Modelling the impacts of in-field soil and irrigation variability on onion yield

摘要

Globally, onion (Allium cepa L.) represents an extremely important crop in termsof production, value and consumption. Similarly, in the UK onion production isconsidered to be one of the most important high-value field vegetables, with ca.300,900 tonnes being produced from 8,448 ha (DEFRA 2010). However, agreat variability in onion productivity (yield) has been identified due to acombination of environmental, genotypic, management and agronomic factors.The increasing demand for high quality vegetables and their supply year roundis adding significant pressure on farming enterprises, which add to thechallenges UK onion producers already face (e.g. crop management, irrigationand pest control decision-making).The aim of this research was to assess the impacts of in-field soil and irrigationvariability on onion yield and quality. Therefore, the scientific evidence on therelationships between onion yield, crop water use, irrigation and crop qualitywere initially reviewed and the evidence corroborated with data from an industrysurvey. In order to evaluate the effects of soils and irrigation variability on yield,under different agroclimatic conditions, a crop growth model (AquaCrop) wascalibrated and then validated using experimental field data. The scientificevidence in the literature and results from the industry survey were used tovalidate and calibrate the AquaCrop model for brown onion (cv Arthur).Statistical analyses were used to assess crop model goodness of fit. A series ofscenario were then defined and the AquaCrop model used to assess theimpacts of different onion cropping practices, production areas and typical andextreme climatic conditions on crop yield.The effects of irrigation non-uniformity (typical of a boom and linear moveirrigation application system) on production were assessed under a series ofagroclimatic conditions (five different years) and two contrasting soil types(sandy and sandy loam). The simulations showed that the lowest yield (8.6 tDM/ha) and greatest variability (standard deviation: 0.23 t DM/ha) occurredunder the driest agroclimatic conditions. Production on sandy soils resulted inhigher yield (in average 0.24t DM/ha) than on a sandy loam soil. The yieldunder hosereels fitted with booms were statistically significant (Kruskal-Wallisanalysis) lower than for the linear move, although the difference was very small(average of 9.52 t DM/ha vs. 9.56 t DM/ha). Under ‘average dry’ conditions, thehighest yield was produced on sandy soils (8.78 t DM/ha), contrary to ‘average’agroclimatic conditions, where the highest yield was produced on sandy loamsoils (9.55 t DM/ha). For the driest season, the effects of irrigation variabilitywere only significant on sandy soils (8.80 t DM/ha and 8.73 t DM/ha forhosereel fitted with linear move and boom, respectively). The study of uniformversus non-uniform irrigation applications showed that onion yield was higherunder uniform irrigation. The differences between yields produced underuniform and non-uniform irrigation increased with increasing climatic aridity(0.01-0.18 t DM/ha compared to average values). Differences were greater incases of boom application systems. Onion yield generated by simulations ofuniform conditions fell within the range found in the literature. The variabilityobserved under non-uniform irrigation was the same (up to 30-40%) as theoverall variation reported by growers.