Yield data has developed a bad reputation for being difficult to obtain, inaccurate and time consuming. It does not have to be that way. There are many simple things you can do before and during harvest season to get better yield maps with out the hassle.
First, ask yourself what you want to use your yield data for. If you just want to get a better feel for how your fields preform, imperfect data will still give you a good idea about the variability of your fields and farm. If you are looking for more detail like evaluating an application trial or building a profit map, you will want to put more effort into making sure your data is accurate.
Where to start? One of the easiest things you can do that has a huge impact on quality yield data is to make sure you properly name all your fields and change the crop type accordingly. Most combine displays can be preprogrammed with set up files that will allow you prepopulate your field names and crops. This is a great solution if you are running multiple combines and want to make sure every operator is using the same field name.
It is important to calibrate your combines properly for each crop at the beginning of the season. Talk to your dealer for the best practices and settings for your machines. Using multiple loads to calibrate and choosing areas of the field with different flow rates will increase your accuracy. Make sure to check the flow delay. Its as simple as using a stopwatch to track grain flow through the combine.
If you must make changes to the settings and calibration try to do it between fields. If the yield is inaccurate but consistent it can always be fixed post harvest. Data that has been recalibrated mid field is very difficult to clean.
Download your yield data after the first field or two to make sure your yield monitors are recording, and you can see the data. If you are using telematics to transfer data to a software platform, log in and look at it to make sure data is uploading as it should. You do not want to be downloading your data at the end of the season only realize nothing recorded.
If using multiple combines, try to have them run side by side when possible. This way if there is an issue with one machine data can be filled in with reasonable accuracy and you are not left with holes in the map.
When it comes to test strips or field trials try combine them with a single calibrated combine. Make sure to harvest the same direction as the test strip and never across it. For accurate post harvest comparison have that same combine do at least two passes around the test strip.
At CropPro, the plan for fall of 2020 is to use the yield maps to determine the relative difference in yield between zones, then take this information to make more informed decision on what target yields should be. In the future, multi-year maps and comparison of multiple years of data can tell us, for example, that zone 1 has never yielded above 30bu/ac. Why? Do we know what the limiting factor is and can we fix it? If the answer is no: why are we targeting 45? Or we find that the yield in each zone is actually similar but the limiting factor between zones 10 and 1 are completely different so even though we are targeting the same amount of bushels we are treating the zones with different nutrients.
You could use absolutely perfect data to estimate your removals for crop inputs but I have never seen a map with enough accuracy to do that. To clean the yield data to the point that it is useable for basing input decisions would take longer than soil sampling.
An agronomist or data specialist that has experience with yield data cleaning can make even the worst maps usable. Reach out to us if you have more questions about yield data cleaning and analysis.
Danielle Epp
Yield Specialist
danielle@croprecords.com