Who makes the decisions in precision farming?
As a precision ag instructor, I try to talk with producers to learn more about their business and technology needs. I’ll ask about how they use precision technology, what’s most valuable for them and what they think about new technology.
Unmanned aerial vehicles, better known as drones, are a good example of a typical conversation. I’ll talk to producers about what they like and what they don’t.
Most are excited about it and agree that it is a cool technology but aren’t exactly sure of its value. Most will also admit that they’re not sure what to do with the imagery data once they have it.
Another example is a sensor network in which a lot of data is collected and transmitted to the home office.
Again, the grower sees how valuable the technology is and agrees that the whole idea is good but is hesitant to get any more data.
This conversation has lately taken place about software. Most of the farm management information systems that are available result in additional data. So again, the main problem I hear is, “I don’t want more data; I have too much already.”
Growers see much of the software that is being marketed to them as analysis software, which creates additional data and no answers.
These growers do see a valuable end result and a lot of tools, but with the mountain of data, there is no path to get there. How do we avoid being inundated with all of the data?
I am going to offer a different perspective.
The problem isn’t too much data or that the software is useless. The problem is who is doing it and that the data isn’t being interpreted for the decision maker.
If the grower is looking at the mountain of data and being overwhelmed by it, then he needs somebody who can analyze and interpret it for him.
The decision maker shouldn’t be tasked with that; the precision farming technician or mapping specialist should be doing it for the grower.
First of all, data should be the basis for a decision in precision farming. If the data is incomplete, then the decision is difficult or incomplete.
We should not turn down data from being collected just because there is too much of it.
As well, we can’t attempt to look at all that data at once to make sense of it.
Analysis summarizes and organizes the data. It may actually increase the amount of data, such as when zones are created or when sensor or sampled data is interpolated to create raster surfaces for pests, nutrients or tissue. Analysis should prepare the data for an explanation, which means interpretation.
Also, interpretation needs to select data applicable to a specific question and put it into context. The grower should ask the questions.
The technician needs to identify the data that applies to the question, and then create an interpretative map that answers the question. Interpretive maps should offer the grower a “no-brainer” view of the decision that needs to be made.
One small example of this process is the assessment of a crop product.
Rather than trying to look at 50 layers of data for some enlightenment, we need a specific question: did this product increase yield and was it economically justified?
For this we need a yield map to determine differences in yield attributable to the product and to calculate income and an as-applied map to determine actual application of the product to various areas of a farm.
Completing a query of yield points from treated areas and control areas of the field and applying a test of significance, such as a T-test, tells us if there was a real and repeatable difference and an increase in yield.
A net profit map uses that difference and the break-even income to show those field areas in which the product paid for itself.
When creating a map such as this, the technician doesn’t just provide more data. Instead, he provides an answer to the grower’s question and a no-brainer decision.
What I hear regularly is that decision makers really want answers rather than just more data.
Just collecting more data isn’t going to do it. Buying another software may not do it and analyzing the snot out of all the data isn’t going to do it.
A systematic approach, in which the decision maker asks focused questions, data is available to answer the questions, software analyzes and creates an interpretative map and a specialist who knows how to do this is hired, will make precision agriculture happen.