1. Preservative effect of chilling

Contents - Next

Why fish go bad

As soon as a fish dies, spoilage begins. Spoilage is the result of a whole series of complicated changes brought about in the dead fish by its own enzymes, by chemical action and by bacteria. It is necessary to understand something of the way in which these changes take place in order to make the fullest use of chilling as a means of keeping them in check.

An important series of changes is brought about by the enzymes of the living fish which remain active after its death. They are particularly involved in the flavor changes that take place during the first few days of storage, before bacterial spoilage has become marked.

Millions of bacteria, many of them potential spoilers, are present in the surface slime, on the gills and in the intestines of the living fish. They do no harm because the natural resistance of a healthy fish keeps them at bay. Soon after the fish dies, however, bacteria begin to invade the tissues through the gills, along blood vessels, and directly through the skin and the lining of the belly cavity.

In addition to bacterial and enzymatic changes, chemical changes involving oxygen from the air and the fat in the flesh of species such as tuna and mackerel can produce rancid odours and flavours.

Thus, spoilage is a natural process once the fish dies, but chilling can slow down this process and prolong the shelf life of fish as food.

Effect of temperature on spoilage

There are three important ways of preventing fish going bad too quickly - care, cleanliness and cooling. Care in handling is essential because unnecessary damage can provide access through cuts and wounds for the spoilage bacteria, thus hastening their effect on the flesh. Cleanliness is important in two ways: (i) the natural sources of bacteria can largely be removed soon after the fish is captured by taking out the guts and washing off the slime from the surface of the fish; (ii) the chances of contamination can be kept to a minimum by ensuring the fish is always handled in a hygienic manner. But most important of all. the fish must be chilled quickly and kept chilled.

The speed with which bacteria grow depends on temperature. Indeed, temperature is the most important factor controlling the speed at which fish go bad. The higher the temperature, the faster the bacteria multiply, using the flesh of the dead fish as food. When the temperature is sufficiently low, bacterial action can be stopped altogether; frozen fish stored at a very low temperature, for example 30 C, will remain wholesome for a very long time because bacteria are either killed or completely inactive at this temperature, and other forms of spoilage progress only very slowly. But, even at -10C, some kinds of bacteria can still grow, although only at a very slow rate. Therefore for long-term storage, of many weeks or months, freezing and cold storage are necessary.

It is not possible to keep unfrozen fish at a temperature low enough to stop bacterial action completely, because fish begin to freeze at about -1C. However, it is desirable to keep the temperature of unfrozen fish as close to that level as possible in order to reduce spoilage; the easiest and best way of doing this is to use plenty of ice which, when made from clean fresh water, melts at 0C.

At temperatures not much above that of melting ice, bacteria become much more active and fish consequently goes bad more quickly. For example, fish with a storage life of 15 days at 0C will keep for 6 days at 5C and only about 2 days at 15C before becoming unacceptable.

The chemical changes that contribute to spoilage are also kept in check by reducing the temperature; therefore it cannot be too strongly emphasised that temperature is by far the most important factor governing the rate at which fish go bad.

How long will fish keep in ice?

Generally, all fish spoil in much the same way, with 4 distinct phases of spoilage. Cod, for instance, will keep in ice for about 15 days before becoming inedible, and this time can be divided roughly into successive periods of 0 to 6, 7 to 10, 11 to 14 and over 14 days. In phase 1 there is very little deterioration apart from some slight loss of natural or characteristic flavour and odour. In phase 2 there is a considerable loss of flavour and odour. In phase 3 the fish begin to taste stale, appearance and texture begin to show obvious signs of spoilage, and the gills and belly cavity have an unpleasant smell. All these changes, which in the latter stages of storage are almost entirely due to bacteria, occur at an ever increasing rate until the fifteenth day, when phase 4 begins, the fish are putrid and generally regarded as inedible.

Other species with different storage times may also have a different division between the respective phases but the spoilage pattern will be similar. Even fish of the same species may spoil differently since factors such as the method of capture, location of fishing grounds, season of the year, fat content and fish size can affect the keeping quality.

Most studies on the spoilage of fish are done under controlled conditions, therefore, the results are more specific than they would be in most commercial situations where conditions can be variable. Published data on storage life should therefore be used with discretion and in most cases assumed to represent maximum values.

Despite the above limitations, the storage life of fish has been well researched and documented and several broad conclusions have been drawn. In general, flat shaped fish keep longer than round shaped fish; red fleshed fish keep longer than white fleshed fish; low fat fish longer than high fat fish and teleost (bony) fish longer than elasmobranch (cartilaginous) fish.

There are many references in the literature to the extended storage life in ice of certain tropical fish species compared to fish from temperate or colder waters. While it is true that some fish from tropical waters can be kept for longer periods an extensive review of published literature showed that this was not universally the case. Table 1 shows storage life of various fish species. For further information on shelf life, reference should be made to the publication "Fresh fish quality and quality changes", FAO Fisheries Series No 29. The reasons for apparent anomalies, or exceptions, are still not fully understood. Another factor which makes comparisons difficult is that different criteria are used to define the limit of storage life and, since relatively few studies have been made on spoilage in ice of tropical fish, direct comparisons are not always possible.

Fig. 1 The effect of temperature on the spoilage of lean, temperate water fish

Table 1 Shelf life in ice (FAO Fisheries Series No 29)

Species Shelf life
(days in ice)
Temperate water
Marine species  
white-fleshed lean  
(cod, haddock, hake) 11-13
flatfish (sole, plaice) 15-18
halibut 21
fatty fish  
summer herring (fatty) 2-4
winter herring (lean) 12
Freshwater sp.  
trout 9-10
Tropical water:
Marine species  
Bahrain(3 species) 13-25
Ghana(5 species) 19-22
Brunei(3 species) 18-28
Sri Lanka(5 species) 20-26
Seychelles (8 species) 15-24
Mexico (6 species) 21-30
Hong Kong (2 species) 30-31
India (4 species) 7-12
Freshwater spp
Pakistan (2 species) 23-27
Uganda (5 species) 20-25
East Africa (4 species) 15-28

In the absence of specific information on storage life, a simple storage experiment will serve to show how long a species can be kept in ice. All the relevant conditions pertaining to the storage period should be met. But, if seasonal changes are likely, adjustments should be made as necessary, or further storage experiments should be carried out at the appropriate time or under simulated conditions.

Although the data are limited, it is generally accepted that the overall pattern of spoilage of freshwater fish in ice is similar to that of marine species, but their storage life tends to be longer.

Definition of storage life

A wide range of terms are used in discussing storage life, such as quality, acceptability, preference, keeping time, storage time, storage life, shelf life and potential shelf life; these terms are not interpreted uniformly.

The simplest definition of the end of shelf life, or storage life, is the point at which the produce is considered to have become inedible, i.e. it is spoilt! Even this simple definition is open to different interpretations since there is no common level of unacceptability, even within small communities let alone worldwide.

At the other end of the quality scale, the "high quality life" (HQL) can be regarded as the point at which the produce retains all its characteristic properties. The equivalent definition in the EEC Labelling Directive is "retains its specific properties" and, in CODEX and US Grade Standards, the cooked product is required to have "characteristic flavours and no off- flavours".

Although HQL is more easy to define and thereby more widely acceptable, in practical terms, it may have little significance with respect to the commercial value of the product. Some personal preferences for instance, may even favour produce which exhibits "non-characteristic off-flavours".

Assessment of quality can either be done by objective or sensory methods. Again, differences in methodology may give significantly variable results.

From long experience in fisheries inspection and research, it is possible to correlate objective and sensory methods but, it is not possible to include consumer preferences in such correlations, since the "end of good or acceptable quality" is not a standard criterion which can be universally applied. Therefore, fish quality standards have to be matched to market requirements rather than absolute standards. Tables of storage lives should therefore be interpreted with caution and related to the situation and conditions which prevail for individual products.

A good deal of attention is given to TTT (time, temperature, tolerance) when compiling storage life tables but the PPP (product, processing and packaging) factors can be equally important. Thus, tables which do not specify all of these relevant conditions should be used only as a rough guide.

In summary, tables of storage lives should be used only to give rough guidance; more accurate information can be achieved by experimentation or experience when all the prevailing factors are taken into account.

Calculation of storage times

It is generally accepted that bacterial spoilage is the major reason for non-sterile unfrozen fish becoming unacceptable to the consumer. As the spoilage flora proliferates the fish become progressively spoiled.

For many years a general rule was applied, that bacterial growth, and hence spoilage rate, doubled for every 5C rise in temperature; this can still be used as a rough guide in making comparisons. For instance, fish which have a storage life of 14 days at 0 C will have a storage life of only 7 days at 5C. Closer studies of the effect of temperature on spoilage have been made, and it has been demonstrated that the square root of the growth rate of bacterial cultures is a linear function of temperature over a significant range up to about 1 5C. This relationship is expressed mathematically by the following equation:

R = b (T -Tc) (1)
where R = rate of growth per unit time
b = slope of the regression line
T = absolute temperature at which growth is measured
Tc = conceptual temperature (k).
Mathematically Tc is the value of T when R = 0.

The minimum temperature at which chilled fish is normally stored is close to 0 C. Therefore, it is convenient to simplify equation (1 ) and re-define the growth rate r as the rate relative to that at 0C. By manipulating the expression in equation (1 ) we get

r = 0.1t + 1 (2)
where r = rate of spoilage relative to the rate at 0C
t = temperature of storage (C)

This can then be rearranged to give: r = (0.1t + 1)2 (3)

Using equation (3) the spoilage rate at any temperature relative to the spoilage rate at 0C can now be calculated. For example, the spoilage rate at 5C will be:

r = [(0.1x5) + 1]2 = 2.25

This means that fish maintained at 5C will spoil at a rate which is 2.25 times quicker than the rate at 0C, or expressed in another way, one day storage at 5C is equivalent to 2.25 days storage at 0C. This differs slightly from the factor of 2 derived using the "doubling rule". A similar calculation for a 1 0C storage temperature shows that the spoilage rate is increased by a factor of 4.

Using the relationship expressed in equation (3) and making the appropriate calculations, it is possible to predict the likely storage life of fish which have been maintained for some time at temperatures higher than the ideal 0C. For example, if fish with a normal storage life of 15 days at 0C are held initially at 10C for one day, and 5C for 2 days before being reduced to 0C for the rest of the storage time, the likely shelf life can be calculated as follows:

1 day at 1 0C is equivalent to 4 days at 0C
2 days at 5C is equivalent to 2x2.25 = 4.5 days at 0C

The equivalent storage time at 0C of the 3 days spent at the higher temperatures is therefore 4+ 4.5 = 8.5 days. This means that 8.5 - 3.0 = 5.5 days of the potential storage time of the fish, if they were maintained at 0C, have been lost and the total storage time is thereby reduced from 15 to 9.5 days.

The above is a simplified example used to illustrate the significant losses in potential storage life if fish are kept at higher temperatures even for short periods. In reality, the temperature history of the fish is likely to be more complicated and, in order to work out equivalent storage times, calculations will need to be made using smaller time intervals. If the facilities are available, this calculation can be done using a computer to give storage life predictions under a wide variety of conditions.

Thus, simple integration of time and temperature functions can provide a useful indication of spoilage, provided that the storage life data at some specific temperature are available, preferably, but not necessarily at 0C.

Instruments have also been developed to continually monitor fish temperature and perform the time-temperature integration function; one model displays the number of days of potential storage life remaining at 0C. Time-temperature growth characteristics of spoilage bacteria vary, depending on, for instance, whether they are mainly psychrotrophic, as would be the case in temperate waters, or mesophilic, in tropical waters. Time-temperature integration instruments therefore need to be programmed for the particular fish species and situation under consideration.

Why cool fish with ice?

Ice as a cooling medium for fish has a great deal in its favour; it has a very large cooling capacity for a given weight or volume, it is harmless, portable and relatively cheap. It is especially valuable for chilling fish, since rapid cooling is possible. When fish are being cooled with ice, heat transfer is achieved by direct contact with the ice, by conduction through adjacent fish and by melt-water flowing over the fish. Cold melt-water takes up heat from the fish and when it flows over ice again it is recooled. Thus, intimate mixing of fish and ice not only reduces the thickness of the layer of fish to be cooled but also promotes this convective cooling interaction between ice melt-water and fish.

As soon as ice is put on warm fish, heat flows from the fish to the ice and melts it. Heat keeps on flowing until there is no difference in temperature between the fish and ice, provided sufficient ice is present. Any further melting that occurs is due to heat from other sources, such as the warm surrounding air during the subsequent storage period.

Ice is its own thermostat and, since fish are mainly water, ice maintains fish at a temperature just slightly above the point at which they would begin to freeze; the point of equilibrium for sea fish, iced soon after catching, is near to -0.5C, since the mixture usually includes some salt and blood.

Why not use other methods of cooling?

There are other ways of chilling fish besides using ice. For example, they can be immersed in chilled water or cold air can be blown over them. Sea water, cooled by mechanical means, Refrigerated Sea Water (RSW) or by the addition of ice, Chilled Sea Water (CSW), is a suitable alternative means of rapidly chilling large quantities of small whole fish, especially on board a fishing vessel: the use of RSW and CSW is discussed in detail in chapter 7. The use of cold air is less satisfactory, except for some applications relating to prepackaged fish which are also discussed in Chapter 7.

When cold air alone is used, as in a chillroom, the heat taken from the fish will rapidly warm the air. The warm air rises and is cooled by contact with the coils of the cooler and then moves by natural convection or fan circulation back to the fish. It does not take much heat to warm the air; it takes 10,000 times as much heat to melt a given volume of crushed ice as it does to warm the same volume of air from 0 to 0.5 C. Thus, it is important to remember that for air cooling to be effective, a good circulation of cold air must be blown over the fish. However, even when a fan is fitted in a chillroom it is difficult to achieve the rapid cooling rates possible with ice and chilled sea water (Fig. 2).

Another disadvantage of air chilling is that, without the use of ice, the fish becomes dry. Continuous air movement evaporates water from the fish surface and deposits it as frost or condensate on the coils of the evaporator. In addition, the air in some parts of the chillroom will be colder than in others. Fish in the cold spots, for example close to the evaporator, may in time become partially frozen, even though the thermostat is set above freezing point at another location within the chillroom. Slow freezing of the fish can be detrimental, since appearance, flavour and texture of the fish may be affected.

Fig 2 Fish in chillrooms still need ice

Contents - Next