Problems with creating and evaluating yield maps
Recording yield maps of your crops when you harvest gives you important information about the variance of yields within individual plots. So, it begs the question as to why this resource isn’t used to customize your material usage to optimize production (e.g. through variable-rate application maps).
To answer that, we must first look at the problem with creating and evaluating the yield maps themselves. For one thing, we have to consider just how accurate of data you are getting; how much can you rely on your thresher giving you accurate information about what it is harvesting? Often, you will have multiple pieces of harvesting equipment operating simultaneously. What happens if one of them doesn’t have a yield recorder? You also need to consider if all of the machines are calibrated the same in order to get a complete and accurate set of data.
In such cases, were the calibration is off, multiple machines can record different yields for the exact same location. And improving calibration is time- and process-intensive operation, one that busy farms will have a hard time managing to make time for. The calibration process includes harvesting an entire bunker, dumping the harvested crops onto a transport, and then carefully weighing and recording the data into the machine. This process then needs to be repeated for every harvesting machine that you are using.
Surely there must be a better way
And now there is. Modern advancements in the field of remote monitoring have resulted in the ability to replace traditional yield mapping with long-term analysis of satellite imagery. Using plots which already had accurate traditional yield maps, this new method was tested and the results matched up almost perfectly. So now there is a much easier way for farmers to analyze their yields, without sacrificing any accuracy.
Left: yield map. Right: a long-term satellite imagery analysis of crop potential