Random Composite Sampling
Random composite soil sampling involves taking samples in a random pattern across a field, while avoiding unusual or problem soil areas. This strategy is most appropriate for fields less than 30 ha (80 ac), that have been uniformly cropped in the recent past and have little natural variation. This is the most common method of sampling presently used in Alberta.
For random soil testing in Alberta, we collect cores from 15 to 20 sites and separate each core by depth to obtain representative bulk samples for each depth.
Directed Random Composite Soil Sampling
Directed or managed random soil testing in Alberta is a modified version of a random soil sampling strategy . This pattern is suited to fields or areas where it is difficult to identify a single dominant area that would represent most of the field. Sub-divide the field into management zones based on unique characteristics. For instance, if there are noticeable differences in yield throughout a field, management zones might be comprised of below average, average and above-average yielding areas. Take 15 to 20 cores randomly from each management zone. A single field may require several bulk samples depending on the number of management zones. This strategy might also be appropriate for areas with more than one soil type, fields with hummocky (rolling) landforms, and fields under strip-crop management.
Directed Benchmark Sampling
Directed benchmark soil sampling is a variation on the benchmark technique. It involves establishing multiple benchmark areas and management zones, based on topography or other characteristics. This strategy can be used when major areas within fields have distinct and well-defined features related to moisture (e.g., texture, slope). Management zones can be identified using soil surveys, detailed elevation mapping, aerial black and white photographs, yield maps or remote sensed images.
Benchmark sampling involves selecting a small (30 m by 30 m) representative site on a field. This site is used as a guide for fertilizing that entire field. Select probe sampling sites in a grid pattern within the benchmark area and prepare a composite sample for each soil depth. Sampling from the same small area each year reduces sampling variability and better reflects changes in soil nutrient level from year to year. Benchmark sampling sites should be marked with a GPS or by other means.
Grid sampling is the most intense and expensive sampling strategy. It uses a systematic method to reveal fertility patterns and assumes there is no topographic reason for fertility patterns to vary within a field. For grid sampling, a field is divided into small areas or blocks. A sample location within each block (e.g., the center point) is sampled 3 to 10 times. Sampling frequency may range from one sample from each 60 m ~ 60 m (0.5 ac) area of the field to one sample from each 2 ha (5 ac) of the field. In general, the smaller the sampling unit, the greater the accuracy.