|
Colby, Kansas | The answer to your question depends on how much potential error in your results you are willing to tolerate. The more cores in the sample the better job you will do of overcoming small scale variability due to natural soil processes, band application, livestock grazing, etc. Here 15 cores seems to be the point of diminishing returns both from a data quality and an economic standpoint. Below shows the relationship between confidence interval around your lab result vs. the number of cores in the sample.
 (SoilTestP CI vs Cores.jpg)
Attachments
----------------
 SoilTestP CI vs Cores.jpg (60KB - 37 downloads)
| |
|
Starting to do my own soil sampling
