Using Soil Potassium Adsorption and Yield Response Models to Determine Potassium Fertilizer Rates for Potato Crop on a Calcareous Soil in Pakistan

Ustochrept soil was collected from a major potato-growing area in Pakistan for a potassium (K) adsorption isotherm experiment. Adsorption data were fitted to Freundlich and Langmuir adsorption models. Results showed that the Freundlich model (R²?=?0.96**) fit the data better than did the Langmuir model. Fertilizer rates were calculated based on the Freundlich model and targeted solution K levels at 0, 3, 6, 9, 12, 15, 18, 21, 24, and 27 mg K L^?1. A field experiment was then conducted on the soil to assess the effect of various soil solution K levels (0–27 mg L^?1, with K fertilizer rates at 0, 24, 49, 75, 101, 128, 155, 182, 210, and 237 kg ha^?1), on tuber yield and quality along with 300 kg N and 250 kg P₂O₅ ha^?1 as basal doses. Yield response models (linear plus plateau, quadratic, square root, quadratic plus plateau, and exponential) were used to calculate the optimal fertilizer rate for potato crop. Linear plus plateau model fit the data with less bias than the other models. There was a significant effect of K use on the yield and quality of potatoes. Potassium fertilizer application at 130 kg K ha^?1, which is equivalent to a soil solution level of 12 mg K L^?1, maximized the tuber yield of potato. However, for the improvement in tuber dry matter, reducing sugars, protein contents, and starch contents, the soil solution K level required was as high as14.62 mg L^?1 (157 kg ha^?1). Even greater rate of K, 17.74 mg L^?1 (190 kg ha^?1), was needed to maximize vitamin C content in potato.