A soil sampling protocol to minimise the spatial variability in soil test values in New Zealand hill country

Because of the large inherent changes over time and space, non-systematic s ampling in hill country can result in high coefficients of variation (CV 18 -55%) in soil phosphorus (P), sulphur (S), and potassium (K) levels between measurements. The components of spatial variability were measured on 20 hi ll-country sheep and beef farms. For Olsen P, 38% of the total farm varianc e was measured within 1 m of a fixed point, 27% within 100 m, and 35% betwe en paddocks within a farm block. With soil pH, quick test K, and sulphate S , there was greater variation within 1 m, less within 100 m, and very littl e between paddocks, compared with Olsen P. From this information, a samplin g protocol was designed to achieve the target CV for Olsen P (15-20%). The sampling protocol was tested on 77 hill-country farm blocks. At annual intervals up to 5 years after establishment, 100 m transects from within 3 representative paddocks were re sampled at 10-m intervals from within a 0.3 -m radius of each original sampling position. All soil cores from the 3 tra nsects except those at each end were bulked (n = 27) for one block. Over 4 years on average, the mean CV for change in Olsen P between years was 21.3% . Corresponding CVs for soil pH were 2.2%, soil sulphate S 37.4%, and soil quick test K 26.2%. These results indicate that the use of a scientifically designed and robustly tested soil sampling protocol will minimise the vari ability in measured soil Olsen P,pH, and other soil test nutrient levels in hill country over time and allow more accurate advice to be provided on fe rtiliser nutrient requirements.