Application of Central Composite Design on Assessment and Optimization of Ammonium/Nitrate and Potassium for Hydroponically grown Radish (Raphanus sativus)

Central composite design and response surface methodology were applied to model bulb fresh weight, bulb dry weight, and whole-plant dry weight, based on ammonium/nitrate and potassium in nutrient solution for hydroponically grown radish between November and December in New Zealand. Three high-precision second-order models were developed. The R2 values were more than 0.90, Pre R2 values were reasonable, and Adeq Precision values were more than 10. The contributions of the two factors to the response variables were studied, and optimal nutrition was discussed using these models. Both factors affected the three response variables significantly and nonlinearly (all p values < 0.05). Ammonium/nitrate played a more important role than potassium in bulb fresh weight (p = 0.00 versus p = 0.01). The negative or positive influence of one factor on bulb fresh weight and whole-plant dry weight depended on the concentration of ammonium/nitrate and potassium. Nutrient solutions with ammonium/nitrate 4:11 (0.36) and potassium 5.6 mM, and with ammonium/nitrate 5:10 (0.50) and potassium 6.0 mM were both better than those of Hoagland and Anrnon for hydroponically Summer-grown radish in New Zealand.