The yield gap of mango in Phrao, Thailand, as investigated through comparative performance evaluation

Abstract

Farmers in Phrao, north Thailand, have often, on a “trial-and-error” basis, planted mango ( Mangifera indica L.) in orchards composed of mango, lychee ( Litchi chinensis Sonn.) and longan ( Dimocarpus longan Loureiro). This met with varying success. In 1993, a comparative performance analysis (CPA) of 45 orchards containing mango was done to identify land and management aspects that condition the level of the mango productivity. The orchards were often situated on podzolic soils on hills, footslopes, and terraces that dry out deeply during the dry season. Yields were expressed in farm-gate prices since middlemen purchased the produce from farmers “on the tree”. With many orchards having “low” yields and 18 having “0” yield, the yield data had a loglinear distribution. Using data from all sites, a final model that estimates Ln(yield+1) was derived; it quantifies contributions to the total yield gap for each identified specific yield constraint. It suggests that yields increased if: (i) it was not an “off” year (caused by the biennial bearing behaviour of mango; use of growth regulators may remedy this); (ii) the orchard was situated on a hill or on soils with a relatively high pH or poor water holding capacity (mostly shallow soils with SCL topsoil; water stress causes crop dormancy and induces flower initiation); (iii) the possibility existed to apply supplemental irrigation water (orchards having a growth flush or in a fruit bearing stage require adequate water management possibly including supplemental irrigation); (iv) in established orchards weeding by tractor was practised (this causes root pruning that affects the trees physiological cycle); (v) pruning was practised (this was normally done to remove branches damaged by stem boring caterpillars, all orchards suffered from this serious problem); (vi) spraying by motor sprayer was done that dispenses pesticides, preferably Azodrin (monocrotophos), deep into the canopy. Based on data covering one production season only, the model suggests that environmental factors (location and pH) account for some 30% of the yield gap defined by the difference of the average production situation with the anticipated best one, that management factors account for 49% and the year effect (species attribute) for 21%. Management of mango orchards requires use of up-to-date technology since responses provide exponential returns. It not only demands that farmers are knowledgeable and experienced but also that a well-informed extension service collaborates closely with researchers.