Practical designs of semi-intensive aquaculture systems integrated with livestock are based on a range of factors. An understanding of the dynamics of aquatic ecosystems is required to appreciate the mechanisms by which fish grow in ponds receiving livestock wastes, and the constraints to their production. Maintaining productivity of natural food growing in the fertilized pond and an environment conducive to fish survival and growth have to be considered. The principles of fertilization, well-established in terrestrial agriculture, have only recently been adequately researched but the three dimensional aquatic environment offers a range of options still to be thoroughly explored.
Intensification of fertilized systems through use of supplementary feeds is a well-known strategy in animal husbandry but, again, is unrefined for aquatic systems.
The characteristics of livestock wastes themselves are also frequently unreported and unremarked but are critical to the design and management of integrated systems. How wastes are used in fish culture systems also affects efficiency and productivity of the system as explored below.
Schaeperclaus (1933) first noted that static water ponds act as both the stall and the pasture for fish capable of exploiting the web of natural feed stimulated by fertilization. In contrast to terrestrial systems managed for herbivores, maintaining a balance between feeding and living environments is both more complex and critical. Whilst the concept of carrying capacity in terms of number, or biomass, of fish or livestock that can grow per unit area is similar, the necessity to maintain adequate water quality, especially dissolved oxygen, is critical in aquatic systems.
Ponds with optimal water quality but little food will be unproductive. Equally, food-rich ponds will under-perform or risk complete loss levels of the different feed organisms and fish, are excessive and use up dissolved oxygen (DO). Both sanitary engineers and aquaculture scientists have quantified how wastes stimulate the productivity of ponds through nutrient enrichment and how water quality changes rapidly as a result. The diurnal cycle of DO that occurs in unfertilized systems becomes extreme as loading of nutrients increases. These swings are mainly caused by photosynthetic production and respiratory uptake of DO by dense blooms of phytoplankton that develop after nutrient loading. Clearly, the key aspects of nutrient management are to ensure that DO, and natural food levels remain adequate and in balance throughout fish growth.
Many factors can affect the management of waste-fed ponds, which may have their origin at macro-, local as well as micro- or pond level. Climatic and cultural factors can affect the potential for integration of livestock and fish production at macro-level (Table 5.1). National policies such as those relating to fertilizer and feed manufacture and import, can affect availability and cost of nutrients at local level, supporting or undermining attempts to intensify livestock and fish culture.
TABLE 5.1
Factors affecting use of animal wastes in ponds
|
Primary factors |
Secondary factors |
Ameliorating factors |
Macro-level |
|
|
|
Local level |
|
|
|
Pond level |
|
|
|
Local resource levels are also affected by both physical e.g. soils, topography, water availability, and social and economic factors that can foster or constrain the development of integrated practices. At the pond or plot level, design and management of fish culture and associated animal husbandry have to balance resource constraints and opportunities.
Stable and high water temperatures and sunlight ensure year-round growth of both fish and their natural feeds. The tropics, in which average monthly water temperatures remain above 18°C, are ideal for culturing fish using livestock waste as inputs, although it is also practiced in sub-tropical and temperate climates during warmer periods of the year.
Manures can be used fresh, or after processing, to enhance natural food production in sun-lit ponds as much of the nutrient content of feed given to livestock is voided as excretory and faecal waste. Although some nutrition may be derived directly from the waste, high-protein natural feed produced on the nutrients released from the wastes mainly in the form of plankton is more important. Fish feeding low in the food web, e.g. carps and tilapias, benefit most from this type of management since they can utilize plankton, benthic and detrital food organisms effectively.
The nutrients contained in organic wastes stimulate a range of natural food organisms that may be suspended in the water column, attached to substrates, or within the sediments. Sediment-water interfaces are key feeding niches for some fish species. The availability and value of phytoplankton and bacteria, the dominant organisms in most waste fed ponds, are linked to specialized feeding habits of the fish. A variety of carps and tilapias can grow rapidly on such natural feeds alone, employing strategies similar to grazing (filter feeding) and browsing (ingestion of attached periphyton). A detritus-feeding niche, in which a large amount of sediment is ingested, may also be important. Most fish are opportunistically inclined to consume macro-invertebrates of various types, molluscs, insects, polychaete and oligochaete worms and crustaceans.
Several factors affect the level of waste loading and standing stock of fish that can be supported. The species of fish raised is important both from a perspective of varying feeding niche and sensitivity to water quality, especially DO in static water ponds without mechanical aeration. Carps are limited to standing stocks of <3 tonnes ha-1 whereas tilapias may be harvested at standing stocks of over 5 tonnes ha-1. Pond design, depth, shape and position also contribute to maintaining water quality since these factors affect exposure to wind, sunlight and run-off from adjacent land.
The carrying capacity of the fish pond, or maximum biomass of fish that can raised in a pond of given feed and water quality, can be enhanced by increasing fish density and using feed to supplement natural food. Supplementary feeds are generally relatively high in energy and low in protein to balance protein-rich natural food.
Water quality, particularly the level of early morning DO, limits the amount of wastes that can be used. Under tropical conditions, input levels in excess of 100 kg dry matter (DM) ha-1 d-1 often overload' the system over a typical fish culture cycle (4-8 months), causing early morning deficits of DO (Figure 15). Balancing the production of wastes and the requirement of the fishpond is a key aspect of management (Colman and Edwards, 1987; Knud Hansen, 1998). An understanding of the livestock densities required to provide optimal loadings of nutrients for given areas of fish ponds is key to the design of manure-fed systems.
The quality of livestock wastes used in fish culture varies greatly. The direct feeding value of manure is poor because of low levels of metabolisable energy and digestible protein (Wohlfarth and Schroeder, 1979). Any spilled livestock feed mixed with the manure will increase direct feed value. Nutrient composition is a useful guide to the value of the waste as a fertilizer, especially levels of total N and P; levels of nutrient density as expressed as a ratio of these major nutrients with C (C:N:P) is also a good indicator. However, these data may tell little about the availability or release of nutrients that can be taken up by the phytoplankton. The rate of release of soluble N and P is useful for predicting manure quality. A range of other benefits from use of animal manures in fish production have also been identified (Box 5.A).
BOX 5.A Benefits of animal manures in pond culture Manure:
Source: Modified after Knud Hansen (1998) |
High fish production in fertilized ponds is mainly attributed to stimulation of autotrophic production, that is the growth of phytoplankton that filter-feeding fish can use directly as feed and which form a major proportion of the detritus in the pond. Phytoplankton is also the main source of DO in the pond. Both living and detrital phytoplankton are protein-rich and are the basis of the food web that can support the growth of a range of herbivorous and omnivorous fish. Technical and economic considerations determine how fish ponds are best fertilized and managed.
FIGURE 15
Mean dissolved oxygen (DO) in mg l-1 at dawn for ponds (200 m2) receiving different levels of manure loading. Error bars show standard deviations.
Source: AIT (1986)
Nutrients
Fertilization aims to supply the optimal level of nutrients to stimulate phytoplankton production. Practically these nutrients are limited to N, P and C; other nutrients are required only in trace amounts and supplied by the fertilizer, supplementary feeding and the natural environment. Analysis of the nutrient content of healthy phytoplankton indicates the level of requirement which is at a ratio of approximately 50:10:1(C:N:P). The N:P ratio can vary from <1.5:1 in N deficient algae to >15:1 in P-deficient algae. Carbon is often ignored but is required in the greatest amounts. Waste-fed ponds are seldom C limited because bacterial breakdown produces carbon dioxide but inorganically fertilized ponds with low alkalinity often require lime to ensure sufficient C. Molluscs can also drastically reduce C levels through removal of carbonate for shell growth.
Selecting fertilizers and fertilization rates to meet the nutritional needs of phytoplakton or algae has a history of confusion that recent research has clarified. Earlier studies, often using low densities of fish or stocking non-filter feeding fish, failed to show a clear relationship between N and yields. Much of this research was based on temperate lakes in which N fixation was important in supporting the relatively low fish productivity. It is now known that in warmwater freshwater ponds, N is usually the most limiting nutrient and that high levels are required to optimize yields of phytophagous fish such as Nile tilapia. Available P can also easily limit algal productivity as it is rapidly fixed by various cations (Fe2+, Ca2+, Mg2+ and Al2+), and bound up in the sediments. Sediments with high clay content or containing acid-sulphate soils are particularly greedy. Fertilizer management should consider previous P inputs or pond history; ponds used for a prolonged period and receiving nutrient inputs will require relatively less fertilizer.
This summary is based on the work of: Goldman (1979); Colman and Edwards (1987); Knud-Hansen (1998); Shrestha and Lin (1996); Knud-Hansen et al. (1991).
Fixed rates
The use of fixed fertilization rates as guides to farmers is based on experimentation at relatively limited sites and cannot be simply extrapolated to ponds everywhere. Moreover, most research has focused on a single phytophagous fish, the Nile tilapia. A range of factors, including livestock and waste management systems considered below, may affect fertilization response. Differences in response by individual ponds even located only a few metres apart indicate that fixed rates should be used only as a guide to avoid over or under-fertilization. Farmers can use general guidelines, if presented in an appropriate way, and adapt them to meet their individual requirements. Indeed, farmers in China and India learned to manage fertilized fishponds before the principles were elucidated by scientists. Clear explanations of how nutrients work in the pond ecosystem are particularly valuable, as opposed to recipes.
There is generally concern among extension staff that excessive fertilizer levels, particularly of N fertilizers, can result in fish mortalities. Generally this is more of a danger if fixed rates are extended. A typical scenario is the effect of such overloading during periods of low temperature or low light conditions when the capacity of the pond ecosystem to absorb and utilize them is lower than normal. When conditions return to normal nutrients can then be present in surplus quantities causing dangerous levels (see Box 5.B) or unstable phytoplankton blooms that can crash causing sudden declines in DO and elevated levels of ammonia.
Animal wastes with unbalanced nutrient levels, typically high C:N ratios and N:P ratios well below optimal, are often compensated for by increasing manuring rate. This can lead to reduced DO levels and fish growth (Box 5.C).
Detritus and periphyton
The availability and usefulness of very small micro-algae (nannoplankton; diameter < 10 mm) that tend to dominate waste-fed ponds and the diets of filter-feeding fish has been the subject of much research. Mechanisms that describe their ingestion e.g. Northcott and Beveridge (1988), digestion (Moriarty, 1987) and avoidance of toxins which they sometimes contain (Beveridge et al., 1993) have been given.
The relative contributions of autotrophic and heterotrophic pathways to the productivity of fertilized fish ponds were once contested (Schroeder, 1977; Colman and Edwards, 1987). Most now conclude that algal-derived detritus, whether suspended, attached or settled is usually the dominant and most nutritionally valuable form of detritus available, even in waste-fed ponds. It supports a huge range of zooplankton, substrate dwelling micro-invertebrates and fish directly. Non-filter feeding fish can still derive adequate nutrition in waste-fed ponds by exploiting these feeding niches but the density at which their growth can be supported, the carrying capacity, is correspondingly lower than for filter feeding species.
BOX 5.B Fixed fertilization rates defined by experimentation N and P loadings of 4 kg. ha-1d-1 and 2 kg. ha-1 d-1, respectively, were defined as optimal for tilapia monocultures in a series of experiment using both organic and inorganic fertilizers at the Asian Institute of Technology and other research stations1. Harvested yields of 4-5 tonnes. Ha-1 of 200-250g fish in around 6 months are possible if 1 fish. M-2 is stocked. Higher stocking densities (up to 5 fish. m-2) can increase net fish yields further. If smaller fish are acceptable and multiple stocking and harvest is practiced, net extrapolated yields of 12 tonnes. ha-1 year-1 are possible. Inorganic N fertilization in excess of these levels resulted in:
Optimal levels of N fertilization are already about five times greater on an annual area basis than the maximum national average in Asia for land-based agriculture. P fertilization needs may be lower than recommended if the sites where the P is adsorbed in pond sediments are filled. This may occur through:
1 The Pond Dynamics/Aquaculture Collaborative Research Support Program (PD/A CRSP) supported by USAID have undertaken research in Honduras, Rwanda and the Philippines and Thailand. |
Fish that can graze aufwuchs, or attached periphyton, are less well supported in conventional fertilized ponds as substrate is limited. Recent studies indicate that increased substrate availability can improve fish yields, especially if the substrate in enriched with nutrients by soaking in manures. Periphyton-based aquaculture may be of particular benefit when size/scale of culture unit, or cost of nutrients preclude conventional fertilization. Significant constraints include the availability and cost of substrate in rural areas where biomass is heavily exploited for fuel and other uses.
Frequency of fertilization
Fertilization, especially if fertilizers require transport to the pond, can be laborious and often result in infrequent application of bulky animal wastes. In contrast, strategic livestock management and system design can result in substantial reductions of both costs for livestock waste removal and fertilization of fish ponds (5.2.3). Integrated systems, in which wastes continuously enter the fish pond and feed the algae probably result in optimal algae productivity since nutrient depletion is least likely to occur (Knud Hansen, 1998). More occasional, heavier loadings are more likely to shock the pond system, leading to explosive bacterial and algal growth and low early morning DO.
BOX 5.C High loadings of ruminant manure Ruminant manures from grazing buffalo had a C: N ratio of 26 and a N and P contents of 1.4 percent and 0.2 percent DM, respectively. To achieve a loading of 4 kg N ha-1d-1, 300 kg DM. ha-1 d-1 of manure had to be loaded. Dissolved oxygen quickly declined to deleterious levels through a combination of:
Source: Edwards et al. (1994); Shevgoor et al. (1994) |
If fertilization adheres to fixed schedules, unbalanced nutrient levels may occur that do not meet the needs of dynamic and unpredictable alga populations; algae in fish ponds are prone to species change or succession that impact on nutritional and water quality. During periods of continuous rainfall, when static water ponds may overflow, the demand for nutrients may be higher and more frequent than dry periods for example. Thermal stratification also occurs, especially in deeper ponds that result in elevated nutrients levels (Szyper and Liin, 1990). After breakdown during heavy rain or wind, nutrients that have accumulated in the lower levels of the water and sediments are released, exceeding the ponds capacity.
Supplementary feeding has been a rather neglected part of aquaculture science, perhaps due to the intensive nature of most modern developed country aquaculture and its reliance on carnivorous fish species. However, the intensification of traditional aquaculture in Asia is encouraging the use of supplements (Yakupitiyage, 1993; De Silva, 1993).
Supplementary feeding of a fertilized pond can both increase the fish yield and reduce the time to harvest but understanding how supplements work compared to complete feeds is critical. The provision of supplementary feed is required if fish of large individual size are required. Thus, although Nile tilapia of between 200-250g can be produced on fertilizer alone within 5 months, individual growth thereafter is slow as natural feed alone does not meet optimal growth requirements (Edwards et al., 2000).
Supplements are generally used commercially if the use of complete feeds is uneconomic. Both scientists and farmers are attempting to optimize technical and economic performance by developing feeding strategies. These require knowledge of local resource and market opportunities.
Supplementary feed may also be required when conventional fertilization cannot be optimized, if sufficient manure is unavailable, its use is unacceptable or high turbidity reduces phytoplankton productivity. Periods of heavy rainfall can make green pond conditions impossible to maintain.
How supplementary feeds work
Farmers providing supplementary feed to any stock, animals or fish, have a fundamentally different aim to giving a complete feed. An intensively raised animal, for example a hen or salmon raised in a cage, has negligible access to natural feed of any type and must be given a nutritionally balanced ration. In contrast, pasture-raised cattle or herbivorous fish require a feed that can complement the low-cost terrestrial or aquatic, pasture respectively, most economically. When herbivorous fish eat only natural feed, part of the protein is used for energy, which could be spared by giving supplementary feed high in energy. Furthermore, if natural feed is in short supply, especially as the density of filter-feeding fish increases, energy becomes limiting before protein. This explains why energy-rich grains and brans are commonly used in fertilized ponds and why, if fed alone or in ponds with little natural feed, they are nutritionally unbalanced and result in poor growth. Extending this concept of limiting nutrients by developing supplementary feeds that will support growth in static-water systems at higher fish densities has become of great commercial significance (Edwards, 1999).
Timing of supplementary feed
The low value of many freshwater pond fish may make supplementary feeding throughout the production cycle unattractive. Use of formulated feeds is still relatively uncommon unless the fish are relatively high value carnivorous species or destined for local luxury or export markets in developing countries, even if they are available. Strategic feeding of low-cost supplements is most cost effective at the beginning and end of the fish culture cycle.
Fish seed are a valued-added product for which economic, complete feeds have been unavailable but for which supplementary feeding is now the norm. Juvenile production of most freshwater species, in which hatchlings are raised to finger-sized fish ready to stock into on-growing ponds to produce table-size fish is normally undertaken in zooplankton-rich, earthen ponds. Even species that are herbivorous later in life require a diet high in crude protein and supplementary feeding, especially if high quality ingredients are used, allows much higher densities of fingerlings to be nursed to marketable size. High survival and quick turnover of stocks is essential for most nursery operations and supplementary feeding tends to improve returns compared to fertilization alone.
The fattening of table fish is a strategic option much employed in livestock production but little researched in fish culture. Conceptually the method may have application for a range of fish species, but especially those that can graze until the size when feeding becomes economic. Fattening will often be most cost effective in cages; the origins of cage culture were probably the fattening of undersized and less valuable, wild fish before marketing. Feeding high quality pellets as a supplementary ration in tilapia monocultures has been found to be most cost-effective if limited to fish that have attained 100-150 g after 3 months pond culture based on fertilization alone (Diana et al., 1996).
Feeding levels
Feeding of readily available by-products such as ricebran in fertilized ponds can be cost effective, but even cheap feeds can reduce returns compared to fertilization alone if optimal levels are exceeded. Rice bran boosted yields of a Nile tilapia monoculture receiving egg-laying duck manure by between 10-150 percent (AIT, 1986), but its use may not be cost effective. The law of diminishing returns was demonstrated when a low feeding rate (1 percent body weight day-1) increased yields, profitably by between 28 and 40 percent, for two levels of duck manure loading, but doubling feeding rate (2 percent) increased yields by a mere 4 percent or reduced them by 16 percent respectively (Yakupitiyage et al., 1991). This situation demonstrates that overall dry matter inputs into ponds must be carefully controlled in waste plus supplementary-fed systems or poor water quality can undermine yield improvements through improved nutrition. Feeding levels of grown tilapia (100-150g) can be reduced to 50 percent of satiation without a decline in individual growth (Diana et al., 1994).
Non-filter feeding fish
Carp farmers optimize the productivity of polycultures in India by strategic use of home-mixed, supplementary feeds in ponds receiving both poultry manure and inorganic fertilizers. Fish are stocked as stunted seed, trained to feed from plastic fertilizer bags with holes at the base. Optimal profitability has been achieved by doubling rates of fertilization over official recommendations (Nandeesha, 1993). The relative scarcity of benthic organisms make the energy sparing effect of supplementary feeds particularly valuable for benthic-feeding carps (Yakupitiyage, 1993); this may explain why the effect of supplementary feed is less pronounced for filter feeders in phytoplankton-rich water.
Supplements to complement natural food
Supplementary feed should complement limitations in the natural food present in the system. Critical standing crop (CSC) identifies the biomass of fish at which growth is retarded due to poor nutrition (Hepher, 1988). Tacon and De Silva (1997) have incorporated the standing crop of fish and natural food into the concept (Figure 16). Edwards et al. (2000) propose that supplementary feeding be classified into four stages. Research on Nile tilapia clearly indicated that only vitamins and minerals (excluding P) did not affect performance as indicated by growth and body composition (Box 5.D).
Supplementary feed under conditions of lack of natural feed
Farmers with turbid water ponds that do not respond to fertilization typically have poor results with supplementary feed alone. The problem is particularly common during the rainy season in ponds with unstable dikes. Fertilizers may also be washed out as ponds overflow under such conditions. Yields from ponds fertilized daily with inorganics alone at optimal levels (4 kg N and 2 kg P ha-1 day-1) were compared with using the equivalent amount of cash to purchase pelleted feed, rice bran or cassava chips. Pelleted feeds gave the best result but the researchers recommended farmers to use equal costed amounts of inorganic fertilizers and commercial pelleted feeds under such conditions (Edwards et al., 2000).
Manures contain considerable amounts of valuable nutrients for recycling through aquatic food webs to produce fish. Both the total amount and the proportion of nutrients available can be highly variable. Between 72-79 percent of the N, 61-87 percent of the P and 82-92 percent of the potassium in the diet was present in the waste of feedlot, egg-laying hens (Taiganides, 1978) indicating the scale of the resource. Generally the wastes from monogastric animals (pigs and poultry) are more nutrient-dense than wastes from grazing ruminants but the range of ruminant waste quality varies enormously (Table 5.2). Solid manure will often be mixed with a variety of other fractions such as urine, bedding, washing water and waste feed.
FIGURE 16
Schematic depiction of changes in the natural food organisms and fish yields, in relation to standing crop of the cultured organism and the ensuing protein needs of the supplemental feed (s)
Source: Tacon and De Silva (1997)
BOX 5.D Categories of supplementary feeding |
||||
Stage |
Category |
Limited natural feed |
Enhanced natural feed |
Notes |
1 |
No supplementary feed |
|
X |
Water may also be turbid e.g. rice bran (dry, wet mash) |
2 |
Energy-rich feed |
|
XX |
|
3 |
Energy-rich feed + some building blocks |
X |
XXX |
e.g. soybean cake and noodle-waste(wet dough, pellet) |
4 |
Energy-rich feed + some building blocks + catalysers |
XX |
XXX |
Di-calcium phosphate (DCP), salt (often pelleted) |
5 |
Complete feed |
XXX |
XXX |
Balanced amino and fatty acids, vitamins, minerals, and attractants (pelleted) |
X - XXX; least to most important
BOX 5.E Disappointing results with supplementary feeding
|
Species, size and sex also affect the quantity of animal manure produced. Husbandry level and waste management techniques can also greatly affect waste collection and use in fish culture. Complete collection of wastes may be impractical or costly, particularly for livestock raised under extensive or semi-confined conditions. Sometimes wastes cannot be used immediately, necessitating processing and/or storage and this will also affect their nutrient content. The quantity of waste produced by any animal is related to body weight, and thus any growth during a production cycle affects the design of livestock-fish systems. The availability of soluble nutrients from wastes of different types also affects their value, and is often related to age and original quality of the manure. Any calculation of the number of livestock required to optimise a given area of fishpond therefore needs to consider a wide range of variables.
BOX 5.F Summary of key factors affecting manured pond dynamics
|
Livestock waste output is often described in terms of nutrients available. unit body weight-1 (g nutrient.animal unit-1 day-1). This can vary greatly within the same species or strain managed under different conditions but for monogastrics raised on formulated diets under intensive conditions is relatively predictable. Relative waste production (g. kg live wt.-1) declines with age and size and food requirement also change. Thus more, smaller animals will generally produce more and higher quality waste than the same biomass of larger animals of the same type. Breeding animals fed high quality but restricted rations tend to produce smaller quantities of nutrient-rich manure than animals being fattened ab libitum. Monogastrics fed complete formulated diets have nutrient-rich wastes whereas ruminants fed fibre-rich feed produce wastes relatively high in dry matter, but poor in N and P. The large individual size of most ruminants means that outputs of total N and P per animal can be substantial. Low nutrient density and presence of tannins negatively affect the value of ruminant faecal wastes in aquaculture, reducing its value as pond fertilizers compared to that suggested in actual levels of N and P. However, intensification through confinement and improved feeding can improve the availability and quality of wastes.
The diverse nutritional needs of egg-laying and meat breeds of poultry also greatly affect manure composition. The calcium and P-rich excreta of laying birds reflect diets designed to support eggshell development and the high fibre content in waste of juvenile (replacement) laying birds fed reflects the content of fibre in their diet. The considerable proportion of total N found in urine and washing water (Figure 17), as opposed to solids has important implications for their management. Generally urine contains little P, except for pigs.
Feedlot, scavenging and intermediate systems
Livestock managed outside of feedlots have more variable waste characteristics and generally less of the nutrients produced can be collected. Animals that are stalled some or all of the time, and fed cut-and-carried forages and/or concentrates, tend to show both increased productivity and produce wastes of higher quality and collectability (Box 5.G).
TABLE 5.2
Input and output of poultry waste fed-aquaculture
|
Inputs (g m-2 day-1) |
Output (g fish m-2 day-1) |
|||||||
System |
Poultry waste |
Other |
Fish |
|
|
||||
|
DM |
N |
P |
DM |
N |
P |
|
|
|
Feedlot |
|
|
|
|
|
|
|
|
|
Egg-laying ducks |
6.71 |
0.3 |
0.07 |
- |
- |
- |
Tilapia |
2.82 |
200 m2 ponds, 6 months (Edwards et al., 1983) |
Broiler chickens |
10.0 |
0.4 |
0.46 |
- |
- |
- |
Tilapia, Common carp |
2.87 |
400 m2 ponds, 3 months (Hopkins and Cruz, 1982) |
Layer chicken |
14.3 |
0.4 |
0.3 |
- |
- |
- |
Tilapia |
1.33 |
1 000 m2 ponds, 5 months (Green et al., 1994) |
Layer chicken |
1.07 |
0.03 |
0.018 |
- |
0.47 |
0.23 |
Tilapia |
2.75 |
220 m2 ponds, 5 months (Knud-Hansen et al., 1991) |
Scavenging |
|
|
|
|
|
|
|
|
|
Muscovy duck |
9.7 |
0.15 |
0.10 |
- |
- |
- |
Tilapia |
1.38 |
5 m2 tanks, 3 months; ducks fed75 percent ad libitum (AFE, 1992) |
Egg-laying duck |
3.0 |
0.23 |
0.03 |
- |
0.17 |
- |
Tilapia |
1.21 |
200 m2 ponds, 4 months |
Egg-laying duck |
1.24 |
0.20 |
0.01 |
- |
0.17 |
- |
Tilapia |
0.53 |
200 m2 ponds, 4 months (AASP, 1996): PADDY bran |
Source: Little and Satapornvanit (1996)
FIGURE 17
Annual production of nitrogen in faeces and urine for various livestock
Source: Modified from FAO (1980)
BOX 5.G Goat management level affects nutrients collectable for aquaculture
Source: Little and Edwards (1999) |
Feed quality
Quality of feed greatly affects both livestock performance and waste quality for aquaculture. However, there are few data for fish yields derived from pig production other than from complete balanced diets. An exception is Long (1995) who found that although fattening hybrid pigs fed only cooked village ricebran and chopped water hyacinth grew at half the rate of pigs fed balanced rations, their waste could still support extrapolated fish yields (more than 4 tonnes ha-1 year-1) higher than the norm in the Mekong Delta.
Performance and waste availability of semifeedlot poultry have also been little studied but preliminary research is encouraging. In scavenging poultry, manure quality is greatly affected by the quality and quantity of supplementary feeds, which in turn affects fish production. Compared to confined birds fed complete diets, nutrients in the waste are low, resulting in poorer performance of fish raised on the waste (Box 5.H; Table 5.2). However, using more waste from a larger flock of ducks can compensate for the smaller amount of lower quality manure produced per duck from scavenging birds penned for supplementary feeding and waste collection only at night.
There are tradeoffs between benefits to livestock and fish regarding the type of supplementary livestock feed used. Egg-laying ducks (Khaki Campbell x local hybrids) fed paddy grain at night produced poorer quality manure than those fed rice bran. The amount of N and P in the manure was 50 percent and 25 percent, respectively, of that found in ducks fed relatively nutrientdense, village rice bran. Fish yields reflected these different nutrient inputs, with ricebran-derived fish yields being almost double those from paddy-fed ducks. However, the number of eggs produced was significantly higher for ducks fed paddy rice compared to ricebran (Figure 19).
Total collectable nutrients in the waste can exceed that present in the feed given as a night -time supplement. Natural food scavenged during the day is usually of higher food value than the supplement (Box 5.H and Box 5.J).
FIGURE 18
Goat production and total collectable nutrients in different management systems
Source: Little and Edwards (1999)
FIGURE 19
Egg laying rate of Khaki-Campbell x local strain ducks fed two different supplementary diets, T1=unhulled paddy rice and T2=village rice bran
BOX 5.H Quantity of supplementary feed affects scavenging poultry wastes and fish production
|
BOX 5.I Integration of duck and fish production in ricefields in the Philippines Ten approaches were compared for the management of lowland rice-fields; conventional rice monoculture, conventional rice-fish culture, rice monoculture, rice-fish, rice-azolla, rice-duck, rice-fish-azolla, ricefish- duck, rice-azolla-duck and rice-fish-azolla-duck. Pesticides were only used in the conventional treatments. In addition to an improved understanding of the specific interrelationships within such complex systems, the study highlighted the potential benefits of integration to the stability of rice production ecosystems. Key findings included
Source: Cagauan (1999) |
Herding of ducks in ricefields typically reduces the input of wastes into associated pond fish production systems. However extensive management of ducks remains a popular system in many places and can be integrated with fish production within the ricefield itself. Cagauan (1999) investigated the productivity of rice, ducks and fish (Oreochromis niloticus) within experimental integrated systems in Central Luzon, Philippines. The management of the hybrid aquatic fern (Azolla microphylla Kaulf. x Azolla filiculoides) was also investigated for its impact on fish and duck productivity and prevalence of weeds and Golden Apple snail (Pomacea canaliculata Lam.) in ricefields (Box 5.I).
BOX 5.J Impacts of supplementary feed quality on waste characteristics
Source: Naing (1990) |
Spilled feed
Spilled feed is a loss to the livestock production system but a gain to the fish because of its direct feeding value. The timing, frequency and location of feeding affect the amount of spilled feed and its direct value to fish. Feed presentation and composition, availability and design of water containers and livestock type all affect the amount or proportion of spilled feed in collectable wastes. Feedlot ducks fed complete diets appear to waste less than birds allowed to scavenge during the day and given access to supplementary feed at night. Feed processing can reduce spillage; up to 15 percent of granulated feeds may be lost compared to 10 percent if the same duck feed is pelletised (Barash et al., 1982). Ad libitum feeding of all livestock increases spillage losses and feeder and waterer designs are important to minimize losses.
Feeding behaviour and the nature of different feeds may increase the amounts of feed available directly for fish. Waste feed left in the water container comprised more than 25 percent of the collectable dry matter from scavenging muscovy ducks fed a supplement of village rice bran (AASP, 1996).
Nutrients contained in manure are not necessarily available for uptake by the food web in the pond. Indeed only a fraction is usually released in a soluble organic form, usually within the first few days of adding the waste to the pond (Knud Hansen, 1998). Cumulative nutrient release curves for buffalo and duck manure indicate that the nutrients within manure from poorly-fed ruminants are relatively unavailable compared to monogastrics fed high quality feed. Manures with high levels of material resistant to decomposition may quickly build up in the sediments, requiring frequent removal.
Animal waste quality can be drastically affected by the method of collection and storage prior to its use in fish culture. Bedding materials and washing water have particular influence on composition of manures, but storage condition and duration also affect their value.
FIGURE 20
Dry matter (DM), total nitrogen (N) and total phosphorous (P) in wastes of ducks
Source: Modified from FAO (1997a)
FIGURE 21
Loss of total nitrogen in fresh layer manure with time
Source: Flegal et. al. (1972) in FAO (1980)
The use of litter or bedding is a key part of intensive livestock husbandry. High densities of animals can be kept in good health as litter absorbs moisture, reduces odor and ammonia and provides a soft floor surface. Typically bedding increases dry matter, ash, fibre and crude protein in the manure. Litter is a living community of detritus and microorganisms that varies in quality with type of substrate, livestock and its management. Fermentation occurs over time and results in losses of N, volatalized as ammonia or becoming refractory and unavailable. The concentration of some vitamins may increase (vitamin B12) and some antibiotics (e.g. chlortetracycline) decrease with duration of storage (Chapman, 1994).
The steady decline in nutrient content of manures not collected on litter during storage is accelerated if wastes are stored under conditions of high temperature and/or exposed to wind and rain (Figure 21).
The needs of fish production for waste input may vary through the year and/or not match waste production by the livestock throughout the entire production cycle. Thus, removal and storage for later use, or sale or use elsewhere on the farm may be required. A storage pit is a common feature on Chinese integrated farms allowing strategic use of manure and prevention of overloading during seasonally cool periods.
Processing of waste, particularly through anaerobic fermentation to produce biogas, can result in concentration of nutrients as C is lost. This option has been promoted as a method of reducing waste volume and increasing its value for fish culture. The reduction of BOD in waste during anaerobic fermentation means that associated aquaculture can process the wastes of greater numbers of livestock.
Overall evidence suggests that both aerobic and anaerobic storage of livestock wastes reduces their net value. Aerobic composting results in losses of N as ammonia, whereas although anaerobic fermentation concentrates N, much of it is boundup or refractory and not readily available for use by organisms in the ponds food web.
BOX 5.K Summary of factors affecting livestock waste characteristics
|
Lekasi et al. (2001) assessed manure management in the Kenya Highlands and found it to be critical for sustaining livelihoods of livestockdependent people. Smallholder dairying was particularly important and the major conduit for import of external nutrients onto farms through purchase of feeds and supplements. Farmers had many ideas for the better management of manures but few were implementing them, especially urine conservation. Manures were also bought and sold at prices above their nutrient value alone, indicating the value farmers placed on the physical benefits to soil quality.
Backyard pig production in Northeast Thailand has been limited to households who can obtain sufficient rice by-products. Greater availability of complete feeds and better transportation systems is opening up further opportunities for backyard production
Washing waste duck feed from the waterer into fishponds nearby. This is a high quality pond input
Maintaining optimal levels of nutrients and water quality condition to meet the needs of pond food web organisms, fish and other components of the farming system requires an understanding and application of basic concepts rather than a recipe. Best fertilization practice is rarely based on fixed levels of nutrient inputs (or numbers of livestock per area of pond) because both the requirement of the fish culture system and the availability of waste are likely to vary. Nevertheless, for the tropics, upper levels of animal numbers per unit area of fish pond can be given, assuming standard feeds and size of livestock and, a need to maintain aerobic conditions (Little and Muir, 1987).
Systems designed for easy removal and use of livestock wastes, allow the most complete integration of livestock with fish, resulting in the least loss of nutrients and their most efficient use in aquaculture.
Livestock pens should be designed for easy collection and removal of wastes as well as high performance of the animal. The design and materials should allow for quick and complete cleaning and reduce exposure of wastes to direct sunlight, wind and rain. Livestock should be confined in a way that does not allow damage to the pond dikes. Allowing swimming birds and wallowing ruminants access to the pond dikes can result in rapid deterioration to pond structure, turbid water and loss of manure. Reinforcement of dikes with brick, concrete and synthetic sheet has been tried to prevent damage with limited impact in many cases. Limiting ducks to a cleanable pen and water surface is desirable.
Collection of urine, especially for ruminants is often overlooked and yet it contains a considerable proportion of the N. Cracked or badly designed floors of pens, although still allowing continued collection of manure, results in losses of urine.
Close proximity of the pen and fish pond generally results in more consistent transfer of wastes from livestock to pond. Sometimes pens can be situated over ponds, allowing constant addition of waste to the pond used, which may optimize fertilization. High land costs and availability of suitable building materials and expertise are the norm for this type of system. The disadvantages include difficulties in access to both livestock and fish and in controlling waste inputs to the levels required as well as higher construction costs, particularly for pigs and large ruminants. Household pigpens in Northern Viet Nam are a major source of manure for the whole farming system. Situating the pen, with well-sloped floors and pipes close to small fish ponds, allows urine to be recycled immediately with little loss of N, and strategic use of solid waste in field crops and horticulture. Location of livestock pens close to fish ponds should consider security, all-weather access and good distribution of wastes in the pond.
Protecting the dikes of duck-fish ponds from damage using fences and ramps to allow access to the water
TABLE 5.3
Effect of feeding and management on waste characteristics of livestock
|
System |
Feed |
Production |
Waste (g/animal/day) |
|
|||||
Poultry |
Feed lot |
Scavenging |
Complete |
Supple-mentary |
Daily live W. gain (g/d) |
Laying rate (percent) |
DM |
N |
P |
|
Egg laying duck |
Yes |
No |
Yes |
No |
1.88 |
46-58 |
44.7 |
1.97 |
0.49 |
Edwards et al. (1983) |
Egg laying chicken |
Yes |
No |
Yes |
No |
- |
- |
44 |
1.3 |
1.14 |
FAO (1980) |
Broiler chicken |
Yes |
No |
Yes |
No |
32 |
- |
20 |
0.7 |
0.92 |
Hopkins and Cruz (1982) |
Egg laying duck |
No |
Yes |
No |
Yes |
0.38 |
16.3 |
59.9 |
1.16 |
0.69 |
AASP (1996) (rice bran) |
Egg laying duck |
No |
Yes |
No |
Yes |
0.42 |
29.8 |
24.8 |
0.52 |
0.16 |
AASP (1996) (Paddy rice) |
Muscovy duck |
No |
Yes |
No |
Yes |
10.4-16 |
- |
40-70 |
0.65-1.28 |
0.5-0.8 |
AFE (1992) |
|
Mean values for ducks on a range of dietary regimes; includes nutrients in manure, dry and wet waste feed |
|
|
Khalil, (1989) |
||||||
Muscovy duck |
Yes |
No |
Yes |
No |
Male 32 |
- |
33 |
1.65 |
0.71 |
|
|
|
|
|
|
Female 20 |
- |
24 |
1.24 |
0.55 |
|
Growing pig |
Pigs fattened on cooked rice byproducts and water hyacinth; values include manure, urine and washing water |
|
|
Long (1995) |
||||||
|
Yes |
No |
Yes |
No |
370 |
- |
1400 |
29.1 |
13.6 |
|
Goats |
|
Based on replicate Katjang goats (n=3) of between 17-30 kg initial weight over a three month period. Wastes removed daily unless otherwise stated |
|
|
Little and Edwards (1999) |
|||||
|
yes |
No |
Yes |
No |
106 |
|
453 |
7.84 |
3.47 |
Stall fed, ricebran and legume fodders |
|
yes |
Yes |
No |
Yes |
62 |
|
256 |
5.12 |
2.27 |
Grazing+ricebran |
|
yes |
Yes |
No |
Yes |
69 |
|
266 |
8.08 |
0.74 |
Grazing+legume fodder |
|
yes |
Yes |
No |
No |
35 |
|
167 |
4.57 |
0.57 |
Grazing only |
|
Yes |
Yes |
No |
No |
16.7 |
|
200 |
1.27 |
0.41 |
Grazing only; irregular waste removal |
Transfer of wastes away from peri-urban areas where land costs are high, to areas where fishponds can be larger and waste use more efficient, can be an advantage. Additionally, it can provide opportunities for poorer people to benefit as waste traders.
In situations where land costs are high and airbreathing fish can be economically raised, farmers may seek to exceed the normal density of livestock per unit area of pond. Alternatively they may add feeds or fertilizers to supplement animal production wastes (Box 5.L). More normally livestock waste is insufficient to optimize fish production, because of risks or costs associated with intensive livestock husbandry or problems managing the waste or because of the limited resources of small-holders.
Collecting manure from underneath roosts of village chickens for use in fishponds. Greater frequency of waste collection increase both the quantity and quality of collectable wastes
Large integrated meat duck-fish ponds in Taiwan are lined with bricks to reduce erosion of the dikes
Pigs being cooled and pens being washed down; water from fishponds is used for this task and the enriched water runs via drains back to the fish pond
BOX 5.L Pornsaks duck slaughterhouse in Bang Lane, Central Thailand
|
When livestock and fish are raised within a broader farming system, experienced farmers design the integration to more closely match their overall needs. The cost of complete nutrient collection may be too high. Concrete floors that allow collection of urine and washing water may not be affordable, or the high cost of nutrients may require manures to be used elsewhere on the farm (Box 5.M).
BOX 5.M Surins use of duck manure in Nakon Pathum, Thailand Lung Surin integrates a flock of 1500 egg-laying ducks with fish and fruit production on his farm in Nakon Pathum, Thailand. He chooses to use duck manure relatively extensively for fish culture (total area of ponds more than 2 ha) rather than optimize the production of fish alone. His strategy allows him to reduce the production costs of eggs, maintain duck health and also to use manure for horticulture. Rice is double cropped in the area, presenting opportunities several times in the year to suspend complete feeding and allow ducks to forage for food at low cost in the surrounding fields. The ducks also scavenge in the margins of the very large ponds year-round, which maintains their condition. Ducks raised in feedlots at high density are more prone to stress-related illness. In addition to the natural feed that the ducks scavenge, he optimizes egg production through using a high quality feed that he mixes himself. One of the methods that he uses to maintain duck feed intake, a key parameter of productivity, is to offer the feed as a moist mash frequently during the day. At the hottest time of the year, he also feeds during the night when temperatures are lower. Leftovers from each feeding, which quickly become unappetizing, are washed into the pond for fish. Additionally he uses pig manure purchased from a neighbouring farm and various supplementary feeds if available. |
BOX 5.N Factors affecting characteristics of livestock waste and its use for aquaculture
Source: Modified after FAO (1980) |
Ducks confined over fish ponds may aerate the water through their swimming and feeding behaviour
Collecting urine and washing water from pigs fattened on concrete-floored pens
BOX 5.O Questions to ask during the design of livestock-fish systems
1 Opportunity cost for use elsewhere on the farm (field crops, horticulture), sold off or used for biogas production. |