Soil phosphorus variability in pastures: implications for sampling and environmental management strategies

Soil phosphorus (P) is an increasingly important consideration in the development of P-based nutrient management strategies. The objectives of this study were to (i) obtain baseline information on soil P variability in pastures amended with animal waste, (ii) examine if current sampling recommendations related to the number of subsamples adequately reduce uncertainty to acceptable limits, and (iii) examine the implications of uncertainty in soil P estimates on implementing a soil P threshold of 150 mg kg(-1). Grid soil samples were collected from 12 pastures. Soil P was determined using Mehlich 3 extractant and an inductively coupled argon plasma spectrometer. The arithmetic mean of soil P ranged from 7 mg kg(-1) in a pasture never amended with animal manure to 437 mg kg(-1) in a pasture that had been annually treated long term with poultry litter. Variance of soil P generally increased with mean soil P. The mean standard deviation of all pastures was one-third of the 150 mg kg(-1) threshold. This study points out that smaller variances associated with mean soil P values that approach, but do not exceed, the threshold can influence estimates of soil P. In turn, management decisions could inappropriately change. When a uniform acceptance criteria (within 15 mg kg(-1)) with respect to measured means was used, the required minimum number of subsamples increased with measured standard deviation. The results of this study imply that following soil-sampling recommendations is critical to obtaining trustworthy measures of central tendency, especially in pastures approaching but not exceeding the 150 mg kg(-1) threshold.