Gaps between farmer and attainable yields across sunflower-growing regions of Argentina: Measurement, significance and implications.

A.J. HALL; C.FEOLI; J.INGARAMO; M.BALZARINI

Mar del Plata/Balcarce

Conferencia; 18th International Sunflower Conference; 2012

International Sunflower Association

ABSTRACT ·   Quantifying the magnitude of the gap between yields achieved by farmers and those attainable using good agronomic practice is crucial for prioritisation of research and policy efforts aimed at reducing yield gaps. In this paper we present yield gap estimates obtained from a comparison between selected reporting-district yields (best available estimate for farmer yields) in eight sunflower growing regions of Argentina and those obtained in comparative yield trials in each of those regions. ·   Eight consensus rainfed sunflower growing regions within Argentina were established on the basis of perceived environmental and management characteristics. Databases, covering 5 to 9 years for each region, were constructed using official reporting-district (county) yield values and data from comparative yield trials. Harvested crop area in the sampled reporting-districts averaged 1.2 Mha y-1, or ca. 50% of the national total for the 1999-2007 period. The attainable yield (comparative yield trial) database included 646 trials conducted over the same time period. Attainable yields were only computed for year/region combinations that included more than 5 trials per region, each including data for at least 7 commercial hybrids. Mixed linear models were used to compute Best Linear Unbiased Estimates (BLUEs) of yield for each region/year combination for both farmer and attainable yields. ·   Gaps between farmer (reporting-district) yields and attainable (comparative yield trial) yields were statistically significant (p ≤ 0.05) for all regions and ranged from 0.37 to 1.18 t ha-1 across regions, for a country average of 0.75 t ha-1 , equivalent to 41% of the mean country yield of 1.85 t ha-1 for the 1999-2007 period. In relative (to average farmer yields) terms, the magnitude of the farmer/attainable yield gaps ranged between 32% and 77% in the five regions that contribute 81% of the national crop. Mean yields for the top decile of comparative yield trial data ranged from 3.2 to 4.2 t ha-1 across regions, and the highest yields for this decile in any of the years of record ranged from 3.9 to 4.8 t ha-1. A notable feature of reporting-district and comparative yield trial data was their variability. At reporting-district level within regions, contributions of spatial and temporal variability were roughly similar. The mean relative contribution of the trial effect to non-error variance of the comparative yield trials exceeded 85 % across regions, dominating the contributions of genotype and genotype-by-trial effects. ·    The analysis shows that, for the main sunflower producing regions of Argentina, farmer/attainable yield gaps are substantial and exceed the floor of 25% of farmer yields which has been posited as the minimum achievable gap in other crops under prevailing economic and biophysical constraints. The magnitude of these gaps underlines the need for research into their causes and their reduction to more reasonable relative values. Our estimates of attainable yields provide a more appropriate benchmark against which to gauge farmer performance than the occasional high yields in comparative yield trials.  To our knowledge, this is the first country-wide attempt to quantify yields gaps for sunflower and the only one, for any crop, to use comparative yield trial data combined with appropriate statistical analysis to estimate attainable yields.