S1E1: 2024 Precision Agriculture Peanut Research & Extension Update
Today we hear from Kendall Kirk, the Director of CU-CAT, as he discusses extension and peanut research into the economics of digger blade replacement, low pressure tire technology, soil compaction, and Clemson apps for peanut production.
Date: Feb 22, 2024
Location: Allendale Extension Office
Contact: Kendall Kirk kirk2@clemson.edu
Clemson Windrowed Peanut Drying Forcaster
Calculators & Web Apps
Precision Ag Web Apps Analytics for 2023
Production Credits:
Introduction: Hannah Mikell
Producer: Kevin Royal
Editor: Kayla Peters
Technical: Trey McAlhany
Music Composer: R.M. Davis
Transcript:
[Hannah Mikell]
All right, welcome folks to Cultivate Ag, where you'll find your premier podcast for all things agriculture. We strive to give you a classroom presentation on your farm. I'm your host, Hannah Mikell, agronomy agent with Clemson Extension, and I'm honored to present to you today, Dr. Kendall Kirk, where we will learn more about certain varieties of soil and how that's going to help us deal with our peanut digger blades, as well as any type of wear and tear indicators and different types of configurations that Dr. Kendall is going to give expertise at helping us figure out. At the end, we'll even maybe have a pop quiz for you.
Thank you for tuning in, and we look forward to gaining some valuable insights from Dr. Kendall Kirk. It's time to take the classroom to the cab. Let's cultivate.
[Kendall Kirk]
I'm going to go through a couple of updates. Two events, though, and I only have one in the slides. The other one, I just didn't have time to get it all together. March 20th in Columbia, we're doing a thing. Seating's pretty limited here.
It's 100 folks. It's going to be a series of one-hour panel discussions. The topics listed there are economics, sustainability, technology, and then fruit and vegetable opportunities.
A lot of supply chain discussion in that one. This is something that we've partnered with the folks at the bottom there, Palmetto Agri Business Council, Department of Ag, Farm Bureau, SC State, and Mixon Seed, to put this together and have these panels set up to where you've got solution leaders, in some cases farmers sitting on the panel, and then you've got solution providers. It's intentionally to have a dialogue to look at ways that we can support one another, what are the trends, what are the things we need to be doing that we haven't been doing, that type of thing.
In the link at the QR code, that's a clickable link if you go to it. By the way too, on the QR code, you don't have to download the app. It should have a, continue to a website if you just wanted to follow along or look at the slides later.
One of the things we did this year, supported by the South Carolina Peanut Board, was looking at what's the cost of not changing blades out when you should have changed blades out on your digger? And also, within that, how often should we be changing them out? That's something that hadn't had a lot of attention if you look back in the literature.
It's a question that we've had. In fact, one of the answers we've always had, and it largely comes from talking to you guys, talking to growers and your insights, you know, sitting in the cab, y'all learn a lot looking behind you. One of the things we've learned is when you're digging and those peanut plants start moving inboard on the conveyor, oftentimes they're getting pushed.
And I'm talking about where you got two rows, you know, whatever plant spacing you've got, you've got two rows behind the blades and they start to come up that conveyor. If those rows start to push towards each other, then that's an indication that either you're too shallow or otherwise, for whatever reason, you're pushing plants rather than severing the tap root before you pull them up on that conveyor or as you pull them up on the conveyor. So that's one indicator, right?
They're moving inbound. One thing you can do to address that is run a little bit deeper in some cases. And one of the questions we have is, what's the penalty from doing that?
We've done other tests that say, if you run too deep or otherwise, if you run too shallow, your yield suffers from doing that. So, one of the things that we're looking at here is when we do have really dull blades, what does it do to your yield when you drop your depth a little bit? We had 16 pairs of blades.
The tests we do are on a two-row digger so we can keep plot size small. And we had 16 different pairs of blades, some that had controlled wear that we had placed on them. It's not going to look exactly like your wear, but nobody's wear looks the same as somebody else's wear.
So, in other words, 20 acres on your blades ain't going to be the same as 20 acres on somebody else's on different ground. Or even for you, 20 acres on this ground versus 20 acres on that ground, it's not always going to be the same. That's one of the things we're looking at within this.
And then some of the blades that we had were from growers that they told us how many acres they had on that pair of blades and generally where they had been with those pairs of blades. So, some of these blades in the test had 120, 140 acres. And when I refer to acres on a set of blades, it's digger acres, it's six-row acres is what we've kind of standardized everything to.
It's not that individual blade because each blade goes through a sixth of what the digger goes through technically. But just to standardize it, we've done it that way. I was disappointed.
We saw the results and I said, man. So, what we're looking at here, this is acres on a blade on the x-axis, you know, and then we've got the yield that we saw on the y-axis. Now, this was sandy soil.
This was really light soil. We didn't see a big relationship. And again, I saw this early on.
We had an early test that came off that was only in light soil. And then we saw this and I said, man, that's disappointing. There is no relationship there, alright?
That line, r-squared of 0. 05, that's horrible. There's no trend, there's no relationship.
Maybe it is petering off here, but why do we see this rise here? We need to do more to learn what's happening in sandy soil, but I was excited when I saw the heavy soil results. So, this is in heavier ground.
Among maybe the heavier grounds, this is a clayier soil, but it's still a pretty sandy ground. You would call it peanut ground. And what we did see here is that when we went to blades that had more wear on them, we generally saw that for every acre of wear we put on the blades, yield decreased by 1.
7 pounds. So, in other words, you can look at it as if you had 100 acres on a pair of blades, move that decimal two times, you're 177 pounds off of the yield you might have been on new blades. I caution in this statement to repeat your wear and our wear isn't the same.
Your yield effect from wear and our yield effect from wear is not the same. So, we're trying to build good, solid recommendations on this. It's going to take a minute to sort it all out and do these tests.
High moisture, low moisture. There are so many other things. Maturity, disease.
There are so many other things that can affect what your yield is as a function of what your blade wear is. So that's one of the things we'll be looking at. If we take that same data though from the heavy soil and flip it around, what I showed you before was yield.
So, this is the exact same data here though we're looking at yield loss. And so all we've done is inverted it. There's nothing different on this data.
What that allows us to do is start talking about costs. Again, I'm not ready to say that we know all these numbers for all the conditions. I'm kind of showing you what we learned in our test this year and where we're heading with this.
So, it follows that as we put more acres on a blade, at that 1. 7 pound rate, your yield loss goes up. And so we can translate that to yield loss cost.
I'm using $400 a ton. Nathan tells me I need to start using $500 a ton. I guess I'm still stuck in the old days.
But at any rate, your cost goes up. The longer you wait to change your blades, the more the cost goes up. However, imagine if you changed your blades after five acres every time.
Well, then your blade change costs go way up, right? So, it's not to say that you don't want to wait. You do want to wait some length of time.
The question is, what length of time do you want to wait to? And so this orange line is the blade replacement cost. So, if we've got six individual blades, the cost I used here is $270.
So, the way to think about that orange line, if you changed your blades after 10 acres, it would be $270 divided by 10 acres. That cost, translated to dollars per acre, would be $27 an acre. And that's important because that allows us to put the blade change costs together with the loss costs from lack of blade change.
That's what this is getting at. And so essentially, we can add the blue line here to the orange line. The farther you go out on the orange line, the more you got out of those blades, the more you've distributed their cost over the acres that you've already gone over.
However, all the while, that blue line's creeping up, right? We don't know, is that a linear relationship? There's a lot we need to learn.
The orange line's solid. Blue line, we got a lot to learn. To say, what does it really look like?
This is just an example on this year's data, what it could look like. So, when we add the orange and the blue line together, you get the gray line. This is total cost, yield loss cost plus the blade replacement cost.
And what you see here, and I do believe this trend will exist, whether the quantities are exactly what they are, this trend should exist. We're doing something similar with cotton picker spindles right now and looking at when should you change your cotton picker spindles. That's actually really interesting data on that as well.
So, in this test, these numbers, it suggested that after 40 acres, I have minimized my blade cost, the sum of the peanut loss cost and the blade replacement cost. I'm not telling you to change your blades after 40 acres. I'm saying there is a point at which you have maximized your profitability, right?
Or minimized your loss is another way to look at that. In this case, if you stretched a linear line from the minimum out to the maximum here on this 100 acres is what I've gone to. It's about for every acre you go beyond that minimum point, it's not quite as high as the slope of the blue line.
Remember I told you that was $1. 70, $1. 80.
This one's $1. 30. It's the sum of those two lines.
That orange one keeps decreasing along the way there. And it's roughly linearized by then. Because my wear and your wear is different, because everybody's wear is different, and because everybody's loss as a function of wear is different, we're trying to objectify this.
In other words, we're trying to say, is there something we can measure? Is there some physical relationship of wear that we can indicate to the grower that it's time to make the change, or some kind of decision aid tool? We don't expect that you're going to do this, but we're kind of digging into this and looking at all the things we can look at to say what relationships do exist, so we've got a leaping off point.
This is looking at blade mass. This is blade mass on these blades that we ran in the same ground over the same amount of acreage so that we know it's consistent in the wear, other than moisture, which we couldn't control at the time of digging. But this took course over a week, roughly.
It was a week to take all the sets of blades we had, all the dots on here essentially, and show what it would be. You see blade bevel up and blade bevel down on these results here, and you see that the mass does decrease with time. So, we could say, okay, well, we can just get everybody to take the blades off and weigh them and use that as the decision aid tool.
Well, no, nobody's going to do that. We're not suggesting that you're going to do that to decide, is it time to change your blade? But it is exciting to see that there is some physical measure of the blade that could relate to something non-subjective, something objective that could say, how much wear do you have?
The other thing we're looking at is digital image analysis. This may be something that you could do. Maybe you could take your cell phone out and take a picture of that blade edge and then run it through some image analysis tool to say, how much wear does it have?
And that's what we've done here. We haven't run it through the image analysis tool or begun to build it yet, but this is a brand new blade from KMC. All of these were KMC blades, and this is that same blade after it's got, or out of the same lot, after it's got 240 digger acres on the blade.
So, you can definitely see some differences and maybe an image analysis tool can help us to get at this. There's some other things we're looking at, wear indicators, that type of thing that might be able to help us get at it. Another project that we've worked on, ported by Peanut Board, Cotton Board, and Michelin.
One of the things we're doing at CU-CAT is trying to increase industry collaboration in all that we do and bringing in industry on the various projects we're working on. This involves a lot of cooperation with some of the farmers in our area for the equipment, the commodity boards, and then here also Michelin playing a role here. So, if you saw a tire that looked like that on one of your pieces of equipment, would you do something about that?
Yeah, I had a lot of nods in the room. All right, these tires are designed to look like that. And Jerry Han here - one of the things Jerry's working on is kind of automating the process of setting tire pressure.
That's one of the things that you'll understand better as I get into it. Most of your bigger ag tire manufacturers have this technology, so not just Michelin. Michelin s are called yield bib tires.
They cost a lot more, but I'm going to show you, we're investigating what the value of that is, and I'll show you what we learned in year one. But, basically the tires, they call them high sidewall flex tire technology or ultra-high sidewall flex tire technology, and they're designed to run at much lower pressures than you can normally run at. So, the advantage of that is the footprint.
Essentially your machine, whatever it is, cotton pickers are among the worst, they're 70,000 pounds loaded, these modulating pickers that we've got, and that weight is distributed on six tires. So, the footprint that that tire sits on directly relates to the amount of soil compaction that you are creating as you drive through the field. So, this study is in a peanut cotton rotation, in part to say where that cotton picker went last year, what does it do to your peanuts next year?
But also the planter went through there this year on both the peanuts and the cotton. So, what did that compaction relate to? We're using a strip till planter for all of these tests.
This is the same tire, all right? So, what we do is we lay a piece of plywood or whiteboard down, we park over top of it, and we spray paint all around where the tire is. From this, we can calculate gross flat plate area.
But this is the exact same tire on the pressures you would normally run at, at a tire that you normally run that doesn't have this technology versus the low pressures that are suitable and recommended for these tires with the ultra-high sidewall flex technology. And it's about a 50% increase in footprint that you can apply. So therefore, it's about 50% decrease, depending on how you do the math, I guess, on PSI imparted to the ground, right?
The weight of the machine divided by that footprint is pounds per square inch that you apply to the ground. It's like walking through the grass in high heels versus something with a small diameter heel versus a big diameter heel, right? Who's going to sink in more?
Who's going to compact the ground more? This does the same. These are eight-row tests.
If you picture each green line as a row, just to show you kind of the row configuration. It's been a challenging test. This digger, we don't have a six-row digger on the station.
We don't do six-row tests. We'd never have enough land at the station. We've got 2,400 acres there, 800 acres in row crop, but we'd run out of land if we didn't do two- and four-row tests, generally two-row tests.
So, this is a cooperating grower that loaned us their digger. This is a cooperating grower that loaned us their combine. Logistically, you can imagine, that's the combine he uses to harvest his peanuts.
We need it for this test. It's been a tough one, but we've had a ton of cooperation and couldn't have done it without them, because we couldn't just use any six-row combine. We had to use one equipped with these tires.
So, every plot has got either the high pressure treatment or the low pressure treatment, and it will remain as such year to year on doing this test. So, in other words, what was high pressure this last year in peanuts will be high pressure this next year in cotton, and vice versa as well. Cotton's a little more straightforward.
We are using four-row picker. So, you say, well, that's not going to be the same as my six-row modulating picker, but we took the duals off, right? So, we took that weight and instead of dividing it by six, we're dividing it by four.
So, it really comes really close to the PSI that a 70,000 pound six-row modulating picker will put on the ground. And folks, in some cases, not everybody's running those. There's plenty of people still using a basket picker.
So, you see the effect of where we had the cotton last year on those eight rows. It'll be peanut next year, and then they'll flip. Ben is our - we can't figure out how to do it.
We said, Ben, can you figure something out here? So, he figured out for us on these six-row plots, how do we get yield off of these six- row plots? So, Ben figured out he can open the clean out at the end of the cross auger and rigged us up this little trailer here.
It takes some manual labor to do it, but all the peanuts out of this six- row plot, they'll fall into that little wagon. We dump them off on a tarp and then we've got like a hanging fish scale. Basically, we take that tarp and weigh it.
Now we got a plot yield weight because we got weighing systems like on Dan's combines, right? We got automated weighing systems, but on this grower combine, we needed an alternative without taking a torch to a grower's combine, right? That's oftentimes the torch comes out if we're modifying something.
We look at rut depth, we saw differences. I know rut depth isn't exciting to you, but what it tells us is, yeah, there's differences. The high pressure versus blue on all these is high pressure tires.
Orange is low pressure tires. There were significant differences. This is after planting and this is after the cotton picker went through.
Behind a peanut combine, we didn't see that much difference. In fact, visibly, Ben and I couldn't find the rut depths, the ruts to measure hardly. We would have had to take a measuring tape to go find exactly which tire we're sitting on right here.
The cotton picker, certainly you can see a big difference on where you ran one tire pressure versus another. On yield, we didn't see much on peanuts this year. But remember, only thing that's gone through those plots this year was the tractor pulling the planter.
That was all that went through the plots this year. We didn't anticipate in year one that we would see a thing. What we anticipated was that after we run that cotton picker through there this year, we're going to see a difference next year in peanuts that we put on that same ground.
But even though we didn't expect it, and this is not significant, so I'm always cautious to look at data that's not significant and then suggest that we can make recommendations based on it. But even there, we did see a difference in the cotton high pressure versus low pressure that was consistent between a heavy soil and a light soil where we did these tests. Couple more things real quick.
Some apps - if you go to Clemson, you can search this. There's a QR code on the next slide. But if you just search Clemson agriculture calculators or Clemson precision ag calculators, you'll find these things.
They're all on the internet. So, this isn't like an app that you go to the app store to get. We put them on the web server.
Trey's a tremendous help with these things, putting the apps together and getting them out there to where all of you can access them. In fact, Trey is the one that brought this brand new one to life. This is one that we're trying to improve on the algorithms.
There hadn't been a lot of work in predicting peanut moisture content, kernel moisture content after the peanuts have been in the windrow over the length of time. If you dig your peanuts today, how long is it before you're going to be able to combine? That's essentially what this one looks like.
So, if you go to that website there, if you go to the webpage and find it or search this, Dylan Burkett's the graduate student and Aaron Turner's ag engineer on it. My role is really small in this. But one of the things we can do is look at frost prediction, that red line there.
You can say, we're going to have a frost event at a time that my kernel moisture is still above some critical threshold. They say if it can rattle, that's one of the recommendations. Don't dig 72 hours before a frost or otherwise, if it rattles in the hole, then it's dry enough that you won't likely get frost damage in a frost event.
So, we're trying to see if we can objectify that too. Can we give you better recommendations on that? That's one of the things this tool does.
But the purple line predicts your kernel moisture content for the next 14 days and where it'll be based on the weather forecast. That's how it's doing it. There's a lot of other peanut related apps.
I'm not going to spend any time on it. Some of these actually came out of discussions at this meeting where people said, man, we need a tool to do this. And so we listen when y'all tell us you need a tool.
Please let us know if you've got things that aren't out there. We don't try to recreate things that other people have created. If it's out there from somebody else, use that.
But if there's tools that aren't out there, we want to make them for you if we can and happy to do that. Over 100,000 page views on these calculators the last year. These are 2023 numbers.
Here's one you can use right now, EMC calculator. Put in your zip code. It'll tell you whether or not to run your grain bin fan.
Coming down the line, peanut maturity app. Instead of counting pods you'll be able to take a picture of it. This is one of the things that Virginia has worked on, on speeding this thing up by about a factor of 20.
She's figured out that we can do it at a lower resolution is the long story short on this. And then this is the last thing. One of the things that we're just getting into is, can you take a picture of cotton or peanuts before you harvest them and make a yield estimate?
But there's been a lot of work on this on aerial imagery, drone imagery. In fact, we did a lot of work early on with aerial and drone imagery. And one of the things we realized is everybody's got this in their pocket.
You all got this. So, if we can put it on here, it's a much more accessible tool for everybody. Quick pop quiz.
If you had to guess, this is not a trick question, point. Which one's a higher yield? Or just come up with the answer in your mind.
If you guessed that one, you're right. So, we took a bunch of pictures of cotton plots, peanut plots, this is after you've dug the peanuts, and used those to predict yields. That's what this project gets at.
Brand new. This isn't an app that's out there yet. We need to refine the algorithm, make sure it works first.
But this is a tougher question. You could have told me which one was a higher yield. Could you tell me, for the left and the right, which one or what is the yield of each of those?
Mike Jones, who does plots down there at Pee Dee REC for cotton, he can. I've ridden in the picker with him. He knows pretty well within a couple hundred pounds, what's the yield of that plot.
He's been doing it so long. So, if he can do it, I think we can make an algorithm that can do it for the cell phone. Just to show you about how accurate it is, these are randomly selected pictures.
The algorithm predicted yield versus actual yield. I swear to you, I did not hand pick this one as being the - I ran a random number generator and sorted them. And I said, give me the top four based on a random number generator.
That's how I pick the ones to show you. This one's not as good. Eleven and one-half percent error.
We took pictures at all different angles. We need to do different times of day, different cameras. We did two different varieties here, tightly held and loosely held like showy cotton.
Another one, this is 16 percent error. That's about as bad as it got in cotton. What do we use this for?
I don't know. Insurance claims, looking at your neighbor's cotton, looking over the fence. I don't know what we end up using this for.
One of the things I know we'll use it for is in research. If it's reliable enough, we can increase the amount of research data we can generate with a tool like this, especially on farm trials. You know, if we're on the station, this doesn't replace the real way of measuring yield.
This is not real. Right? Or this is not as absolute as putting it on a scale.
Peanuts, same story. Virginias, runners, if you had to guess, which would it be? That one was higher.
That was a tough one because it's runners and Virginias in the same picture. So, what is the yield though on these Virginias? What would you guess?
Come up with a number in your head. If you guessed 5,800 pounds, you were right. The model said 5,200.
So, we got some work to do. The algorithms got runners and Virginias combined right now. So, I think we can improve these models when we start doing that.
On the runners, if you had to guess, 3, 2, 1, 4,800, 4,900 pounds on this one. And the model did pretty decent on this one. Again, those were randomly selected pictures.
[Hannah Mikell]
Wow, Kendall, that was some insightful stuff on all peanut related and even cotton in some cases. We look forward to seeing what else he has available for us in the future. Remember, keep it tuned in here to Cultivate Ag where we provide you with all insightful and educational material related to all things South Carolina agriculture.
Thank you and we hope to see you next time where we learn more from our experts in the field.