thus chalk streams or spring fed headwaters of many rivers may appear abnormally hot in winter or conversely cool in summer (Rodda et al, 1976). As a downstream progression is made temperatures are more influenced by the temperature of water from tributaries than by basal flow or by heat exchange from atmospheric contact.

3. Human intervention, e.g. reservoirs and weirs, thermal effluent discharges and canopy alteration by logging or planting. As more of the world's rivers are being impounded, then temperature variations applicable to standing rather than flowing water bodies, become more common. Similarly, thermal effluents largely emanating from electricity generating plants are increasingly encountered. Though heated water in some rivers may be dissipated quickly, or simply form a heated layer on the surface, in others the total effect may be quite noticeable.

Figure 2.2 Mean Maximum, Minimum and Average Monthly Pond Water and Air Temperatures for Two Sites in Malawi

There is very little chance of manipulating water temperatures on a commercial basis, though some intensive (mainly research) facilities are experimenting with production units which are totally enclosed in a polythene housing. Use can be made of the artificially heated water, e.g. at some power station sites the higher water temperatures can greatly speed up growth rates for various species.

The prospective fish producer should be aware that, at the macro-scale, the degree of water temperature variations will not be great - they will be related largely to latitude, altitude, season and the extent of basal flow. Figure 2.3 shows how the degree of variability of water temperatures varies with latitude for various African lakes or rivers (Talling, 1969). At the micro-scale the degree of variability would be more affected by those human intervention factors previously mentioned plus the size of the water body, the nearness to the source of the stream and the prevalence of inflows of tributary water. However, marked micro-scale variations are relatively rare since water readily mixes or adapts to achieve a temperature equilibrium in sympathy with ambient air temperature.

Figure 2.3 The Annual Variation of Surface Temperatures in a Series of Fourteen African Lakes

The annual variation of surface temperature in a series of fourteen African lakes, arranged by latitude. Successive curves are displaced downwards and the common temperature scale (left) refers to differences only. Absolute values can be located using the single temperature marked on each curve.

(Reproduced by permission of Schweizerbart'sche Verlagsbuchhandlung from Talling,1969)

2.3.3 Water Quantity

When considering water availability, the fish producer must be concerned not only with the total quantity normally available and potentially accessible, but also with the variability of water flow. It is impossible to state that “x” quantity of water will be needed over “y” period of time since the amount needed will vary according to:

1. The total volume of fish being produced.

2. The density of stocking.

3. The age, size or the species of the fish or aquatic organism.

4. The temperature of the water.

5. The dissolved oxygen level of the water.

6. The fertility of the water.

7. The evaporation or seepage rates.

Generally, the larger the quantity of water available, the larger will be the size potential for individual fish growth and for total fish production.

The causes of water quantity variations at any particular location will be many. They will be a function of the temporal and spatial vagaries of the hydrological cycle at that location. Table 2.5 lists the main causes, noting whether they are primarily spatial variations or temporal variations (or both). The variability ratio of the water flow is of fundamental importance to aquaculture production. It represents the difference between a measure of high and low flows thus giving an indication of the consistency of flow expected. Areas having consistent flows will be less liable to flooding, will generally have less turbid water and the fish producer will be able to site gravity flow intakes with more certainty. Table 2.6 gives an indication of flow rate variability for nine selected hydrograph stations in England and Wales, and Figure 2.4 shows the frequency of flow variability for all 576 water gauging stations in England and Wales, both actually and cumulatively.

(a) All data was obtained from computer print-outs by the Water Data Unit at Reading.
(b) Excepting all of the 1940's.
(c) The Kielder Reservoir will prohibit the river drying up in the future.

Figure 2.4 Frequency of River Flow Variability Ratio for All Gauging Station in England and Wales

Wright (1974) has estimated that, in temperate areas having largely sedimentary rocks, geology and catchment size may account for over 90% of total flow variance, though in many mainly sub-tropical areas seasonality of rainfall will dominate. As a general rule water flowing entirely across, or which is mainly derived from, areas of permeable rock are likely to exhibit small variability ratios - typically in the region of 5 to 20. Conversely, streams which are dominantly on impermeable rock exhibit highly variable flow rates, with ratios of >200. This variability is frequently exacerbated by the coincidence of areas having high relief rainfall with areas having hard impermeable rock.

Under normal conditions, this function of water quantity can be readily controlled by the fish producer - FAO (1981) gives a detailed practical account.

Clearly, for inland fisheries, water quantity is seldom a problem in larger water bodies, with the notable exception of areas of inland drainage in arid, semi-arid or draught-prone regions, e.g. around the Aral Sea, Lake Eyre, Lake Chad, the Okavango swamp or the Nile marshlands of southern Sudan. For fisheries in reservoirs, or downstream from them, fluctuations in water area and level are important for their effect on biological production as well as the efficiency of fishing. Knowing the biology of the fishes and fishing techniques potentially provides the basis for managing flows to optimize fisheries, or at least to minimize losses.

For smaller standing bodies, the dry season evaporation potential is a limiting factor which is difficult to control. For producers utilizing flowing water, apart from periods of flow extreme, water quantity is readily adjusted by monk or sluice gates, and the land on flood plain or deltaic areas can frequently be modified (by excavation or impoundment) to provide a succession of ponds which can utilize large quantities of the parent river water.

Extremes of flow are of critical importance to aquaculture location. Excessive water flow, leading to flooding, can literally “wash” fish away and damage capital works, whilst periods of low flow usually coincide with periods of greatest water needs elsewhere, e.g for irrigation, and with enhanced evaporation. Here it is relevant to note that for some kinds of aquaculture it is desirable to achieve rapid water exchange, e.g. to increase oxygen throughput for salmonid production, whilst for other kinds it is desirable to limit water exchange rates, e.g to enable nutrient build-up which encourages plankton concentrations.

The actual variability of this production function over macro-scale space largely accords with the world's climatic zones, i.e. having regard to total precipitation, precipitation variability and evaporation rates. Almost all the low latitude tropical areas receive abundant convectional rain, though this has an element of seasonality related to changes in wind and pressure belts. Monsoon areas of S.E. Asia and higher coastal latitudes having prevailing westerly winds receive abundant rain water.

There will clearly be an overall tendency that - “The nearer the coast, the more water there will be available”. Obviously rivers increase their size in this way, rainfall is usually higher here; plus the land will be lower and frequently flatter which encourages marsh, lagoons, etc. Estuaries will be wider and there will exist a diverse range of potential mariculture environments. Conversely, inland areas are more frequently deficient in water. Areas having been affected by Pleistocene glaciation frequently provide abundant water quantity, both because glacial erosive action has resulted in exceptionally long coastlines, e.g. Norway, Southern Chile and western Canada, and because land areas have been deeply scoured so as to provide numerous lakes, e.g. Canada and Finland. Areas receiving past or present tectonic disturbances, frequently have large standing bodies of water where down-faulting has occurred, e.g. the African rift valley.

Since water is confined to channels or standing bodies in a far from ubiquitous manner, then the variability of water quantitatively at the micro-scale is very large. Additionally, nearly all the factors shown in Table 2.5 show micro-scale variations which cause further spatial or temporal variability. And further to this there are many areas where there is still uncertainty as to even the existence or not of potential water supplies (Anderson, 1989; Shimang, 1989; Vincke. 1990). These factors all underline the immense importance of selecting for this production function when seeking production locations - Table 2.7 summarizes individual selection criteria.

2.3.4 Transport Accessibility

Transport costs per se frequently represent only a small proportion of total fish production operating costs, e.g. for trout production in the U.K., Lewis (1984) found that they were typically 5% and for shrimp production in Malaysia, Muir and Kapetsky (1988) reported costs of only 0.3%. However, accessibility to transport is important in a number of non-quantifiable ways (Edwards, 1978; Kapetsky, 1989).

Firstly, transport can be seen in terms of relative access advantages, whereby cost savings are made by selecting sites which have advantages of good direct access from the water side (or pond bank) to the public highway system. Since fish products are highly perishable, they are nearly all transported by road. This is because of the flexibility of scheduling and to prevent break-of-shipment costs associated with alternative forms of transport. Road transport is especially important if large-scale production is envisaged since necessary inputs and outputs could well be very bulky. Occasionally, usually in more remote areas, water transport must be considered, i.e. if it is feasible, cheap and there are few alternatives. Relative access advantages help to reduce vehicle running costs, expedite inward and outward deliveries and reduce personnel time/travel costs.

Secondly, transport can be viewed in terms of a cost function associated with distance between the production site and the market place. If markets are at a distance then transport penalties would necessarily occur. It is difficult to give likely cost/distance curves because these would vary as a function of total distance, condition of the roads, mode of transport, vehicle type and size, quantity transported, form of fish product (iced, frozen, fresh), etc. It is clear that in this sense many locations, especially in developing countries, suffer severe disadvantages, e.g. Lindquist and Mikkola (1989) show the potential problems of distributing fish products from Lake Tanganyika within both Zaire and Tanzania, and in FAO (1975) it was noted that the lack of transport routes presented a huge impediment to fish production development.

This impediment of the lack of transport routes highlights a third aspect of transport accessibility, i.e. should transport be viewed in terms of a cost, a distance or a time function? We noted that it is difficult to postulate transport cost penalties in site selection. Though distance penalties are easier to calculate, they too might give an incomplete picture, i.e. if the transport was slow, if break-of-shipment, a ferry crossing or any similar hold-up was involved. It could be argued therefore that time taken for transportation should be the norm used for assessing transport penalties, though this method too has some drawbacks.

There are several other ways in which transport accessibility might control fish production, and ways in which transport costs themselves could be controlled. If a site is selected which involves the construction of an expensive transportation connection, then this could well make the site non-viable (Muir and Kapetsky, 1988). In these cases viability might be dependent upon who was prepared to pay for construction costs, e.g. public or private or even co-operative funding sources. Transport costs themselves can be ameliorated by various co-operative buying or selling ventures, or by buying in bulk quantities. In some areas transport cost differentials have been eliminated entirely by suppliers adopting universal pricing policies on their products, i.e. irrespective of distance shipped.

The degree of spatial variability of this production function is strongly related to population density. Areas having a dense population are likely to have a correspondingly dense road network (or railways) and conversely rural areas will have few roads. Variations on this pattern are worth noting. Some rural areas have quite a high density of minor roads, i.e. where agriculture is important. There will frequently be a network of main roads which cross less densely populated areas in connecting major population centers. Fortunately for fish producers, many routeways choose to follow waterways since the formation of the waterways themselves have created relatively flat routes through areas which otherwise may be hilly. Also, access to water supplies for various purposes has always been a necessity. At the macro-scale there will be huge variations in transport accessibility, i.e. since it is a linear function, and there will also be qualitative variations in transportation provision. Table 2.8 summarizes the main transport criteria which need to be considered.

2.3.5 Market Accessibility

In this section we will not be concerned with the form in which aquatic products are marketed. even though this form may influence markets served, e.g. in tropical areas fish may need to be disposed of quickly (usually locally) unless smoking, sun-drying or ice facilities are available to allow more distant markets to be served. There are many possible combinations and permutations in marketing linkages for the disposal of fish. Figure 2.5 shows a schematic diagram for the marketing of trout in England and Wales, and Balarin (1987) shows a diagram relating to markets in Malawi (Figure 2.6). Different flow diagrams would be appropriate in differing countries or regions, for different species, for different operational scales or at different temporal periods.

Figure 2.5 Flow Diagram Showing Table Trout Marketing Networks in England and Wales

Figure 2.6 The Fish Marketing System in Malawi

In most of the more developed countries the largest market outlets would be the wholesale sector. This might comprise of coastal or inland merchants, many of whom were associated with particular sources, e.g. a local fishing fleet, or with different aquatic species. There are often many merchants at each market center. Most fish handled would be fresh and purchase would be by day-to-day negotiations or by auction. Other wholesale outlets include fish processors, who would buy fish for freezing, canning or smoking. They often operate on a contract purchase basis and they might supply large retailers using particular brand names or they could be processing for export.

In order to achieve higher returns, many fish producers would endeavour to dispose of a high proportion of their production to markets which are “closest” to the final purchaser. Highest returns can be achieved by selling via the “farm gate” or at the “dock side”, i.e. direct to the public. Lewis (1984) has shown that trade can be obtained, often with considerable success, if there is a busy routeway passing by the production site (having appropriate parking). “Dock side” sales are only really possible where regular markets can be guaranteed at sites having public access. The next best return might be achieved by operating a fish delivery round or by selling to local hotels, restaurants, peddlers or caterers, i.e. outlets whose own “mark-ups” might be very high. Some larger producers have contracts to supply supermarkets direct and some might supply direct to fishmongers or other retailers.

An increasing volume of fish is being produced for live sale to supply various stocking or restocking waters. The natural reproduction rates of most fish species in many developed countries are no longer able to sustain the requirements for angling purposes, especially among the salmonid species. Periodic stocking to enhance food fisheries is common practice in much of Asia, particularly in China. Water pollution has totally eliminated aquatic stocks over large tracts of northern Europe and parts of north America, and where clean-up operations have been successful restocking has followed. This growth in serving the needs of leisure fishery is likely to slowly develop in many other parts of the world.

If fish production is operating in a free market economy, then there exists the potential to exert varying degrees of control over the market, e.g. by securing cost advantages through bulk input buying or by horizontal or vertical integration practices. The extent of these economic manipulations will vary according to various statutory controls, the size of the economy, the perceived investment rewards and the state of the economic climate. Because, at the present time, fish production is such a varied and complex economic activity, i.e. involving a risk element, much specialized knowledge and a large degree of commitment by the entrepreneur, then markets are not being dominated by large-scale fish producers.

Following the opening or closing of any fish processing market outlet, there can be quite dramatic changes in the advantages of differing locations for fish production. In a study of the spatial diffusion of catfish farming in the lower Mississippi area of the U.S.A., Meaden (1978) showed that the “center of gravity” of the industry changed quite noticeably during the 1960s and 70s, i.e. in response to the opening of new processing plants. The existence of any processing works is likely to be the focus of attention for many inland water fisheries. Transport routes or facilities can also determine the feasibility of using specific markets, especially in areas where fresh fish are demanded and/or where processing facilities are absent (Pathak, 1989).

Marketing information is now becoming more generally available, e.g. Satia (1989) shows how the fortnightly INFOPECHE bulletins, the bi-monthly INFOFISH INTERNATIONAL trade magazine and the GLOBEFISH (FAO) computerized data base are likely to have an effect on fish production and marketing strategies throughout Africa. A similar effect will be felt by the growth of specialized marketing organizations in other areas.

The degree of spatial variability of this production function varies quite significantly and can best be reviewed according to market outlet. Wholesale markets in most countries are distributed on a macro-scale. As indicated earlier they are mainly sited in larger cities or towns and in coastal or lake side fishing ports. Some countries may only have one or two wholesale centers, whilst traditional fishing countries might have many. As some wholesale centers, or particular merchants, specialize or deal in certain market sectors (or species) only, any prospective fish producer would need to be very certain where his nearest wholesale outlet was, i.e. if an increase in fish production materialize then an increasing proportion of output is likely to be disposed of via this outlet.

At a different marketing level, catering, hotel, restaurant or retail outlets would usually show a more dispersed spatial distribution, though clearly there would be many exceptions in remoter areas. Indeed in many countries, hotel or restaurant outlets would be confined to a few larger cities or resort areas. This market level will have a limited sales potential which is usually related to variances in disposable income levels within a region or country, or to variations in dietary preferences. It is likely that isolated market sites could only support the development of small fish farms since transport or storage costs would prevent further growth.

The spatial disposition of “farm gate” or “dock side” markets will clearly show no specific pattern apart from adjacency to urban areas, busy roads or fishing ports. Again the intended fish producer would be advised to investigate the possibilities for these outlets at individual locations.

For reasons previously given, the spatial variability of the stocking/restocking market sector will clearly relate to those waters which need extra stock. This will often accord with a combination of:

1. The degree of urbanization, reflecting a measure of the number of people likely to be fishing (anglers).

2. The degree of industrial reclamation, reflecting likely attempts to clear up water pollution caused largely by older industries which have been made obsolete, or reflecting areas of newer and cleaner industrial growth.

3. For recreational fisheries, the degree of affluence, reflecting the likelihood of having the ability to pay for recreation and to have the time necessary to participate.

4. For commercial fisheries, the perceived increased returns due to stocking.

In some smaller developed countries, virtually all of the country could satisfy these criteria, so restocking is very widespread. In larger developed countries which might have areas of sparse settlement, restocking markets would be more spatially variable. In most developing countries the amount of restocking is presently very limited, indeed in most cases unnecessary or non-existent, so the potential of this market sector would need careful investigation; Table 2.9 summarizes the main considerations pertaining to site selection relative to the markets.

2.4 Considerations of Some Other Spatially Variable Production Functions

Having shown in some detail the manner in which the major production functions control fish production, and how these functions might vary spatially, it will be apparent that the fact of spatial variability is of great importance in any decision concerning optimizing the production location. The intending fish producer, or any concerned person, will also be aware of the types of practical considerations which spatial analysts must make. We are now able to look briefly at some of the other spatially variable functions which control the complex enterprises of aquaculture and inland fisheries. Here again we will only concentrate on the ways in which the function might vary spatially.

2.4.1 Climatic Factors

Around the world there are well identified climatic zones and these have been intensively documented in most atlases and elsewhere. The main climatic problems to aquaculture or inland fisheries are associated with “unreliable” climatic areas. At the macro-scale these mostly accord with:

1. Areas where the timing of seasonal rainfall may be unreliable.

2. Inland areas in the mid-latitudes where precipitation totals are unpredictable.

3. Those usually warm areas which occasionally experience freak frosts.

4. Areas having tropical cyclones, hurricanes, etc.

Micro-climatic variations would seldom cause problems to fish producers.

Prospective fish producers would need to be very certain of extremes in the temperature regime and be aware of the temperature tolerances of any species being produced. Excessive cold, leading to snow or ice, can cause access problems to production sites and severely impede the management of any waters (Cordell and Nolte, 1988b). Excessively high temperatures will lead to evaporation problems if water replenishment is difficult. Very heavy rainfall can, of course, lead to flooding, but equally importantly, both Sage (1980) and Kapetsky (1989) demonstrate that this is a major factor in lowering the p^H of waters to sometimes dangerous levels. Heavy rain can also severely disrupt transport access. A lack of precipitation may lead to problems of water shortages which have been discussed in section 2.4.3.

2.4.2 Agglomeration

It is difficult to give a formal definition of what constitutes an agglomeration of fish producers, though it usually implies that some degree of cooperation is being purposefully obtained arising from the adjacency of production units. The size of agglomerations might vary greatly. The existence of agglomerations will give an excellent indicator of the suitability for fish production, since they are only likely to occur in areas which show outstanding production function mixes. The potential advantages to be obtained within agglomerations are:

1. The opportunity for bulk buying.

2. Already established potential markets for the newcomer.

3. The sharing of transport costs, particular pieces of equipment or facilities or knowledge.

4. The likelihood of a nearby processing plant being established.

5. The opportunity for fish transfers to meet particular customer orders.

The prospective producer would need to find out whether agglomeration advantages were actually functioning as a positive factor. It might in fact be better to site at a distance from other producers, i.e. even if sub-optimal conditions existed, in order to have less competition or to supply a previously untapped market. It might also be the case that the area has sufficient producers, i.e. from the considerations of water quality and quantity, markets saturation, potential site availability, disease risks, etc.

2.4.3 Soils

Soils will vary greatly at the macro-scale because soils result from complex physical interactions which themselves take place in areas having different topographic, geologic, climatic, vegetation and human influences (FAO, 1985). Geology often gives a clue to soil type suitability since it indicates the parent rock, and topographic studies help since they reveal steeper slopes and hill tops where soils are invariably adverse. These studies will also indicate river flood plains, deltas, estuarial locations and coastal plains where promising alluvial soils are likely to occur. Kapetsky (1989) and others have illustrated how soils may vary greatly at the micro-scale. This makes it essential that the intending fish producer be aware of the particular properties of soil as they apply to the suitability for aquaculture. The main properties are shown in Table 2.10, though FAO (1985) should be consulted for further details. The producer should also be aware that soil types may vary with depth, and that it is useful to know where accessible soils maps are. If soils are deficient at particular locations, then there are means of controlling or manipulating these though they too will exhibit spatial variations, e.g. by bringing in quantities of a suitable soil, by liming, by adding organic materials, by lining ponds with plastic sheeting, by puddling, by adding fertilizer or even by constructing alternative production systems.

(L) = Laboratory test
¹ Particle-size analysis on total sample: particles are then separated by size into twogroups: the tests for “sand” are performed on the group with particle size larger than0 075 mm and the tests for “sills” are performed on the group with liner particles.
² A faint odour may be intensilied by slightly hcating the sample.
³ See Section 60

2.4.4 Relief (Topography)

Though this function may vary greatly at both the macro and micro scales, over significant proportions of most regions relief will not pose a barrier to fish production. The main problem areas will be of two types:

1. Excessively flat areas. Here flooding, plus the inability to provide for a gravity flow of water and poor drainage, may be the major problems. Additionally, sluggish water flows found in flat areas are associated with low dissolved oxygen levels, high summer water temperatures and, near coastal areas, saline intrusions.

2. Areas having steep slopes. Here enterprise sites may be difficult to locate and develop, and any adjacent flat waterside area would be prone to spate flooding conditions. Areas of steep relief may also cause problems with other production functions such as transport accessibility, isolation from markets, high rainfall and run-off, etc.

Underwater relief, i.e. bathymetry, can also be important. For example, in shallow areas water circulation is restricted under cage installations. In deep waters, anchoring costs for raft culture can be high.

2.4.5 Land costs (or Rent)

It is impossible to state actual land costs or rents for any general location since constant spatial and temporal adjustments occur. However, land cost variability will usually be a function of three factors:

1. The supply of, and demand for, land - as reflected by population density.
2. A perceived regional economic potential - as reflected in commercial investment.
3. Land quality - as reflected by land use capability or perhaps soil quality.

There would generally exist a strong positive relationship both between land costs and each of these three factors, and between each of the three factors per se, e.g. areas would tend to be densely populated precisely because land quality was good or because perceived economic potential was high. At the micro scale, any of the three factors could prove dominant in a particular location and at this scale there would be large cost variations over quite short distances. At the macro-scale the land cost surface would be much more smoothed. Actual land costs or rent will vary depending on factors such as whether the land is owned or rented, what its alternative use value might be, how much land is needed, etc. Additionally, most waterside land can command premium prices because of the huge range of competing uses.

Although the initial cost of land may be a major item of expenditure, and it may therefore greatly affect the location decision, it can be insignificant when costed over a long period. The fish producer should be aware that, since a hypothetical location-economic rent curve would show that future income expected from fish farming, rather than most other types of farming or rural land use, would be advantageous, he could afford to pay more per unit of land.

2.4.6 Availability of Credit (or Capital)

In developed countries there are few spatial variations in access to credit. Limitations are related to personal factors concerning the entrepreneur, plus perceptions of aquaculture as a risky enterprise, whose costings may be hard to assess because of output quantity and pricing uncertainties, and because of temporal delays in achieving any returns. With dwindling stocks being available to inland fisheries, it is likely that capital for future developments here will be hard to obtain. In less developed countries credit or capital may be very difficult to obtain and very high interest rates are common. Part of the reason is that aquaculture is relatively new. Consequently, bank loan officers are not well informed on techniques of assessing the economic viability of fish farming. The likelihood of obtaining credit would vary spatially, with urban oriented locations finding it easier.

The prospective fish producer should seek funding from not only traditional commercial/banking sources but also from various organizations specially established to help with financing development projects, e.g. EEC. loan funds; various UN funding programmes; government funds to help diversify local economies; regional development schemes; fishery development organizations and various development agencies. Balarin (1987) lists nearly twenty different agencies, etc., who were providing external assistance for fishery projects within Malawi alone. He also shows how financial assistance may be given for various facets related to fish production, e.g. personnel training, demonstration farms, R. & D. projects, etc. Other credit sources are cited in various FAO documents e.g. Baluyut (1988); Girin (1989); Pathak (1989); Satia (1989). The potential producer would do well to seek both documented sources which reveal funding availability and advice about the appropriateness of any desired investment. Since costs of capital items (plus labour) may vary spatially (and indeed non-spatially), then several alternative costing estimates should be sought and some idea gained as to the relative cost advantages of different areas.

2.4.7 Availability of Fertilizer and Agricultural By-Product Inputs

The natural fertility of water varies greatly as a function of the complex interrelationships between a number of environmental factors, e.g. available nutrients in the soil, strength of solar radiation, water turbidity, depth of water, etc. Maar et.al.(1966) give a simple explanation of this. Variations in natural fertility could well be discussed as a separate production function but, since it is readily manipulated and since some of the other functions we have reviewed may control natural fertility, we will look only at the availability of fertilizer inputs per se.

Fertilizer inputs are frequently not even considered in cooler areas because plankton growth rates will be much slower meaning that better (or more cost-effective) results can be achieved by supplementary feeding (see Figure 2.7). This logic may also apply in warm water areas where intensive, highly capitalized production systems operate.

Figure 2.7 Relationship of Phytoplankton Production During the Growing Season to Latitude (from Brylinsky and Mann, 1978)

The distance from farms having livestock is obviously of importance to the spatial availability of fertilizer, i.e. since animal manures are optimal, especially poultry manure, and therefore the incidence of these sources will need researching for any specific location or area. There will be large cost variations in acquiring fertilizers because of variations in its quality, the amounts available seasonally and transport cost efficiencies. In many areas, e.g. much of Africa and southern Asia, there is a distinct lack of fertilizer even for more normal agricultural use (Satia, 1989).

The balance of fertilizer inputs must be carefully considered, i.e. both the economic balance, involving cost of inputs v. extra outputs obtained, and more importantly the environmental balance such that maximum productivity is obtained but not too much fertilizer is added which will lead to water quality deterioration. This balance (in both senses) will vary temporally and spatially. The type of production system will influence the demand for fertilizer, e.g. where water exchange rates are high then fertilizer inputs will be less effective, as will inputs made to very large bodies of water. The fish producer might wish to consider the viability of polycultural systems as a means of supplying fertilizers, e.g. by combining ducks and fish, or by integrating crop rotations with fish production in dyked fields so that residues from the cropping cycles help fertilization. If natural fertilizers are unavailable the producer might need to consider the economic viability and spatial availability of inorganic fertilizers or supplementary feeds. Palm (1989) gives an excellent summary of variations in fertilizer needs.

2.4.8 Availability of Extension Services

The actual provision of this function, in one form or another, exists in most countries but the number of extension workers would closely correlate with the wealth of a country, or with its interests in fish production, e.g. Baluyut (1989) reports a well developed network throughout much of eastern Asia, though with insufficient workers in the poorer areas of Laos and Vietnam, and Satia (1989) notes that extension services are insufficient in most of sub-Saharan Africa. This lack of extension services in most developing countries means that the availability of this function is spatially poor and probably “patchy.” A major problem for areas having few extension workers has been their lack of spatial mobility, though this is now being rectified in many countries. In developed countries this function could be looked upon as being almost ubiquitous, i.e. in the sense that they would nearly always be on a fairly short service call. The provision of extension services in ten major countries has been well documented in Chakroff (1981).

Prospective producers would do well to not only familiarize themselves with formal extension service provision in selected areas, including likely temporal delays in receiving advice, as well as the quality of the in formation being provided, but also to seek out alternative arrangements regarding sources of aid and information. This might be from neighbouring producers; from farmers unions; from veterinary sources; from trade journals or magazines; from “Peace Corps” workers in many countries or from aid agency officials, etc. Some degree of self education in fish production methods and problems would appear to be quite vital, especially in those areas which are likely to be in temporal isolation from the extension service.

2.4.9 Current Inland Fishery Levels

It is difficult to work out exact levels of fishing which may be possible, especially in tropical areas, because this depends on the natural fish biomass production rates, which in turn depend on a large number of variables (Payne, 1986). The levels of potential fish procurement will also vary spatially depending upon the type of fishery environment - lake, river, estuary, etc. Though some fisheries have possibly not yet reached their maximum sustainable yields, e.g. many river fisheries in Latin America and Africa could support higher fishing levels, with rapid population increases in these areas this is increasingly rare. In much of the rest of the world, current fishing levels are already too high and most countries are adopting restrictive measures in order to conserve stocks.

Controls over fishery levels can be realized in several ways:

1. Statutory quotas for fishermen, boats, species, etc., can be laid down.

2. A closed season can be stipulated.

3. A minimum fish size per species can be ordered.

4. The number of lines, nets, boats or fishermen can be limited by licence or taxes.

5. The size of the net mesh can be adjusted.

6. Given areas of water can be excluded from fishing for various time lengths.

7. Natural stocks can be supplemented by fish “ranching”.

Many of these measures may be very difficult to regulate, especially when the fishing effort is distributed around a large water body. In practice, much fishery regulation is accomplished by traditional authority. Central governments may have laws on the books, but often the political will and personnel are lacking for enforcement.

2.5 Other Production Functions Showing Some Spatial Variability

Having now examined a large number of production functions in varying degrees of detail, the prospective fish producer or land holder will at least be conscious that the simple fact of having some accessible water nearby may not be good grounds to embark upon fish production. Likewise, the fishery planner will know that a variety of criteria will have to be taken into account to properly identify the areas with the best prospects for success in fishery development. In any case we have only reviewed some of the major functions which vary spatially. Table 2.11 shows a large number of other functions which vary in this way and which might also need to be considered before a suitable site could be selected.