Michigan State University Extension
Soils & Soil Management - Fertilizer - 06039704
07/10/97
Tri-state Fertilizer Recommendations for Corn, Soybeans, Wheat and Alfalfa Extension Bulletin E-2567 (New), July 1995
M.L. Vitosh, Michigan State University; J.W. Johnson, The Ohio State University; D.B. Mengel, Purdue University; Co-editors
Four variables are generally considered when taking soil samples:
1. The spatial distribution of samples across the landscape.
2. The depth of sampling.
3. The time of year when samples are taken.
4. How often an area is sampled.
Proper consideration of these variables ensures that the sample accurately reflects the fertility of the area in question and allows for the best possible fertilizer recommendations.
Sample Distribution
Sample distribution usually depends on the degree of variability in a given area. In relatively uniform areas smaller than 25 acres, a composite sample of 20 to 30 cores taken in a random or zigzag manner is usually sufficient. Larger areas are usually subdivided into smaller ones. Non-uniform areas should be subdivided on the basis of obvious differences such as slope position or soil type.
Banding fertilizer creates zones of very high fertility in soils because the fertilizer is mixed with only a small portion of the soil. Samples taken in the band can greatly overestimate the overall fertility of a site. Because the position of fertilizer bands is rarely known with certainty, one should take more random samples than usual in fields with fertilizer bands and vary sampling position with respect to row location to ensure that the bands do not bias test results.
For non-uniform sites, a systematic sampling approach is best. Sampling in a grid pattern can give an idea of variability in a field and fertilizer application can be adjusted according to the distribution of soil test results within the grid. The grid spacing can vary from as little as 30 feet to several hundred feet. Often the grid spacing is some multiple of fertilizer applicator width. Grid geometry can be adjusted to account for characteristics of the site in question. For example, a rectangular grid may be more useful than a square grid when fertilizer applications have been primarily in one direction. Eight to 10 cores are usually taken and combined for analysis at each sampling point in the grid.
Sampling Depth
Soil samples used for nutrient recommendations should be taken at the same depth that is used in the research generating the recommendations, normally 0 to 8 inches. A major exception involves sampling sites subjected to little or no inversion tillage, including those in established forages, no-till and ridges. In such cases, additional samples should be taken at a shallower depth (0 to 4 inches) to assess acidification of the soil surface and make appropriate lime recommendations. Surface soil pH may greatly affect herbicide activity and/or carry-over problems. Occasionally sampling the soil profile in 4-inch increments also may be useful for assessing the degree of nutrient stratification in fields managed with conservation tillage, but no recommendations are being made at this time based on the results of such samples.
Time of Year to Sample
Sampling after harvest in the fall or before planting in the spring is recommended. Fall sampling is preferred if lime applications are anticipated. Sampling during the growing season may give erroneous results due to effects of crop uptake and other processes. In-season sampling should be used only to test soils for nitrate as a guide to sidedressing additional N. Recommendations for sampling soils for nitrate are not consistent across Indiana, Michigan and Ohio, so those interested in such tests should use in-state recommendations.
Sampling should occur at the same time of the year each time a particular field is sampled. This allows better tracking of trends in soil test values over time, which may be as important as the test values themselves.
Intervals Between Sampling
Most sites should be sampled every three to four years. On sites where rapid changes in fertility (particularly decreases) are expected or when high-value crops are involved, shorter sampling intervals (1 to 2 years) are recommended. Regardless of the sampling interval, records of changes in soil test values over time should be kept for each site tested.