Developing A Strategy For Precision Soil Sampling

 

There are many different tools and approaches available that, if used correctly, can help to improve your nutrient management (variable rate application, precision placement, crop sensing via NDVI, late-season application, nutrient BMPs, etc). However, selecting the correct tools and using them to your advantage is not always an easy process, since the best tool and the best approach can vary by farmer and field. The key to a successful soil fertility program is to identify your goals and develop a plan to meet those goals each season. Identifying both short and long term goals make it possible to develop a strategy to use precision technologies to systematically improve your soil fertility program. Some goals you may consider are:

 

1.     Improve mapping of field variation that affects soil fertility

 

2.     Maximize the economic return of fertilizer applications

 

3.     Reduce off-site movement of nutrients

 

 

One of the most important decisions that you will make as part of your fertility program is how to divide (the area within a field boundary) a field into representative areas and what the area represents yield soil type etc. Currently, there are two widely used methods: grid and zone sampling. Deciding between the two is not as simple as it may seem, since these methods require different sampling techniques, different analysis, and different applications. It is important to keep your fertility program goals in mind when making this decision.

 

 

Grid sampling involves taking samples at regular intervals across the landscape of a field. Grid size is selected to provide the desired data resolution. A 2.5-acre grid size is commonly used (360 by 360 feet); however, choosing a grid size that matches up to spreader equipment widths is recommended. Smaller grids may be necessary to accurately capture differences in fields with a high degree of variation and it may be possible to increase grid size if a field is fairly uniform. Cost increases as the number of samples increase; however, research has shown that smaller grids provide higher resolution, and often more useful, data.

 

 

Grid sampling should be used when there is little information available about the variation in nutrient levels across a field. Grid sampling may be useful in fields where variability is expected but the field history is not well known, topography is uniform but differences in soil type occur, varied management patterns have been used in the past or manure applications have occurred. Proper grid sampling makes it possible to identify variation within a field and is an important data layer when determining future management zones for fertilizer applications.

 

 

The goal when grid sampling is to determine the best estimate of each soil test value near the center of the gridded area. The changes that occur in unsampled areas of the field are then modelled using interpolation to determine a likely pattern of variation. Several different geostatistical models can be used; for example, point kriging, inverse distance, and splines. Studies have concluded that the initial selection of sample number is more important in successfully reflecting actual fertility levels across the landscape than the statistical model used. The interpolation method may vary depending on the software used to generate the prescription; therefore, it is important to check with your consultant before sampling.

 

Figure 1 shows a recommended method for collecting soil cores when grid sampling. Samples at each sample point are collected in a 10-foot circle with a two cores pulled from each quadrant or a total of eight cores.

 

 

 

Zone sampling involves dividing the field into zones that are uniform enough to be managed as a whole and then sampling to determine the average soil test values for those zones. The success of the zone sampling relies on the amount and quality of the data used to determine the zones. Layers such as soils maps, aerial photos, yield maps, topographic maps, management history and personal field experience can provide valuable information about the variation in a field. This information can be used to define sample zones or management zones in a field. As the number of management zones in a field increase, the number of samples needed increase. If only a few zones exist, samples can be combined to reduce the cost of analytical expenses.

 

 

Management zones are a better choice than grids when the operator has a long history of working with the field, topography varies and can be used to define zones, where yield map data over time has defined high and low yielding areas, the soil type map represents yield zones or other remote sensing data is available to overlay with operator experience to define yield patterns in a field.

 

It is important to note that differences in yield may not be always be caused by differences in soil test values. Identifying other yield limiting factors will help fine tune your soil fertility program for each field.

 

 

The goal when sampling by management zone is to determine the best estimate of the entire zone. If the data used to determine the zones is accurate, the soil test values should be relatively consistent. In this case, taking multiple soil cores is necessary to reduce the chance of pulling one from a “bad spot.”

 

Figure 2 shows the recommended pattern for pulling soil cores when zone sampling. Sample points should be taken randomly (recommended to walk in a zigzag pattern) with 10-15 cores per sample area up to 25 acres. Georeferenced sample points may give a better opportunity to compare sample trends over time by returning to near the same point in future years. This can be beneficial to tracking soil fertility recommendation program effects on soil test levels over time.

 

 

Since the soil test values will represent the average for the entire zone, interpolation should not be used. A blanket fertilizer application rate within each zone is most appropriate when zone sampling is used.