Objectives Content of Olsen-P in soil is an important basis for phosphate fertilizer application on cotton. Olsen-P in cotton soils in Xinjiang has increased greatly because of the application of phosphate fertilizers for many years. The old standard was used to evaluate the Olsen-P status in cotton soils, and the recommended rates of P fertilizer could not adapt to cotton production. Therefore, the critical level of soil Olsen-P was determined to provide theoretical basis for managing phosphate fertilizers on cotton.

Methods Relationship between the relative yield and soil Olsen-P was established basing on 281 test sites in Xinjiang, and a logarithmic equation model was obtained. The relative yield was taken into the corresponding logarithmic equation, and the corresponding soil Olsen-P classification indexes value were obtained. At the same time, the critical level of soil Olsen-P was calculated by linear + platform model fitting. Phosphate fertilizer was recommended according to the Olsen-P deficiency index.

Results The correlation between cotton yields and soil Olsen-P contents was regressed with linear, exponential and logarithmic models. All the correlation coefficients of the 3 regression equations reached 1% significant level, with the R² of 0.500, 0.470 and 0.590, respectively. The logarithmic regression equation gave the highest correlation. The relationship between relative cotton yield and soil Olsen-P content was fitted by linear + platform model. The response of relative seed cotton yield to soil Olsen-P content was divided into two stages. The inflection point was the critical level of soil Olsen-P, and the threshold of soil Olsen-P was 22.0 mg/kg. When soil Olsen-P ≥ 22.0 mg/kg, y = 93.77, showed that relative yield of cotton was constant with the increase of soil Olsen-P, and application of phosphate fertilizer almost did not increase the yield. When the soil Olsen-P < 22.0 mg/kg, y = 62.86 + 1.405x, phosphorus fertilizer could promote cotton yield (R² = 0.63**). The phosphorus utilization rate of cotton decreased with the increase of soil Olsen-P. When the soil Olsen-P < 5 mg/kg, phosphate fertilizer utilization rates had no significant difference between flood and drip irrigation, which were 25.7% and 21.6%, respectively. When the soil Olsen-P was 5‒12 mg/kg, the phosphate fertilizer utilization rates of flood and drip irrigation were not significantly different. When the soil Olsen-P was 12‒25 mg/kg, the phosphorus utilization rate of drip irrigation was 15.0%, which significantly increased by 4.6 percentage points compared with that of flood irrigation. When the soil Olsen-P was 25‒38 mg/kg and > 38 mg/kg, phosphate fertilizer utilization rates were not significantly different between flood and drip irrigation. Average utilization rate of phosphate fertilizer was about 15.9% under flood and drip irrigation.

Conclusions The Olsen-P contents in cotton soils could be divided into 5 grades: extremely low, low, medium, high and extremely high, corresponding to the relative cotton yield of < 70%, 70%‒80%, 80%‒90%, 90%‒95%, > 95%. The critical level of Olsen-P in soil is 22.0 mg/kg. The average phosphorous utilization rate is 15.2% under flooding condition, and 17.1% under drip irrigation.