Ep. 25: A Look Into SWAT Map Technology with Cory Willness
Before we get into the fields this year, we wanted to introduce another approach to generating your maps. This week on Precision Points, we rounded out our conversation about variable-rate maps with a discussion with Cory Willness of Croptimistic Technology about SWAT Maps.
SWAT – which stands for soil, water and topography – maps are focused on site-specific management of soil-applied inputs. The process is patented and unique to each field, bringing several layers together and then determining how those layers can create a comprehensive look at your field.
“It's a map that incorporates all kinds of base layers,” started Cory. “For example, like soil color and electrical conductivity and elevation topography and water models, and it builds it all into one single soil type map.”
These soil-based zone maps look at factors like soil salinity, texture, organic matter and topsoil depth. The maps also take into consideration how water moves across the field and your field's topography impact. Using electrical conductivity is a key part of creating the maps because the device they use – a hardware product called SWAT Box – helps collect this data at multiple depths.
“Well, one of the challenges with soil-based precision ag is just it doesn't scale very well. When you want to get soil maps, at some point in time you've got to go out there in the field and then manually collect the data,” said Cory. “So we built a product called a SWAT Box, and it's basically a Geonics CM-38, or a geo-prospectors topsoil mapper sensor inside a box that collects the data remotely. So you can attach it to a piece of farming equipment, like you could attach it to an air cart or a planter or a seed drill, you can attach it to sprayer booms, and it has no operator interface. You just plug it into power and it'll automatically send all the measurements to our servers in real time. And then the staff can build your maps from there.”
All of the different layers are processed by an agronomist and, from there, the field is ground truthed. Once the map is created, it can be used to assist in any soil-applied input like seed or fertilizer. SWAT Maps have a process that is 50% technology and 50% agronomist. This ground-level specialization provides a unique experience.
If you’re interested in learning more about SWAT Maps solutions, you can visit SWATMaps.com. Cory and his team are also very active on Twitter. To listen to our full conversation go to Precision Points.
Have you tried SWAT Maps? Leave a review here.
Host: Morgan Seger
Guest: Cory Willness
Morgan Seger (01:17):
Welcome back to Precision Points, an ag tech podcast from precisionagreviews.com. I'm your host, Morgan Seger, And in each episode we strive to bring you unbiased ag tech information and ideas. And on today's show I'm joined by Cory Willness of SWAT Maps. And SWAT stands for soil, water and topography maps. So this is a patented process that they use to create maps to help growers better understand and visualize their fields, and then use them to create variable rate seeding and nutrient recommendations. As we are approaching spring, I thought it would be great to have another perspective on how to build variable rate seeding maps. So here's my interview with Cory Willness.
Morgan Seger (02:01):
Welcome back to Precision Points. Today I am joined by Cory Willness of SWAT Maps. Cory, welcome to Precision Points.
Cory Willness (02:09):
Well, thank you very much for having me.
Morgan Seger (02:11):
So I'm excited to talk about SWAT Maps and what you can do with this mapping program or service, but before we dive into it, could you introduce yourself to our audience and just give them a little bit of your background?
Cory Willness (02:23):
Sure. Yeah, I'm in central Saskatchewan, I was raised on a farm, got my degree in agriculture from the University of Saskatchewan, spent eight years as a retail agronomist. And then, in 2003, I started my own consulting company doing scouting and fertility services for farmers. That company has grown into about 30 staff now. And then, three years ago, we became operational on an ag tech company that's called Croptomistic, And this company is taking the SWAT Maps variable rates service brand to international partners.
Morgan Seger (03:04):
Interesting. So we were able to connect on Twitter, and I noticed from your profile that you have not only this consulting, but you're the president of SWAT Maps, and you're going back to MIT to take some classes on digital transformation. So you seem really busy. Is there one area that you're kind of focusing on right now? I know we're just starting to get the planners into the field; what type of information should they be thinking about right now?
Cory Willness (03:30):
Well, I guess you could say I've moved on from the things that I enjoy, and I think most people enjoy, is working in the fields and working directly with the farmers on putting down prescriptions and understanding ways to site specifically, apply things, and production-related stuff at planting season. And now, in my current role, I'm more of a person that's wearing all the hats, juggling a lot of balls, kind of human resources to tech projects to sales and marketing. And we run a smart farm here too. That's kind of the only field I get into it anymore, is this little smart farm that we have here that we do tours and research and stuff on. So that's kind of what keeps me busy.
Morgan Seger (04:21):
Got you. Got you. Well, can you kind of kick us off here by telling our audience what SWAT Maps are?
Cory Willness (04:29):
SWAT Maps stands for soil, water and topography maps. It's the world's favorite soil map. It's a foundational piece to doing site-specific management for any type of inputs that you'd be applying in the soil. So whether that's fertilizer, seed varieties and rates, it could be soil amendments, soil-applied pesticides for various things. So it's a map that incorporates all kinds of base layers, for example like soil color and electrical conductivity and elevation topography and water models, and it builds it all into one single soil type map.
Morgan Seger (05:11):
Okay. So is this information or data that you and your team are collecting, or is this something that growers would provide to you that you then put together? Or how would a grower engage with this?
Cory Willness (05:23):
Well, typically the business model is through a business partner. So the software and hardware solutions that we provide to people to do this, there are a few farmers that will invest in it and have the expertise and desire to do it themselves, but we generally find that probably 95% of farmers, they would rather hire it out to a third party, all their very technical site-specific management stuff, just because there's a lot of specific skills required for it. So 95% of the time we would be working with a ag retailer or an agronomy firm or some type of a group of agronomists that would take this technology and then utilize it in their own business to take it to a farmer.
Morgan Seger (06:15):
Okay. That makes sense. And then, as they are getting these maps back, so say they work with a retailer and they have this service provided to them, what benefits are they seeing from using SWAT Map versus like a conventional map that they might've built off of yield data or something like that?
Cory Willness (06:31):
Yeah, I would say that definitely it's the best product fit for the soil-applied products. So you would start right at when you think of soil sampling. We don't recommend people soil sample from a yield goal, a yield potential, a yield map or satellite image, because it's very common for some areas of the field to yield the same. Like maybe the top of a hill is yielding the same as a depression that's slightly flooded out, and well, that doesn't mean that they're a zone. Like you're not going to go out there with your soil testing truck and go to a top of a hill and have this dry thin topsoil and low organic matter and maybe a coarse texture, and then go down to the deep black dirt that's high in organic matter and you can get smears, you can hardly get it out of the probe, and it's wet; they're two totally different things.
Cory Willness (07:24):
So yield doesn't really define soil response and how nutrients move in the landscape. And when you start out soil sampling and you start to think of it like that, they say, "Well, when I'm going to go and apply fertilizer," well, the response is soil related. You're putting seed and fertilizer in the soil and it's related to the water, it's related to the soil. And if it's deep black, high organic matter soils, or thin top soils that are low organic matter and dry, that's how your fertilizer response is generated. So yeah, that's typically what it would be used for.
Morgan Seger (08:04):
Yeah. And I mean, I have the same argument a lot, I was actually just writing a blog post on it, that you can't just go off of ranking productivity because you have to understand why, right? The underlying reason as to why the productivity is the way it is.
Morgan Seger (08:18):
So as you're bringing all of these layers together, I saw that it was a patented process, so I don't know if you can share this or not, but are they weighted? Is there any importance on a certain piece of information, or is it just the way that they come together that makes them reflect the field so well?
Cory Willness (08:36):
Yeah, that's a great question. So it depends on the type of soils that you're working with. Like if soils, let's say, have a lot of texture variability; sometimes you have sands in certain areas and you go to other parts of the field and there's clay and the sub soils are heavy clay. Then that's where having electrical conductivity layers is really, really nice. Same thing if water moves around a lot. If water's moving around a lot in the field, like it's shedding off of certain areas, it's collecting in certain areas, maybe you've got high groundwater tables, electrical conductivity is really nice, because between the water and the solutes dissolved in water, you're going to see that in the maps.
Cory Willness (09:22):
Now, if you get into land that's super flat, and when it rains, it just sits there in the whole field, it doesn't move, and it's all one texture, like it's all clay, well then that layer isn't that valuable anymore, right? Then you have to go look and to say, "Okay, well, let's see what my water layers look like. Let's see what maybe a soil color map looks like. Maybe it's more influenced by historical wind erosion and there's top soils blowing around." So every area that you move to is all maybe a little bit different in what layers provide the best ... the key layer that have higher weighting on the map. But in all cases, you can't build a map from a single layer. It's not like you can ever go out and say, "Oh yeah, this specifically," or, "An EC map, this is the map for this field. That's it, we're done." That never happens.
Morgan Seger (10:22):
Yeah, I can see that, because there are just so many different pieces to measure. The trick is bringing them all together and then making sure that they are appropriately reflecting the field, like you were saying.
Morgan Seger (10:33):
So when you get into different soil environments, like you were saying you might have some sandy loamy clay stuff all in the same field, when you are running to get your EC measurement, do you usually use the same depth? Or what's your protocol for evaluating the field that way?
Cory Willness (10:52):
Well, the sensors that we've been running are either two depth or four depth, typically up to a meter or say three feet in measurements. And it's very common, I would say a very high percentage of the time, that they're mirror images of each other as far as the depths go. Like the second depth or third depth or fourth depths, as you go down in the profile, the readings always tend to go up because you're getting into probably higher water content, higher solutes in the soil, probably a finer texture. That's pretty common out there.
Cory Willness (11:27):
So the actual depth part of it is not really that important. The more important part is that, when you're mapping, you're able to capture the spots, because it's irregular shaped, you know? So being a mapper often involves you're driving around, making sure you capture these little hills or areas where the stubble look different. And it's sort of like your job is to capture variability. That's the most important part.
Morgan Seger (11:54):
Yep, okay. So this makes sense. And I think maybe I just wasn't thinking through this right. So you're already collecting both, and then is there any human interaction? Say the maps do look a little bit different depending on the depth; is there a human interaction to determine which one you use? Is that kind of the boots on the ground part?
Cory Willness (12:14):
When we first started this process, I'd say it used to take us two or three hours to build maps. By the time you took all the information, there was no simple tools to do this. So a common person just trying to use existing software is going to find it's going to take you two or three hours to try and just get something that makes sense to you. So in our module we've now got that down to 20 minutes of human interaction. So we put together what are the most common things and ways that things would go together? Because we've ground-truthed millions of acres now over many years and all over North America, plus some other countries. So you just get to see how things are going to go together.
Cory Willness (12:59):
And then all the data always synchronizes back to an agronomist and ground-truthing is the next step. So they take in each of the layers. So it might be waterflows, a depressions layer, a topography layer, the groups of hills, mid-slopes, depressions, an EC layer. All these different layers. So you see the actual maps, plus then they get a set of SWAT Maps that are predicted to be the ones that will be the best for them. So it's their job then to drive around the field and stick the probe in the ground and learn this field and say, "Oh yeah, yeah, this is a little bit better, that's a little worse."
Cory Willness (13:35):
And then oftentimes there will be some fixes required, like a little bit of what we call them mods. Mods due to certain things that aren't picked up properly or need to be rezoned or something like that. So, if that answers your question. But yeah, there's always ground-truthing. We never actually would ever send a sample out and say, "Well, here's the map, go and sample it." No, you have to ground-truth the maps. And unfortunately, nothing is push button yet. Yeah.
Morgan Seger (14:04):
Got you. Nope, that makes sense. And I think that we see that with a lot of different things in ag, especially when you're covering, like you were saying, millions of acres.
Morgan Seger (14:13):
Once they have their map then, how are you seeing people use them? Is it mostly variable rate fertilizer or variable rate seed and fertilizer? Or what combination do you see most often?
Cory Willness (14:25):
Yeah, it would be both, so depending on where you work. So in some areas, applying a lot of products prior to seeding in the soil is not that common. Like lots of places that use planters and stuff, they seed place a little bit of stuff, or side bound it, and then most of the rest of the nutrients come in with different top dressing applications. They might use the maps for their initial stuff that's going into the soil, but in general, they're probably more dominated by using seed rate prescriptions and that type of thing. Whereas, like in Western Canada, for example, it's an area of the world that's a very short season, all the inputs basically go in before the crop is even germinated. It's either prior to seeding, or it's with the seeding operation. So all your fertilizer, all your seed rates and everything's going down in one massive cart. And so you're varying everything that way. So yeah, it kind of depends on the area, but it's pretty much dominated by fertilizer and seed rates. Yeah.
Morgan Seger (15:32):
Got you. And then what kind of lead time would a grower need if they were interested in trying this? Knowing that we're inching up on spring here, is this something they could still pull off for this year?
Cory Willness (15:43):
Oh yeah. Typically all our partners ... we do significant business in the spring. It's very common that people would try it in the spring, like do a couple fields, get them mapped up, so then, by the time you map it, you can get the maps back within an hour, you could ground-truth it, soil sample it, send your samples away and have prescriptions probably within four days, type of thing, in a best case scenario. So there's generally always time to try out some fields.
Cory Willness (16:14):
We've finished farms in the spring before. It's not uncommon to map tens of thousands of acres and do a whole farm that's 25,000 acres. That can all be done if it was all planned in advance. But yeah, there's plenty of time for people to try. And I would say it's the fall when people say, "Yeah, I liked how this worked and this makes sense to me." We'll finish it up in the fall. That's pretty common too. Yeah.
Morgan Seger (16:37):
So as we were kind of chatting before we started recording, you mentioned that there was both software and hardware options. Do you mind sharing what some of those hardware options are?
Cory Willness (16:49):
Sure, yeah. Well, one of the challenges with soil-based precision ag is just it doesn't scale very well. And that's why there's a tendency to use satellite imagery, because anyone can buy a computer and access imagery daily and all over the world for next to nothing. Whereas, when you want to get soil maps, at some point in time you've got to go out there in the field and then manually collect the data. So we built a product called a SWAT Box, and it's basically a Geonics CM-38, or a geo prospectors topsoil mapper sensor inside a box that collects the data remotely. So you can attach it to a piece of farming equipment, like you could attach it to an air cart or a planter or a seed drill, you can attach it to sprayer booms, and it has no operator interface. You just plug it into power and it'll automatically send all the measurements to our servers in real time. And then the staff can build your maps from there.
Cory Willness (17:54):
So yeah, that's the SWAT Box. That's sort of our major piece of hardware that we've built to help people scale this and get the job done without having to have a person with a vehicle driving around the field, capturing data.
Morgan Seger (18:09):
Yeah. And then they can multitask.
Cory Willness (18:12):
Yeah. Well, then it's cheaper.
Morgan Seger (18:15):
Yeah, for sure. And is this something that you mostly see through ag retailers? Or again, is this something that growers could access themselves?
Cory Willness (18:24):
Yeah. If someone's interested in being a partner, you can go onto our website and you can apply to be a partner. I think we've got 56 partners right now, mostly in North America, but there's some in other countries too. And we do have a few farmers that are partners, that they've got their own agronomists and they've got all the software and they're going to invest the technical time and everything to do all this themselves. But like I said, that's only probably about 5% of the people that would really chew off that big of a job. 95% of the rest is ag retailers and agronomy firms that are using our technology.
Morgan Seger (19:09):
Sure. And everyone places a different value on this, but is there a range of acreage that you would say would be worth it for a grower? We were talking about scalability; if they wanted to go out and do it themselves, do you recommend them being a certain size, or is it just the value they place on the data?
Cory Willness (19:29):
Well, growers doing it themselves, I guess the barrier to that is that they probably have their own agronomist, they have their own soil testing trucks, then they're writing all their own prescriptions. And so you have to have the skills and you have to have the people, and not many farms have that. So it's not like I'm trying to discourage farmers from becoming partners, but in general we encourage farmers to work with people that have dedicated their life to this. At my companies, with our teams, between our companies we've got 40 people. Well, this is all they do, right? So it makes it as easy and simple as possible for a farmer to go out and execute on these things. Like farmers are already wearing way too many hats; they're welders, they're marketers, they're combine experts, they're sprayers. How many things can they be experts at without going crazy? That's why they aren't accountants or lawyers. And I believe that precision ag in general is another one of those specialties where it's like, don't bother, just go hire experts that do this for a living and make life a little easier for yourself.
Morgan Seger (20:43):
Yeah. Well, and it comes back to the time thing. I know back when I was building prescription maps, the first couple took so long. But if you're looking at maps all the time, you really start to pick up trends and things that would signal to you that something is right or something is wrong. So it definitely speeds up the time and efficiency as well.
Morgan Seger (21:02):
So is there a place that people could go if they wanted to learn more about the work that you guys are doing?
Cory Willness (21:10):
Yeah, well our website of course has lots of information, swatmaps.com. We're a very active company on social media, so like on Twitter, our company website is @swatmaps. And from there you can see lots of our staff, and our partners around the world are also very active on Twitter. It's a great place that people are always sharing pictures and farmers are interacting and stuff. Those would probably be the two biggest ones, I would say.
Morgan Seger (21:39):
Okay, great. Well, I'll make sure that I link out to those in the show notes. Is there anything else about SWAT Maps that we didn't cover today that you want to make sure our listeners can hear?
Cory Willness (21:49):
Yeah. I think it's important that people recognize, like a lot of this precision ag space is things or processes aren't finished. I would say we're well down the path of doing things very well, but we're on the path of doing things well because we chose to have a technology solution that we provide; it's 50% technology, and as far as implementing on the ground, the other 50% is agronomy. And we've chose to say the person that's implementing it at the ground level is specialized in this. They're very good. I think people are going to find that a SWAT Maps partner is a very good person to do business with. I can think of in the United States, like all the partners that we have down there, in Illinois there's Bodie and in North Dakota there's Josh. And I could keep going through the list and these people are all like superstar agronomists in their regions.
Cory Willness (22:49):
So we have great technology, but paired with great agronomists and people that really care about doing the right thing. Then you've got something. Even though it's not a finished polished product where you can do data analytics and all the things everyone talks about, it's like, "Yeah, but we've got the main part right." And with good boots on the ground, you're in good hands, you're going to do well. This is going to work for you.
Morgan Seger (23:16):
Yeah, for sure. I love that. I mean, I think that there is a lot that we can't replace quite yet. And the technology's constantly changing, but having a solid agronomist on the ground on your fields is definitely a high-impact thing.
Morgan Seger (23:31):
So one question that I like to ask our guests is if there is one technology that you're most excited about? And it can be yours or not yours, no limits. Is there any one thing that you're most excited about?
Cory Willness (23:43):
Yeah. I think the technology that I'm most excited about now is most of the AI and machine learning stuff. So we've got what I would consider a small project on machine learning where we're just building a basic camera system that identifies crop and weed populations. It's nothing fancy, it's going to be really cheap and easy and inexpensive to operate. No longer does an agronomist have to walk out to a field or drive a quad and stand there and count plants. And you only get to count them in 15 spots; is it really representative of the whole field? And now we can do thousands and thousands of images and it's processed in real time. And so, I mean, that's just a little project we're working on, but around the world, there's a lot of fantastic projects going on like this that's just scaling things that we've done so slowly and inefficiently, and not only just making them faster and cheaper, but way better. The data that's coming out of these systems, it's phenomenal. So yeah, that's the technology I think I'm the most excited about.
Morgan Seger (24:51):
Yeah, I totally agree. And I can't help but think of the manpower that went into doing those tasks before. I just don't believe that that was their full potential. So it'll be interesting too to see, as they realign those resources that they had, instead of counting weeds or stand counts and things like that in the field, how they might be able to apply their skill set to something else. So I'm excited about the potential that that has, kind of on the flip side of the machine learning.
Morgan Seger (25:20):
I appreciate your time today and I'm so grateful you were able to hop on. And like I said, we'll link out to your contact information in our show notes if anyone wants to learn more.
Cory Willness (25:28):
Thanks, Morgan. Appreciate it. This has been fun. Thank you.
Morgan Seger (25:31):
Morgan Seger (25:32):
Thanks for tuning into another episode. If you want to learn more about Cory and his work, you can find them at swatmaps.com. And again, they're very active on Twitter. If you'd like to learn more about Precision Points and the work that we're doing at precisionagreviews.com, we encourage you to check out our website. From there, you can look at grower-sourced reviews for products and services that you may have used or maybe considering using. You can also check out our expert advice on our blog. We really appreciate you tuning into another episode and we hope you all have a safe and successful spring. Let's grow together.
Thanks for tuning into today's episode. To hear more podcasts like this, please rate, review and subscribe to Precision Points. Visit precisionagreviews.com for show notes from this episode and read expert advice on the blog. Share your experience with the precision ag products you use and check out our network of farmer reviews. Let's grow together.
Host: Morgan Seger
Morgan Seger grew up on a small farm in northwest Ohio before studying agriculture at The Ohio State University. She spent 10 years working with ag retail – specifically in ag tech – prior to hosting the Precision Points Podcast. She lives and farms in western Ohio with her husband Ben and their four children. Morgan has her own blog, Heart and Soil, where she talks about her experience farming, gardening, and raising her family.
Guest: Cory Willness
Cory is the cofounder of Croptimistic Technology, an ag tech company that provides software and hardware solutions to build SWAT Maps (the world’s favorite soil maps). Soil, water and topography (SWAT) is the foundation layer for soil sampling and soil-applied products and strategies. With an international staff and a growing network of partners in the USA, Canada, Australia and South Africa, the SWAT team is helping agronomists and farmers around the world maximize their product efficiencies and minimize environmental impacts. Millions of acres are mapped and retention rates are over 95% annually as partners and farmers see excellent value from SWAT-enabled variable-rate services.