Nitrogen and irrigation effects on yield and quality of potatoes (Solanum tuberosum L.)

The link between potato yield and quality was examined using experiments planted in 1985/86 (Experiment 1), 1987/88 (Experiment 2) and two planted in 1993/94 (Experiment 3 and 4) with the cultivar Ilam Hardy. The main aim of this study was to relate yield and quality so that the crop could be better managed to produce high yields in the desired size grade, and of consistent quality for processing. Total yields averaged 61.3, 39.1, 76.6 and 80.7 tha⁻¹ for Experiments 1-4 respectively. The lower yields in Experiment 2 were associated with high moisture stress prior to tuber initiation. Tuber initiation occurred 21-25 days after emergence in all experiments. Experiment 3 and 4 experienced water stress about 10 weeks after tuber initiation, but there was no yield response to applied irrigation. Thus, Ilam Hardy is sensitive to moisture stress early in growth, but not late in growth. Tuber number and yield were shown to be interdependent, so that low yielding crops had lower number of tubers. Tuber numbers were reduced by moisture stress, and 98% of the seasonal variation in tuber number >10g was explained by the maximum potential soil moisture deficit experienced before tuber initiation. Maximum ground cover was only 85% with high moisture stress prior to tuber initiation, but >95% with late season moisture stress. The maximum canopy cover, which occurred 30 days after tuber initiation, and canopy duration were dependent on the canopy size at tuber initiation. The yield response to nitrogen fertiliser depended on the amount of moisture stress experienced before tuber initiation, and not the soil mineral-nitrogen content. The optimum rate of nitrogen therefore, depends on the amount of moisture stress early in the season. For growers with irrigation it was recommended that nitrogen be applied at 200 kgNha⁻¹, and irrigation applied to keep water stress at a minimum in the early stages of growth and during the main phase of tuber bulking. Variation in grade yields depended on the number of tubers that fell into those grades, but the average tuber weight in the grades did not change. Therefore, tuber number in discrete weight grades was used to fit a distribution, rather than yields. Tuber numbers were normally distributed during growth, and had parameters of mean tuber weight and standard error. The mean and standard error of the tuber number distribution increased with nitrogen and irrigation. There was a linear relationship between the mean and standard error, which varied with season, depending on the amount of moisture stress experienced prior to tuber initiation. Therefore, the CV of the distribution is set early in growth, so the best way of altering the proportion .of yield in a grade size is to change the mean of the distribution. The mean tuber weight of the distribution fitted with tuber numbers was directly equivalent to average tuber weight (yield/number). To examine factors which influenced the increase in mean tuber weight, a model was developed using equations to predict tuber numbers >10g and fresh yields. Mean tuber weight was shown to be dependent on the growth rate of the crop, the duration of growth expressed in thermal time units, and the number of tubers. Changes in these three parameters explained differences in distribution between seasons, and calculated values of the mean and standard error were within 10% of observed values. The tuber growth rate was partly dependent on the number of tubers set, but also appeared to be influenced by the timing and severity of moisture stress during the main phase of tuber bulking. Nitrogen applications did not affect tuber growth rate, but did increase the duration of growth. Prolonging the duration of growth with applied nitrogen seemed to have the greatest effect on increasing the mean tuber weight of the distribution. Management of nitrogen fertiliser is therefore important to control mean tuber weight. The ratio of petiole nitrate to nitrogen content per unit area leaf, measured 1 week after tuber initiation, ranged from 300-390 and was found to be related to the duration of growth. At a ratio of 300, the duration of growth was 620°Cd and increased to 837°Cd at a ratio of 370. Further increasing the ratio to 390 did not increase growth duration, suggesting a limit on the duration of growth for Ilam Hardy. With optimum irrigation available to produce a good canopy cover by tuber initiation, an application of 200 kgNha⁻¹ resulted in a ratio of 370. Nitrogen side-dressings applied 2 and 4 weeks after tuber initiation did not further increase the duration of growth, or yield. Therefore, measurements of the ratio at tuber initiation will give an indication of the likely duration of growth, but correcting of nitrogen status will need to occur before tuber initiation, which occurs about 21 days after emergence. Changes in the rate of increase of the mean tuber weight of the distribution were found to be related to changes in the DM% of tubers, an important quality criteriion. This relationship formalised a link between yield and quality. Changes in DM% were associated with changes in culinary quality of sloughing and mealiness, but not crispness. Consumer preference tests of French fries showed that a combination of crisp and moist fries were preferred. Crispness was decreased with nitrogen applications greater than 200 kgNha⁻¹. Moistness was improved by late season irrigation, but since yield was not increased, late irrigation cannot be recommended unless premiums are paid for culinary quality. The appropriate DM% for the desired moistness was calculated to be 19.2-21.9%. Based on these relationships, a management approach to control yield and quality by controlling factors influencing the tuber distribution is proposed. The approach would enable growers to produce crops of the required quality for the market, with more efficient use of resources. This approach also enables processors to monitor crops and predict harvest date, tonnage and quality characteristics.