Podcast Extra: Rooted in Organic Podcast - What AI in Ag Can Look Like with John Kempf

You're listening to Rooted in Organic podcast. I'm Mike Klein and along with Brian Wood. Today we have a returning guest. He's been on a couple times before. One of the early episodes, John Kemp, who is the founder of AEA Advancing Eco Agriculture and is very involved in the regenerative ag movement and he has some exciting new things we want to talk about today. So, John, thanks so much for joining us again.

Yeah, thanks Mike, and thanks, Brian, for having me back on. I've always enjoyed our conversations.

So, yeah, John, usually we start these with a good introduction and that usually takes us down wherever we wind up. But since you're a returning guest, I think folks should be aware of who you are and if not, I mean, you can certainly give a quick introduction, but I think it would be more most interesting to just talk about what, what it is that you're. Some of your latest projects you've been working on and how that's advancing agriculture, what your, what your goals are behind them.

Yeah, thanks, Brian. There's, you know, the world of agriculture, and particularly this world that's now being called regenerative agriculture or biological agriculture, whatever you want to call it, has changed so much in the 20 years since I've been involved. And even just in the last half a dozen years, there has been this upswelling of accessible information where now there are dozens of YouTube channels, there's all the information out there. And you know, This method of farming differently, of a different type of agronomy that seeks to enhance soil health and enhance ecosystem health generally tends to be much more, somewhat more management intensive sometimes and certainly more knowledge intensive, more information intensive. It's like there's a need to know more and to constantly be asking different questions and be asking better questions. And while it's true that we live in an information age, we have access to information in ways that we've never had before. It's also, it's sometimes hard to know where to find that information because the best information relative to this type of agriculture, a lot of it is being put on the Internet very fast, but the best of it is still held in books and held in people's experience. It's not on YouTube. There's a lot of great information on YouTube, but there's still a lot of information that is more difficult to access. And so one of the things that I've been thinking about for quite some time is in our work at Advancing eco agriculture, our, our job as a consultant is to have read the books, to have benefited from the experiences of hundreds of growers that we have worked with and to bring that to the farms that we are coaching and that we're working with and to, to help to give them access to that information. Well, the problem with that is that if you really want agriculture, regenerative agriculture, to scale rapidly, then there are limitations to how rapidly we can bring people on board, how rapidly we can train them, how rapidly they can get exposure to that type of experience. And there are limitations to how many people we can support, how many people we can help. And so for the last decade I've been thinking about how can we democratize knowledge, how can we digitize it, how can we make it readily accessible? And I first sat down and outlined a sketch, a scope of what at the time I called an AI agronomist. And well, AI today turns out to one. It's developing extremely rapidly, but it's developing very differently from what I had envisioned. And I don't particularly care for the term AI anymore to describe what it is that we're doing, but for the last several years our team at AEA has been working on, on digitizing all the knowledge that we have collected and making it more broadly readily accessible. So in just at the beginning of this week, well, actually end of last week, I guess, we launched our, that regenerative AI agronomist that I first envisioned and sketched 10 years ago, 11 years ago, and we're calling it Field Lark. So it is the features that are currently available, it is a chatbot that you can use and you can ask it any type of questions about. It's focused obviously on agronomic questions, but any type of soil and plant ecosystem management questions related to agronomy, related to what types of biologicals, what nutrition interactions, what causes different diseases and insects. You can ask it any of those types of questions and the answers that you get back are, I mean this, this is technology that is not dissimilar to ChatGPT or Gemini or Grok or these other AI tools that are out there. But this is, we are not, when you have, when you're using those general chatbots to scour the Internet and to come up with publicly available information and the results you get from those versus the results you get from Field ARC are, they're not even in the same league. We're not, we're not making slight incremental improvements of 10 or 20%. We are making step change improvements and get, getting much higher results back than we used to, than we're able to get with those general AIs. So that's pretty exciting. We're quite excited about that. And then there's some new product stuff that we've been working on the last year or two that we, that we're now just releasing that I'm also pretty excited about. So there's, there's, there's a lot of momentum, a lot of energy.

Yeah, I'd say so. And my first question on the AI would be or the chatbot or field lark. I think I'll just go with that for consistency sake. So, you know, when I think of what is out there in terms of, you know, human agronomists, so to speak, you know, their basis of their knowledge usually comes from varied sources, whether that.

Be.

You know, university educated or experiential, you know, their local experience in the area or I'm trying to think of the, the author that wrote some of the books and I'm drawing a blank right now that Dr. Paul always referenced. But either way, that, that's kind of where my question starts with AI is, you know, is there a basis, a base point that feeds into the intelligence that it has or where does that come from?

Yeah, it's a great question. So the base points are a number of written resources, materials and books out of my library and the types of books, I'm not sure who you're referring to in reference to Dr. Paul, but books by Acres USA and just generally the scientific reference books and the open published literature that are aligned with regenerative agriculture, that talk about the interactions of microbes and minerals and soil and organic matter and so forth. So it is published information. And then the other significant data source is all of the database that we have developed in our work at advancing eco agriculture, where we have this large database of soil analysis and SAP analysis data and the crop reports that came with those. What were the yields like? What was the performance like? What were disease and insect susceptibilities, what were, what's, what is the microbiome like? And what does the DNA profile of the soil and the plant look like? And you combine all that data together and the, this is one of the places where computers and machines in some ways have, they have complementary analytical abilities to humans because we have, we can only hold so many data points in our mind at once and compare so many variables at once and they can compare many more. So all of a sudden you start looking at this data and you realize, oh, wait a minute, for this particular crop, every time we have this disease show up, we have this particular nutritional profile show up, or we have this particular microbiome profile show up. And so there is an abundance of very rich, high quality data there that feeds into the background of this model that is. And that is information that's been collected over the last 20 years. But I'm not aware of any other similarly integrated database that exists.

So it works similarly to any other chatbot, but you would have given it a different baseline to begin with.

It's got a completely different training set. It's not trained on the information that is available on the Internet. It's trained on the information that we have vetted and proofed and said, all right, this information is worthy of including. So you know the old, you know the old saying of garbage in, garbage out. And the only way, the only way you get something better than garbage out is to not put garbage in.

So it does not scour the Internet like the others for the information or does it using certain parameters that you would have put in?

It uses the parameters that we have put in.

Yeah, I'm going to definitely play with it.

What have you observed so far, Mike? What have you. What have you. Have you played with it at all?

Yeah, I did. When you had sent me that link a couple of weeks ago, I started, like, playing around with it, and my wife actually played around with it more, but it was. We have some brown spots on strawberry leaves. So she went into that and came up with a whole plethora of problems or, you know, things we should be doing. And then also asked on, like your strawberries runnering, like, do you let them run or do you not let them run or. And it was. The response was quite different than a Google search.

Yeah, yeah, that's. That's exactly what we're looking for, Mike, is to be able to. You know, one of the things that has always intrigued me, and you know this from experience, if you, if you think about the knowledge base of a dairyman of who is very good with cows and who has 20 years of experience treating cow health organically, their knowledge of the physiology of the cow and the husbandry and what is required to care for cows versus a homesteader that has had one cow for five years, the delta. The gap in those two knowledge sets is vast, and yet it would be very valuable and very beneficial for the homesteader to have access to that knowledge of the guy who has 20 years of experience. But how do you do that and how do you accomplish that? And I think we now have a pathway where it is easier to get access to that type of Knowledge, that type of information. But we also, we have to be very clear. I've always been very clear in thinking about, all right, how do we develop technology that is ethical and that is appropriate and that is designed not just for regenerative agriculture, but that inherent in the way that it is designed, it inherently regenerates relationships. And, you know, that's an interesting conundrum, an interesting challenge. Challenge to think about. And I think at its very foundation, we have to think about the highest and best purpose and design of technology, in my opinion, is not to replace people, but to make people better, to get people to ask better questions, to be more observant. That's the way that we should be developing this type of technology. And anytime that we try to use technology, when we look at our history in agriculture, all the way back to Wendell Berry and Wes Jackson and your father Mike to David, one of the challenges with the way technology has evolved in agriculture over the last 70 or 80 years is that generally technology has meant the displacement of people. We've designed it to replace people, generally speaking, rather than to make people better. And so that's one of the things that I hope to come about with this technological innovation, is that we can help people become more informed and ask better questions rather than trying to replace them, because there is no replacement. There is no replacement for human hearts and hands and minds in the field. Anytime we remove people from landscapes, the landscape and the ecosystem suffers.

I think Wendell Berry, when we had him on, mentioned the eyes to acres ratio is out of whack and how detrimental that is. But boy, that is a, that is a fine line to walk because technology has really been used, like you said, to replace people and to kind of be able to get big and you know, regular agriculture has grown in size per operation significantly because of advances in technology. So where do you draw that line? How do you build something that doesn't create that replacement for humans?

Well, I think you can only go so far in the design of the tool or design of the technology. You can design it to, to a degree, deliberately to try to make people better and to not replace them. But there is also a component of the user of the tool has to use it appropriately as well. So there's only so far you can go from the perspective of creating the tool. But. And then the other contributing piece is who is using it and how are they using it. So, yeah, it's, it's an interesting place to be. But as an example of the ethos that we're trying to capture behind Us AI agents today are many users. And the language around them is that they're referring to them as if though they were a person. And we're very deliberate in the way that we've developed Field Arc to, to use language so that it makes it clear that this is not a person, that this is not a soul or a mind behind this, if you will. This is just a technology and just a tool. And even just in the way that we thought about the name I, we, we had. It's getting to be quite difficult to come up with original names in the world today because there are so many names have been captured already from a trademarking perspective. Well, we came up with the name Field Lark because of job 12. 7. Job is one of my favorite books of the Bible. I'm constantly thinking about the lessons about agricultural ecosystems that he teaches that are communicated there. And there's some fascinating things there, reference to the minerals of the soil and so forth. But you know, in. Particularly in the Anabaptist circles, there is this constant warning or aversion to worshiping the Creator, this constant warning to not give the creation precedence over the Creator. And I understand that, but I think something is lost commonly in the way that that is communicated. You know, in job 12, I think somewhere in there it's the beginning of the chapter. It's like verse six or seven or somewhere in there he says something to the effect, but ask the animals and they will teach you. Ask the birds of the air and they will tell you. Speak to the earth and it will teach you. I think this is one of the foundational aspects that we get to when we think about good stewardship. And what Wendell Berry was referring to is the eyes to acres ratio, or what I've been calling the heart to acres ratio, is you have to observe, you have to pay attention, you have to ask better questions. And when you ask questions of the birds and the earth and the animals and ask them to show you what is happening and going on, they can tell you.

I like that. I always refer to it as farming in harmony with nature, not fighting it. Because this goes back to one of our early conversations we had know farming with the warfare paradigm, where you find the problem and you kill it. Instead of like, why do you have that problem? It's, it's kind of like our society wants to fix the problem or make the problem go away without fixing anything, just treating the symptom. So I really like that that you're trying to work, like, look at your fields and spend time and see what the problem is, I think this is really interesting.

Well, I've adapted Wendell Berry's phrase where he talks about the eyes to acres ratio. And for the last dozen years or more I've been speaking publicly about the heart to acres ratio because it's the eyes alone are not what makes the difference. It's the eyes of a caring observer, someone who really cares and who really engages with the things that they observe and implements from that. So there is, there is that, there is that observational component which is fundamental. That's the first step. But then the next step past that is intentional and deliberate stewardship. And that has to come from a place of care.

Yeah, and that's the, that's the conundrum I'm, I'm trying to balance in my head right now. And I think, you know, I'm trying to put myself in the, in the shoes of a, of a day to day farmer that is, you know, when we're talking about this, technology advancements, I certainly see the opportunities and I see a level of resistance to that because of that, of what you're talking about around farming being. When you're farming with that heart to acres ratio in mind. And with that approach, it turns into more of an art than a science, which, or is viewed by the farmer in a lot of cases as being an art more than a science, I should say. And that feels to me like an antithesis to technology because you need to live it to do it, whereas technology obviously doesn't live it. So that's where I guess that's a long rambling of what's going through my head. And my question in there is, you mentioned the AEA database is part of the input to this tool and using the observations that have happened through, you know, your team of folks, how does that even get documented and included? Like, what is the process behind that? Because a lot of times, you know, like Mike or I, when we're on a farm, we're just kind of verbally communicating what somebody else has done and it's more of a, like a, like a story that's being passed from one to the next and it doesn't always get written down, which I recognize is a, is a loss or could be a potential loss. But I'm wondering how you got into that practice and how that happens.

Yeah, Brian, that's, that's a great question. And I'm not, I'm not dodging or sidestepping the question, but you said something important that I want to get back to and not miss Just in the way that you, the way that you framed it. And thinking about the stewardship perspective and the art versus science perspective, you know, when we think about this from the dimension of not replacing humans, there are what humans bring. And the art versus science component, you could almost think of it as heart versus head. It is, it's the, the art versus the science is the intuitive knowing versus the intellectual knowledge. And perhaps it's not appropriate to include the framing of one versus the other because they perform the best when they are used in parallel in conjunction with each other. And so the way that I think about this, rightly or wrongly, the way that I'm thinking about this is that this is a tool that when used. Well, the purpose of this tool is to sharpen the mind, to sharpen the intellect, to ask better questions and to ask the question why? Why did this scenario emerge in that, in this setting, but not in that setting? What are the foundational factors? Could it have been one of these five things? So it is a tool to sharpen the mind, but it should no way. It cannot be a replacement for intuitive heart connection piece. But it can help sometimes those questions that we can think about those questions intellectually, but the answers have to come from the intuitive process because the answers aren't there intellectually. There's not research. Well, there might be research that we can go do, but also for our specific context, we have to go out, we have to be present. We have to ask the animals and the earth and the birds to really understand what is happening and what's going on. I'd love to hear your thoughts and your response to that. What do you think about the way that I'm framing that?

I like that. That again, I'm. When I'm hearing you talk and hearing the concept here, I'm. I'm attempting to put myself in the shoes of. I mean, the group that I would be most familiar with would be a dairy farmer. You know, of going in and, and using this tool and there is a level of resistance and sometimes I think it comes. And some people may be offended if they hear this, but I think I'm now applying my own experience to trying to put myself in somebody else's shoes here. When I say that sometimes what stops people from going to seek further information and improving themselves is not even knowing where to start.

Exactly.

Conversation.

Exactly.

That's where I see this technological tool being a big benefit of. You can have that conversation. You can, you can fine tune what it is that you need to know or even identify what Questions or what is the, what is the piece that I'm missing or not understanding here? And then you go seek human interaction.

Yeah, Brian, you're exactly, you're so spot on. You know, I just, I had this conversation with an early user of this tool just before we started having our discussion. He was an apple grower in Washington and he sent me a message and he said, you know, I just had this incredible conversation with one of the agronomists that I work with. And the, the way it happened is I'm dealing with this certain problem that I haven't been able to figure out how to resolve. And I was having a conversation or having an exchange with with Field Lark and it helped me. Once I started getting into the conversation with my agronomist, I was able to ask so much better questions. And he had, my agronomist had the benefit of experience and context specific knowledge. I was able to get much better answers from him that I would never have been able to get from the chatbot alone. But I wouldn't have been able to ask as good questions if I didn't have the previous exchange with the chatbot. So that's, that's, you're exactly spot on. That's, that just happened.

Yeah, that's, that's amazing because, you know, I, you hit the nail on the head earlier on and I would apply this, whether it be regenerative or organic or any of those types of systems that the, the landscape of available resources for any of these or any of the generalized topics, whether that be agronomy, nutrition, you know, animal health, animal welfare, whatever it may be, it's expanded significantly over the last 10 to 15 years. There's been, you know, research by universities, by organizations, there's been books written so on and so forth. And there's people, professionals working day to day, which is fantastic. That's great. It's what we've wanted to see. It's what the early adopters, you know, didn't have available to them. But you know, like you, like you said at the beginning there, that then along with that abundance of, of people in the space comes an abundance of information that is impossible for everybody to, to keep track of.

Yeah. Some of which, or perhaps even a large part of which is contextual and context dependent. Grazing practices that work well in more dryland environments are perhaps not ideal for wetter environments. And so it is, it's the process of parsing out that information, figuring out how to adapt it and how to make it context appropriate that is again the Piece where I have a difficult time envisioning technology being able to accomplish what it is necessary for humans to do as caring stewards of the landscape.

Yeah, absolutely. And you know, just one other tail off or tail question here that may lead us down a whole nother rabbit hole. The, the information that the chatbot or field arc receives, do you have access to the questions that are being asked? And the only reason I'm asking that is, you know, I'm seeing that as a potential opportunity to get better understanding for what type of research or where are there gaps in the knowledge that it has. And so how do you continue to fine tune the tool?

Yeah, that's a good question. So there are multiple, several thoughts that come to mind. One is the users of the tool are able to provide feedback to every response. Every response can get a thumbs up or a thumbs down or even a written response that says, you really missed the boat on this one. So there is that very direct feedback component that gets, that then gets reviewed and incorporated. And also, yes, we do have, we have to have careful conversations about invasions of privacy and so forth, but we do have access to, at a high level being able to see the types of questions that people are asking. And yeah, there is, there's a lot of opportunity here for, for further development, further information development and further fine tuning.

So, and that is basically focused on the agronomy side. As of right now.

I would say agricultural ecosystems more broadly. So there is the training set. The training data set is constantly being refined and added to, but I envision it including very thorough information on water cycles, for example, the small and large water cycle on, and all the factors that contribute to agro ecosystems more broadly. So sometimes some people interpret agronomy as being very narrowly focused on just nutrition management, where this is going to be much more, much broader than that. We're also including very in depth information on or a lot of resources on grazing management, for example. You know, one of the things that I find so intriguing is some of the resources that I have access to that we have access to are books and resources that, that aren't readily publicly available anymore. Edward Faulkner when he wrote Plowman's Folly. Andre Voison. Andre Voison wrote half a dozen books that are absolutely incredible resources for grazing management. A lot of information and knowledge that we are now rediscovering bits and pieces of part by part. But his book on grass productivity and soil, Grass and cancer, they're not in print. You can't readily buy them anymore. But this is a pathway that we can incorporate that information. And when, you know, there's this old phrase that when the student is ready, the teacher will appear. Now, when you, if you could, if you figure out the right question, if you're asking the right questions, that information all of a sudden pops up for you and you can have access to that information that they've been working on 60, 80, 150 years ago. In some ways it's almost. I'm not going to say it's better than a library because I do believe again in the concept of. I think it's very important from my perspective, it's very important for people to still read books and to still learn the knowledge for themselves. You know what happened when we had Google Maps on every iPhone? People lost their ability to navigate.

I got lost.

They got lost. They lost their ability to navigate based on a map or just their sense of direction. And that can so easily happen with this technology. And it can happen regardless of whether we develop field ARC or not. As chat, GPT and these other AI tools become mainstream and commonly used, people are the potential, the very real potential exists for people to lose their ability to think critically, even more than they already have. And so this is not a, this is not a replacement for library, and it's not a replacement for learning and reading on your own, but it can point you in the right directions very rapidly.

I think what you're getting at is, you know, it's a fine line between reliant on technology versus an indicator from technology. I guess, for lack of. I'm not thinking of the right term to use for that second one, but that reliance that hit me when you just use that Google Maps. I know I've attempted to not use it because when I don't have it, I can't find my way around my own neighborhood.

Yeah.

Which, which is sad, you know, and that, that is a good analogy to use because if the farmer, you know, doesn't have a map for, for lack of a. Better to continue to the, the analogy here. A map of their farm and know what direction they're going to go in, they just become reliant on being told.

Exactly. And you know, what's, what's worse than being told bad information by a human being Told bad information by an AI.

Yeah. Because it's like a Google search. If you want, if you can find the answer that you want, you know, may not be the right answer, but you can find an answer that you want to hear, which is not necessarily a good thing. Would you, would you see this expanding into the animal agriculture world as well? Or is that already a piece of.

Is not a piece of it in the sense of animal nutrition and animal husbandry. But it's very easy for us to add that piece. I mean it's, it's very easy for us to add that piece. And you should. And so yeah, it's, it's, it will not be difficult to add that at all. But you know, my goodness, Mike, something you said triggered an idea for me, something I wanted to respond to and now it's gotten wings and flown away. I forget. Oh, non linearity. So we know, good stewards, good farmers know that in many cases the solution if something is not working well, the solution is not just to add more of something. You know, for a long time agronomy has been based on this relatively simplistic perspective that you collect lab analysis and you add more of whatever is missing. If you're low on calcium, you add more calcium. If you're low on zinc, you add more zinc. And this chatbot is deliberately trained, Fieldlark is deliberately trained to describe all of the nonlinear interactions. And this is the one. This has been happening even better than I ever imagined. And things are coming forward that I did not know and that it's. Once you start digging, you find the information is there, but it will say things like, I love asking it questions about diseases, how can I manage rhizoctonia on potatoes, or something like that. Because what you get back is you get this very thorough and detailed answer back that says, well, Rhizoctonia thrives in this type of soil environment and it has these characteristics. And because of this, here are the mineral nutrients that you should add. And by the way, in terms of nitrogen fertilizer, you should only use ammonium based fertilizers because ammonium based fertilizers has this effect on these nutrient availabilities and has this effect on rhizoctonia indirect. And it gives you kind of this domino chain of two or three or four interactions where the type of nitrogen fertilizer that you use is very non obvious that it would have an impact on rhizoctonia to most people. But in our work at aea, we already know that very well, we've documented it very well that the type of nitrogen that you apply will have a disease suppressive or disease enhancing effect on different diseases. This fieldwork can describe the sequence to say A leads to B, which leads to X which equals J. And you have this domino effect where something that you did over Here ends up having an effect on causing a problem. And you might never have connected those two things in your mind.

That's what I find so interesting is observation will show you so much. You know, farmers look at cow manure to see if they have enough protein or not enough, you know, too much protein, not enough protein. The observation is it's a domino effect and there's. But to be able to find all these little things, you're trying to fix a problem. Well, you don't realize that there's a whole chain of events that are going to happen. And if this, if field lark can explain that, that is huge. I know, I know I'm going to use it.

It's still early days, but it is explaining many of those nonlinear action, linear nonlinear interactions in ways that I'm very, very pleased by.

So you said you, you got this concept or this idea 11 years ago. Have you been working on it 11 years? Or was it just an idea until recently?

So we started using plant SAP analysis in 2011, and we tested it extensively in 2012 and 13 and 14. And by the end of the 2013 growing season, I realized the implications of what this technology could offer because it was so sensitive and so accurate that we could, we. I was successful at predicting disease and insect susceptibility of a given crop into the future by about three or four weeks. I could say, well, this crop is fine right now, but if this trend keeps going, we're going to have powdery mildew in four weeks from now and be quite accurate doing that. And so that became. I realized the implications of that, that this is a very powerful tool and stoic. I sat down and sketched an outline of. All right, if we turned this knowledge base and with this tool of SAP analysis, combine it with soil analysis data, we turn this into an algorithmic engine, we would be able to make all kinds of recommendations about nutrition management and disease management and pest management into the future that we're not able to make right now. And so I sketched out all of these, all, all these various factors and nutrient interactions and relationships and ways that we should think about this. And then I put the idea on the shelf because I believed that at the time, with the evangelistic work that Dan Kittredge was doing with the Bionutrient Food Association, I believe that we were just a couple years away from having in field handheld sensors that could do something similar to SAP analysis. And my perspective was that let's wait until we have the sensors and then develop the software to be able to make those predictions to match with those sensors. Well, here we are a decade later, decade plus. So this all happened. This would have been December, December of 20th, November, December 2013, when I first sketched all of this out. And a decade has gone by. We now appear well, we've made significant investments on our own in a company called Croptix. And we have successful sensors in the field right now that are measuring mineral profiles and phytonutrient profiles on a couple of crops. We expect to release them in spring of 26 for a handful of crops and then to expand beyond that in 27. So that is now that technology is almost ready to go out into the field. But in the meantime, the digital technology has continued to develop and we've continued to develop a very large data set of SAP analysis data. So all of a sudden I realized 18 months ago, 12 months ago, you know, we have such incredible, incredible depth of data, quality of data at this point, that it is worthwhile for us to create this tool even before the sensors are fully deployed and fully out there. We can build it now based on the SAP analysis database that we have. And so I should communicate a couple of things clearly and just create some clarification here. So what we are building with fieldlark has two separate components. One of them has already been released this last week and one of them is still in development. So the piece that has already been released is what is commonly referred to as a chatbot. It's chat GPT or grok that has been fine tuned and focused on agriculture and agronomy with a regenerative focus. That's it. That piece is public, launched, live, ready to go. There is another piece that is quite different and that is an algorithmic engine. So the challenge with the chatbots, these large language models, is that they are inherently variable by design. This is not a flaw, this is a feature by design. If you ask the exact same question six different times, you will get six different answers. Now, they might overlap by 80% or 60%, but there'll be some degree of variation. And by the way, if you want to dig deeply into a topic, that can be a way of finding out all the edge cases, on finding out all the edge information, but that's a feature, not a flaw. When the reality is when you are making agronomic recommendations and you're making those recommendations based on data, you need precision, not variability. So the second component of fieldlark is an algorithmic engine where you then can put in, you can upload all of your Data you can upload soil analysis data, forage analysis, SAP analysis, microbiome analysis. There's we are building it and preparing it to be able to absorb and to utilize almost any type of data, laboratory data and harvest yield data, et cetera. And then from that data you put in now instead of, you have this algorithmic engine that can make very precise and consistent and repeatable recommendations to say, all right, you have these nutritional imbalances. Here are the possible solutions. Here are the COVID crops that you can grow that are known to address these nutritional imbalances. Here are the microbial inoculants that change these nutritional imbalances and make a very thorough set of recommendations that is based on the knowledge that is in the algorithmic engine and the knowledge that is in the chatbot. These, these two systems talk to each other, but one is inherently variable and one is inherently precise and consistent. Does that make sense? It does.

Because I don't want to put in my, my soil sample and get an analysis and then ask it six times and get six different recommendations on what, what's going to go on my land. But then would it even go. Because I know lots of times when you get a soil sample and you look at what it's all going to take, especially if it's a rundown field, you know, has no inputs for 30 years, would could you then also put your budget in and see like if I've got X to spend per acre, it would give me the recommendation based on that spend?

Well, in order to do that it would need to know what things cost. And so you would need to input that information as well. But you could ask it the question such as I have very limited economics, very limited economic resources, very limited budget, what is the most cost effective thing for me to do? And the key here, and this is another aspect of nonlinear agronomy that I'm really excited about, is what if the recommendation to solve low levels of calcium in your soil is not a recommendation to add more lime but a recommendation to grow a specific cover crop? Or what if the recommendation to solve a particular disease problem is not to spray a fungicide or even an organic fungicide, but to add a specific inoculant or again to grow a specific type of COVID crop. So it's that non linearity component. I'm of the persuasion that, well, there's this whole conversation about being able to regenerate soils with particular plant species or combinations of plant species versus mineral amendments and so forth. And the reality is we should thinking I'm now, drifting from the topic and focusing on agronomy here a little bit, but we should be thinking about the economic costs of soil mineral amendments as buying time. That's really what we're doing. We are buying time in many cases. So if we have soils that have large phosphorus reserves and we want to release those phosphorus reserves, we have two approaches. One of them is that we can grow plants that are good at solubilizing phosphorus, and we can graze livestock and we can add mycorrhizal fungi and we can increase the phosphorus availability in that soil. But that process might take a year or two or three or more. Or you can buy soluble phosphorus fertilizer like map, or you can add chicken manure and get a flashbang response on phosphorus. When you pay money for those types of soil inputs, you are buying time. You're getting results immediately in the first year instead of needing to wait five years. And so I think this has been a recent realization of mine is that this whole the ideology of minimizing inputs or using no inputs can be successful in some soils and in some settings, but it takes more time. So are you willing to have that delayed response and to take that time, or do you need a faster response from an economic perspective?

So I'm just thinking about, I'm looking out my office window and at my pasture fields. So if you have pasture fields, you know, cover cropping to fix something isn't necessarily really in the cards. It can be. But if you have a complete no till operation where everything is just grazed, what are the tools that can be used to unlock some of these nutrients?

It can still be specific plants and specific plant species that perhaps you just overseed or intercede into your existing pasture. So what if you overseeded fenugreek, for example, and got a much better, got a significant milk production response. So there's a significant economic response in the short term. But you also increased phosphorus solubilization. There are, there are tools for that setting as well.

I had a conversation with you, I think it was last fall maybe about native prairies and how there were so many Forbes in there. And, you know, if you graze really short, you'll get the Forbes established again. I'm doing a test run on that.

We'll.

We'll see how it goes. Heifers are looking great.

Yeah, I'd love to chat with you more about that. I've continued to learn more since we spoke and I actually on my podcast recently, I don't know, it's now half a dozen or More episodes ago, I interviewed Jim Elizondo from Florida, who, you may know if you don't.

He.

Let's just say that Jim and I clicked deeply. For the first time I got to have a conversation with someone who has in depth grazing experience, but also understands plant physiology in ways similar to the way that I understand it. So he has thought deeply about sugar flows within the plant and about producing high fat content plants and how to facilitate all that with grazing management. And as a result, his grazing, you know who's, you know what his grazing system resembles? It resembles Andre Voisan from friends from 80 years ago. He's talking about producing plants. I, he introduced as a concept to me that was. Made me scratch my head and never heard of it before. But we talk about happy lines on cattle. He talked about having happy plants. And he said, when you have a perennial plant, a perennial grass that is truly happy, it doesn't go to seed, it doesn't produce seed heads, it remains in a vegetative state. I'm like, say what? I never heard of such a thing. But he's doing it, has done it on multiple operations all the way from 8 inches of rainfall in the Chihuahua desert of Mexico to 60 inches of rainfall in Florida and everything in between all around the world. It's like, wow. So there are. And I'm quite excited by this because I've seen this in fruit and vegetable production systems where we start managing differently and plant expression changes significantly. And he's the one person who I've had a conversation with who has observed that same pattern on perennial forages and grasslands.

Boy, would that be a game changer when the weather starts warming up and you're trying to get through before all the seed heads shoot out and your quality goes downhill.

Wow.

If he can do that with Kentucky 31, he's a genius. But boy. Sorry. Go ahead, John.

Maybe we can use greasy management so that Kentucky 31 is no longer the dominant species.

Well, I'm, I'm trying actually to have this my second year. Grazing it every 15 days.

Yep.

And that is working. It's still palatable at 15 days. At 20, they don't want to eat it in the heat. But at 15 days it works great. And what's happening is I'm getting a lot more white clover back. Yeah. Simply by using the concept of try to kill it with grazing and you can't. But it's been interesting to see. I'm, I'm actually really impressed with the feed and how the animals look on it by going every every 15 days.

Well, you know the definition of trying to do the same thing you've always done and getting different, expecting to get different results. So kudos to you for trying something different.

I do sometimes do the same thing over and over and expect different results though too.

John. I want to. I haven't listened to that episode yet. How do you get happy plants? Is it the same concept as cows, nutritionally or.

Well, it is. The concept, in essence, is to time grazing with the. The moment and the window in which your desired forage species have the optimal ratio of sugar content in the forages versus sugar content in the roots. And sadly, there isn't a kind of simple, abbreviated explanation that I can give other than what he's pointed out. He said that Andre Voison, 80 years ago spoke about the exact same concept where he spoke about timing grazing management to the point where the lowest, oldest leaves on the grass were just starting to. They're starting to lose chlorophyll and the plant is starting to sacrifice them. And, you know, there is this perception that. That is associated with a certain stage of maturity. Let's take ryegrass, for example. Rye grass is supposed to have four leaves and then it starts forming a seed head. Except I was on a ranch just a couple of months ago where we had rye grass that had six leaves that were all still green, and the oldest leaf was just beginning to lose chlorophyll, just beginning to senesce, and had no seed head.

John, you were talking about the. The ranch you were on that had rye grass at six leaves and no seed heads.

So there was this observation that we can have. There isn't necessarily just an association between leaf count and reproduction that we can have these plants that remain reproductive even when a normal common knowledge would be, excuse me, I misspoke. Plants that can remain vegetative even when common knowledge would indicate that they need to go reproductive based on the stage of maturity they're at. So what. What Jim described for me, that really caught my attention. He said that in. In the. In a very harsh environment in the Chihuahua Desert, with less than 10 inches of rainfall per year, they didn't get to zero, but they got seed heads down to about 10% of what they were historically. They got about a 90% reduction in seed heads and a reproduction. And other than that, the plants remain vegetative. And that was just a reflection of grazing management and timing the grazing to come in at the optimal time when the sugars were balanced between the vegetative part of the plant and the root system. Because when you do that, when you. So now you remove the vegetation. And by the way, this, this also completely makes the observ. The, the conversation about take half, leave half obsolete because he's taking 100% or next 100%, let's call it 90% of what's above the soil because he has these large sugar reserves in the root system. And so when you have these large sugar reserves in the root system, the fact that you've removed 100% of the forage is not problematic because it just comes back with a roar because it has all of this energy reserve. So it's essentially figuring out and understanding once you have sugar reserves in the top and you have sugar reserves in the root system, but you haven't built up so much sugar reserves that you're ready to go reproductive. So I have a feeling that when Jim Elizondo and Andre Boisean's work becomes more widely recognized and appreciated, that we are going to be talking more about grazing management in terms of, of carbohydrate flow within the plant than we have been up to this point. We've not really been talking about plant physiology very much in the grazing conversations that I've participated in.

So would, would part of that be possibly because the rule of thumb is, you know, a 30 day ish rotation in your grazing and you know, grazing tall, you know, you said take half, leave half. After my conversation with you last year, I'm taking 90% right now to try to get clovers and stuff established and just trying different stuff. Would you say that, would his rotation be based on days or. No, based on the actual testing the plants?

Well, observation more than testing. But this is again coming back to the conversation we were having earlier. Understanding and knowing what to look for, having keener observation. And so his determination of when the optimal stage for grazing is, is based on what is happening with the lowest, oldest leaves of the plant. The moment they start losing chlorophyll and they begin senescing, it's time to graze. Sometimes Perhaps on Kentucky 31 fescue, that might be in 15 days. Sometimes it's 75 days. So all of a sudden this, this conversation about some arbitrary days to grazing number. We all know that in different environments, in different ecosystems, the, the amount of days it takes plants to reach a certain physiological stage is going to be very different. It's going to be dependent on rainfall and temperature and everything else. And so this is a scenario where we need to start asking better questions and being more observant. It's not about days. It's about or even growing degree days. It's a reflection of the stage of physiological development.

And how would you do that in a diverse pasture with 15 different species growing in there? Would there be. Would you look at the more dominant species that are in there to see, or will they all be at the same stage at the same time?

No, they're certainly going to be at different stages at the same time. And here's where I would love to have Jim or others with more experience in the conversation. But I would suggest there's two ways of approaching it. One is what's the optimal stage for your primary forage species, if that's rye grass or Kentucky 31 or whatever the case might be. And the other approach, if you have a very diverse mixture, is what's the optimal stage for the forage species that you want to be dominant? Because if you, if you want to, if you graze for the optimal health and the optimal performance of that of a particular species, then that particular species will thrive and others will be suppressed.

So I'm going to wrap this back around to what we have spent a lot of our time talking about on the field lark. You know, this is a perfect example of, you know, grazing just in general. There'd be a lot of information that would align with some of the more, I guess I'll call them conventional methods of, you know, amount of time of rotation and take half, leave half. There's been a lot of materials written about that and spoken about and this different approach. And how, how does. When I'm again putting myself in the shoes of a farmer, trying to get information on, on grazing and how to best manage their plants. How do you balance out what kind of, you know, there's two varied responses there, and maybe that's where then you get those responses. And then you talk to a human about what actually needs to happen. So that's a perfect example of kind of. That encapsulates this. There's a lot of different answers out there, but it's a matter of what do I focus on and then how to actually implement that information.

Well, I have, And I'm speaking from a voice of a certain element of naivete here because I am not a grazer. I've certainly consulted on a lot of grazing operations, but, you know, there's no replacement for experience and for having done it yourself. To some degree, you can learn from the experiences of others. But in our work in fruit and vegetable production and in crop production generally, I think one of the foundational things that has made us so successful in our work at AEA is deeply understanding foundational principles, kind of first principles thinking of plant physiology. So an example of what I'm talking about is we've been able to have to produce some significant yield responses on a broad range of different crops. And our approach is to understand exactly at which stage of plant development and what the process is for determining a reproductive bud. So if you take, if you take an apple tree, for example, the reproductive buds for next year's crop are set somewhere in the neighborhood of about 30 to 45 days after this year's bloom pollinates. So at the moment you have dime sized to quarter size small apple fruits on a tree. That tree is setting the buds for next year's crop. So if you want to influence the number of buds you have next year, if you miss that window, if you go and put on an application or nutrient application in August or September, you've missed your window of opportunity. If you take blueberries, for example, blueberries will transition buds later in the fall as they get close to dormancy on their new shoot growth. They will transition buds from vegetative buds to reproductive buds. So if you want to enhance reproduction, you have to time your applications then. So it's, it's this. So then the, the foundational principle, the first principle becomes you have to optimize nutrition at every stage that is associated with reproduction. So you have to optimize nutrition at bud initiation, at bud development, at blossoming and pollination, at the cell division stage. There are these four very distinct stages. And once you understand those stages, then you can go do the research like the research exists. We know that every reproductive crop will have these stages. They just occur at different times. Blueberries versus apples are at very different times of the season. But the reality is, from us, for us as agronomists, we look at that and say, oh, we manage blueberries and apples exactly the same way. And somebody who doesn't have that context says, what are you talking about, you idiot? You're putting on this foil application to enhance reproductive and apples, you're putting it on in June and the blueberries you're putting on in September, you're not managing them the same way. But in fact, from a first principles perspective, you are that type of understanding this first principles thinking of how to manage forages, at least in the conversations that I've participated in. And the information that I've been exposed to has been largely absent from grazing management. The whole concept of take half, leave half or taking the majority, or grazing on a certain 30 day cycle. Is there, There is no, there is no. There are the foundational principles that are behind that, when that was first communicated, first discovered, are not generally communicated. And the problem with that is that it makes difficult, makes it difficult to adapt it to different contexts. So if I describe why we are, describe why we are putting on a foliar application of nutrients on apple trees at a bud initiation stage, and I describe that it's because of bud initiation, a blueberry grower can take that information and figure out that he needs to do the same thing at a very different time. And if we had a conversation about forage management from the same first principles perspective, and we said, all right, it's this particular stage of plant development that expresses itself in this way. This is when the optimal time is for grazing, for optimal regrowth and optimal root system and plant health and stand longevity. And this is also the optimal time for livestock performance. Now all of a sudden you can take that foundational knowledge and you can apply it in the desert or in 60 inches of rainfall, you can apply it further south, you can apply it further north. And what you'll discover is that in, in one soil type, at a certain latitude, that might mean grazing every 25 days and somewhere else, only 100 miles away, it might mean grazing every 45 days. But now you can take that first principles knowledge and you can adapt it to the local setting, you can adapt it to the local context. And from my perspective, and again, I'm offering a lot of the caveats here, that this is not my core area, but from my perspective, we've been having a rules based conversation around forage management of follow these rules rather than follow these foundational principles. And those rules have gotten, they've created a lot of challenges. Because the reason I'm interested and the reason I'm having more of these conversations the last couple of years is because I've observed entirely too many farms who did exactly what they were prescribed to do. They followed, they listened to multiple teachers, the best people in the space, they followed all the recommendations, they did everything to their optimal, the best that they Knew. And after 20 years of, of intensively managed rotational grazing, in some cases with dairy and in some cases with beef, they hit a plateau initially, there is some initial improvement in soil organic matter gains and productivity gains, and then they hit a plateau. And I've seen too many farms who have been grazing for 15, 20 years, in some cases even as much as 30 years, and their organic matter Levels are not significantly better than they were 20 years ago. And that's a problem of not understanding the first foundational principles. Anyway, I feel like I'm on a little bit of my soapbox here. I need to get off. But that's, that's my perspective on what I've observed. I think if we really want to get off this plateau and have much better performance from a grazing system, soil ecosystem, soil health and livestock performance perspective, we have to really understand the foundational principles of plant physiology of our, of our target forage species and manage for optimal performance there combined with optimal livestock performance. You do those two things and you should not have stagnation of reaching a plateau and having stagnating levels of soil organic matter accumulation.

That makes complete sense. And keep going on your soapbox if you have more to add. I think when as you were saying that, it was just really clicking in my head of, I think of the KISS method. Keep it simple. And that is what you're talking about with foundational principles. I've heard it said, if you're trying to learn any new topic, don't complicate things by learning all the little details. There's usually some set of foundational rules or. And rules isn't the right word here, but I think you know what I mean. Foundational principles that lead you down all kinds of other paths of actually applying those foundational principles. And that's what you're talking about. And you know, you said we were talking about grazing, but that can apply to everything. That can apply to anything.

Absolutely. It's when, when you, when you understand the foundational principles well, you can work in almost any setting. And you know, I've shared this story a number of times, but now, five or six years ago, we started working with cotton. And as a team, we knew zero about cotton. I could have stumbled across a cotton plant and not recognized it, had no idea what it looked like. I'd never seen it. And in our first year working with cotton, having no history and no knowledge, we got a 50 to a 60% yield increase across the cotton farms that we were working on. How do you do that? Well, we just looked at this first principles like when are the reproductive buds of a cotton plant being formed? What are the nutrient ratios that are optimal for reproduction? How do we minimize stress? How do we position stress at different stages of the plant life? It's like those foundational principles are transferable across any reproductive crop. And similar to those reproductive crops, there are foundational principles for a vegetative crop. If you're growing spinach or kale or alfalfa or grasses, then the ideal is to not have them be reproductive, but to have them be vegetative and remain vegetative. And there are foundational principles for nutrition management of how you accomplish that. You, you manage nutrient ratios differently. So I know that I can, I can put on foliar applications to make an alfalfa plant go from a cutting to bloom in 14 days instead of in 28 days. Or I can do the opposite. I can put on foliar applications that will delay bloom from 28 days out to 36 to 40 days. All of that is a reflection of nutrition management of some foundational principles of how particular nutrients influence plant hormonal expression. Anyway, I could go on and on, but I think if there is, if there is a new frontier for grazers to explore that holds tremendous untapped potential, it is figuring out the things that Jim Elizondo has figured out and figuring out these first principles. This first principles thinking, what are the foundational principles of optimal grazing management based on stage of plant physiological development? I think we do that and all of a sudden this conversation about take half, leave half, we will say, oh, this makes a great deal of sense. Yes, this is exactly what we do, what we need to do for optimal performance if we are grazing livestock on fescue in Missouri, but if you are grazing this species in Minnesota, take half, leave half makes no sense whatsoever. And all of a sudden it will make, we will understand that the way some of these ideas came about had merit. They were not meaningless. They weren't just contrived. They were based on real world experience in a certain setting. But then when you change the setting, the rules also need to be changed. And you can more easily change those and know precisely how to do it when you understand the foundational principles behind them.

I think there is so much that will happen in the crazy movement with these principles. I, I really am interested in that. Will that also be a part of like Fuel Lark? Will it tell you the plant species.

You cut out a little bit? Mike? Will it tell you the plant species.

And what, and, and like the, the optimal time, you know, at what stage? What is the foundation principle of that species, you know, for to keep it vegetative?

Based on this conversation, we're going to expedite the training of that so that we can develop it. But it will probably be based on plant signals, I would guess because these, these vegetative plants, unlike a reproductive plant, has specific stages. Specific reproductive stages. A Vegetative plant has plant signals. So I'm, I'm just pulling stuff out of the air here. One signal on, let's say maybe a forb plant like plantain or chicory is when there is a change in leaf size ratio, when the last fully developed leaf becomes smaller than the leaf that emerged before that. Or perhaps in the case of grasses, when the lowest, oldest leaf starts losing chlorophyll, there will be, there will be physiological expressions of sugar flow inside the plant that we can observe. And this is, this is exactly the type of thing that I, that we have already trained lark on for the reproductive crops and that would be awesome to get into for the forage crops as well, is what are those exact physiological expressions for different types of plants? And also remembering that, you know, it's interesting much. There's a lot of, there are a lot of really wonderful wise people with a lot of knowledge in the world today. And also some of the best observations that were made and that were actually written down were made 80 to 100 years ago and sometimes on very different forage species. So now we have to take that old information and we have to adapt it and update it and say, okay, this worked for brome grass when they were grazing a lot of brome grass, but how do I adapt that information to these different forage species?

This, this is getting me a little. Sorry, Brian, go ahead.

I just gonna make two observations here that kind of ties it all together in my mind at least that, you know, with field narc. What I envision happening here is a, is a mindset change. And maybe, John, you deal with a lot of people that already understand the foundational principles, but I know there's a lot more that need that level of education, which is where I think Fieldmark can help wrapping it back around to asking the right questions. However, in order to apply all of those foundational principles, what do you need to do? You need observation.

Exactly. Exactly. That's where the human component comes in.

Yep, yep. So I think our, our, you know, we got down the rabbit hole of talking about grazing, but I think it was a perfect encapsulation of, you know, how this can be applied and the idea behind it and then what it takes behind, you know, for that observational piece of. You're not just relying on a piece of technology, you're relying on that piece of technology to send you in the right direction.

Exactly. To make you a better observer.

Yep, yep. Well, John, I, I really appreciate the time here today. And two, two things would be where can folks access fieldmark would be first question and then second one is of course I do. We didn't mention it at the beginning, but you've mentioned it here here in the. In the recording that you also have your own podcast, a regenerative agriculture podcast. Wanted to make sure that we said that where more and more information can be found as well. But back to the fieldmark question. Where can people access that?

People can find it just by searching for it online. Searching for Fieldlark. The URL I think is Fieldlark AI and you can also find it on the Advancing EcoAgriculture website. And I haven't posted the link yet, but I need to also put it on my website@john kempf.com so there's a bunch of places you can find it and prepare to be delighted.

And if people want to give feedback, you mentioned that it can be done through the AI, but then also is there any other avenue that you wanted to send people to to give feedback on. On the tool itself?

The team at AEA has built this and developed this. So you can communicate with the people at aea. You can communicate with me directly. I'm fairly readily accessible. My contact information is out there. I'm not going to share it here. But you search for it, you can find it.

Well, thank you, John.

Thanks.

Thank you, Brian. Thank you, Mike. This has been fun as always.