Back to Home Page of CD3WD Project or Back to list of CD3WD Publications

PREVIOUS PAGETABLE OF CONTENTSNEXT PAGE

5 Quantifying the losses: What happens in storage?

For a long time it was generally assumed that losses caused by insects in post-harvest operations are high, without these losses having been precisely quantified. For instance, a report published by the National Academy of Sciences in Washington (National Research Council, 1978) not very helpfully states a loss rate of 1-100% for maize in developing countries. Due to the lack of universally applicable and exact methods of loss assessment, it became habitual to generally assume a figure of about 30% for post-harvest losses. More recent, although controversial estimates also generally assume high losses (thus Oerke et al., 1996).

In order to gain a more reliable data basis, a comparison of different methods of loss analysis in farm maize stores in southern Togo was carried out from 1983 onwards in the context of a doctoral thesis supported by project funding (Pantenius, 1988). The counting and weighing method proved to be the most appropriate approach with due regard to the workload required and the reliability of results. In addition to methodology development, this project determined for the first time precise loss rates for farm maize stores, both such with LGB infestation and without LGB infestation (cf. Box 1).

Box 1: Maize losses in southern Togo

Pantenius (1988) examined, in a two-year study, the storage losses of husked cob maize in traditional farm stores (of the Ebli-va and Kédélin types, and above the hearth fire) in southern Togo. In various trials with untreated maize he arrived at the following results:
· after 6 months without LGB: 7.1% loss (dry matter)
· after 6 months with LGB: 30.1% loss (dry matter)
· Hybrid varieties are more susceptible to loss than local varieties.
· Unfavourable weather conditions diminish storability.
· Losses are very much higher in the short storage period than in the long storage
period.

However, it needs to be noted that the widespread practice of continuous withdrawal of maize from storage for consumption or for sale throughout the storage period leads to the actual storage losses being overestimated. The 30% loss found by Pantenius (1988) is thus a relative figure that relates exclusively to the maize remaining in storage in the last month of the storage period. Over the whole storage period and quantity, total losses in fact figure some 17% of the quantity put into store. This is shown by Table 1:

Table 1: Determination of storage losses in absolute terms from Box 1

Storage duration
(months)

Quantity in storage
(kg)

Withdrawal
(kg)

Storage loss
(%)

Storage loss
(kg)

Put into store

100

-

-

-

1

90

10

2.0

0.2

2

80

10

4.0

0.4

3

60

20

8.6

1.7

4

40

20

18.9

3.8

5

20

20

25.0

5.0

6

0

20

30.1

6.0

Total

-

100

-

17.1

Accumulated storage loss after 6 months: 17.1% of the quantity put into store

Shortly afterwards, Böye (1988) quantified losses in cob maize in Costa Rica in the context of his doctoral thesis. Here the focus was placed on the LGB. Box 2 summarizes the results, which were similarly determined by the counting and weighing method.

Box 2: Maize losses in Costa Rica

Böye (1988) determined in the course of three storage periods (1985-87) the losses in husked cob maize in traditional timber storage structures of the "Troja" type in Costa Rica. The measured dry matter losses averaged:

· when put into storage in the dry season: 5% after 6 months
· when put into storage in the rainy season: 12% after 6 months
· when put into storage in the rainy season: 13.6% after 9 months

The differences were attributed to increased levels of infestation and better conditions of development for storage pests in the rainy season.

The significantly lower losses as compared to Togo gave cause to search for natural limiting factors in the region of origin of the LGB. This search led to the discovery of the beneficial insect Teretriosoma nigrescens (Tn) and its release in several African countries.

The extent of losses in small farm storage systems continued to play an important role in a series of further doctoral theses. Helbig (1995) largely confirmed the bandbreadth of losses described by Pantenius (1988) and showed with surveys over three storage periods that the LGB causes dry matter losses ranging from 5% to 26.5% in Ebli-va stores after 6 months, depending upon season and site. After 8 months a maximum value of 36.5% was reached. Further loss analyses prior to the release of Tn were carried out by Henckes (1992) in Tanzania and Albert (1992) in Togo. The results are shown in Boxes 4 and 5 (section 6 below).

Maize losses display considerable variability, both across the individual trial series and among the different studies. This variability is due to a great number of influencing factors, of which the most important are:

· Maize variety and quality of the stored product (ripeness, moisture content, etc.)
· Type of store, duration and technique of storage
· Climatic factors as a function of location and season
· Infestation by storage pests in the field before storage
· Spectrum of pests (species and relative frequencies).

The loss assessments that have been carried out are therefore only case studies that must be interpreted in the specific context. The findings available from southern Togo do give an overall picture of the losses in this region. However, care needs to be exercised when extrapolating these findings to other regions for which less information is available. It must further be kept in mind that the findings are approximative values that depend upon the type of analysis chosen.

One a half years after the release of the beneficial insect Tn in southern Togo, Mutlu (1994) carried out first tests to determine the loss-reducing effect of Tn upon LGB infestation. She was able to demonstrate that Tn had actively spread on its own, and quantified the reduction in LGB populations and the loss-reducing effect in stored maize (cf. Box 3).

Box 3: Loss reduction due to biological control of the LGB
in Togo

Mutlu (1994) examined during two storage periods (1992-93) the reduction in LGB-related losses after the release of the beneficial insect Teretriosoma nigrescens (Tn) in traditional small farm stores in southern Togo. The stores were artificially infected with the LGB, while Tn actively entered the stores on its own. After storage of 8 months, the following losses were recorded:

· in the 1st season 11.6% without Tn and 8.4% with Tn
· in the 2nd season 24.1% without Tn and 15.9% with Tn

While the generally low losses in the 1st season do not permit any conclusions concerning an impact of Tn, a onethird reduction in losses was observed in the 2nd season. This loss reduction was associated with an LGB population that was reduced by 73% compared to stores without Tn.

Since 1994, Schneider (unpublished) has continued to work on the question of the effect of Tn. Schneider is presently examining the spread of the LGB and of Tn, damages, losses and parameters of the Tn-LGB predator-prey system. In addition, economic parameters are being surveyed in order to estimate the benefit of Tn introduction. Schneider stresses that it is not purposeful to make universal statements for entire countries or regions on the losses caused by the LGB or the reductions in these losses associated with the release of Tn. This is because LGB occurrence is erratic, leading to high losses in certain areas and low losses in others. He concludes form the evaluations conducted to date that under certain conditions Tn effects a drop in LGB populations and thus a reduction in storage losses, and that over the longer term, when a stable predator-prey equilibrium has established itself, Tn will prevent a widespread destructive impact of the pest. However, even then, infestation outbreaks will always have to be expected.

PREVIOUS PAGETABLE OF CONTENTSNEXT PAGE