There are at least three conceptual problems with crop-loss estimates. The first is the distinction between actual losses by farmers given that certain loss reduction practices (e.g. pesticide application, crop rotation, etc.) were used and potential losses if such practices were not used. This distinction is important because the incorporation of resistance typically will produce both lower actual losses and reduced pesticide use. The reduction in actual losses will not be a full measure of trait values. Potential losses may be a better measure, although pesticide use data should ideally be used.
The second conceptual problem is that crop losses may not be 'additive'. That is, total losses from two or more insect pests may be greater or less than the losses attributed individually to each pest (and similarly for diseases). Crop-loss data are typically attributed to individual pests.
The third conceptual issue is that the incorporation of insect or disease resistance into modern varieties (MVs) may result in the adoption of the higher yielding cultivar in locations where the pests are endemic. Crop losses may not actually be reduced, but yields will have been increased.
A related issue that requires consideration in actual estimates is that the natural incidence of pest and disease pressure varies by location and over time. Variation by location creates a 'left-out' variable problem. Locations with high natural incidence may have high losses even though resistance to pests and diseases is quite valuable. Variation over time means that pest and disease resistance may be of little or no value in some periods and of high value in others.
A recent study of rice crop loss reports loss measures for Indonesia, China and India (Evenson et al., 1996). The Chinese data show actual losses, given existing protection, and potential losses without protection.
This study showed that actual loss levels were relatively low. For disease losses, they were less than 1% in Indonesia (in 1986). They were between 1 and 2% in China. Insect losses were higher in Indonesia and India but not in China, generally being less than 2% (although they were 4.25% in Indonesia before MVs were introduced). Potential losses in China were in the 5-7% range. (The Nepal estimates were in the 15% range.)
In Indonesia, crop losses from insects were higher prior to 1980 when first generation (see below) modern rice varieties were replaced with second and third generation MVs.
Was this article helpful?