In this episode of the Nutrient Management Podcast, four U of M researchers discuss the Maximum Return To Nitrogen (MRTN) approach to corn N rate guidelines. What is the MRTN and why was this system implemented? What are the pros and cons of the MRTN approach and how is it performing? What should corn growers know about alternative approaches to the MRTN? Will there be any changes made to the MRTN approach in the future? Thank you to Minnesota's Agricultural Fertilizer Research and Education Council (AFREC) for supporting the podcast.
Welcome to University of Minnesota Extension's Nutrient Management Podcast. Each month we bring you the latest research in nutrient management for crops and how you can incorporate the latest tips and best management practices to your farm.
(Music)
Paul McDivitt:
Welcome back to University of Minnesota Extension's Nutrient Management Podcast. I'm your host, Paul McDivitt, Communications Specialist here at U of M Extension. Today on the podcast, we're talking about the maximum return to nitrogen, or MRTN, approach to N rate guidelines. We have four members of Extension's Nutrient Management team. Can you each give us a quick introduction?
Dan Kaiser:
My name's Daniel Kaiser, I'm a nutrient management specialist located at the St. Paul campus. One area that I do work on are the nitrogen guidelines. So the MRTN is one of the things we're looking at now that we'll be talking about today that I work with, with those guidelines.
Fabian Fernandez:
I am Fabian Fernandez. I'm a nutrient management specialist as well, located in the St. Paul campus. And my area of emphasis is in nitrogen management. And I collect a lot of data that goes into this MRTN calculator and I work with that data, give it to Dan. And Dan is actually the one that composites all of that data together with other trial and puts it for the calculator.
Brad Carlson:
I'm Brad Carlson. I work out of the regional extension office in Mankato. I'm an extension educator, and I do the lead on our Nitrogen Smart education programs, which really emphasizes flexing your management based on conditions and so forth. And so we look extensively at the MRTN approach. But then I'll also talk about why adjustments need to be made under different circumstances.
Jeff Vetsch:
Hi, this is Jeff Vetsch. I'm a researcher here at the Southern Research and Outreach Center in Waseca, and I've accumulated a lot of this N rate database or data over the last 20 some years. And N management, nitrogen fertilization of corn is one of my specialties.
Paul McDivitt:
All right. So starting off, what is the MRTN and why was this system implemented?
Dan Kaiser:
Well, the MRTN, what that stands for is maximum return to nitrogen. Sometimes you're at meetings, we'll talk about MRTN or what we'll talk about is the economic optimum nitrogen rate, which essentially are the same thing. What the MRTN does is... or really why it was put into place was to factor in economics. So looking at essentially the return on investment for nitrogen, because if you look at a normal nitrogen rate response curve, it's what we call a quadratic plateau, where the response will rise towards a maximum, and once it gets towards that maximum, you'll tend to see less yield produced per unit of nitrogen applied. So really what the MRTN is trying to do is to determine what it deems to be the economic optimum nitrogen rate, where we're managing around that plateau value, or around that point at which you'll hit a point at some point where the response to nitrogen is negative, where you're not getting... or the dollar you're investing in nitrogen isn't giving you a dollar back in crop value.
Dan Kaiser:
So why this came into play was, when you start looking at it, we've had some of these price spikes, like what we're seeing, before. So there were some discussions among some of the regional researchers in terms of implementing a practice where we could factor in economics, because it wasn't done before. Most states had some factor times yield that they were using to recommend nitrogen. And what a lot of the researchers were seeing is really when you start looking at just a yield goal based system, it wasn't working out. And if you look to data, there was really no linkage between the optimal nitrogen rate at a given site and yield goals.
Dan Kaiser:
So that's where they decided to look at another approach and make the recommendations data driven. I think the key for a lot of this, when we start talking about the MRTN approach is that it is a data driven system where we're continually adding data to a database that evolves over time to match environmental conditions, hybrids, and those type of things that we can factor that in. And it's a little more defensible than just using a simple factor based equation that I don't know if anybody really knows where that came from. So it's one of the things that digging back into some of the data, some of those older recommendations, it was getting harder and harder to defend them.
Fabian Fernandez:
Yeah. Well, and in addition to that, there was an interest in trying to come up with a regional approach to the guidelines, which definitely the MRTN approach is going in that direction. We still have recommendations based on each state individually. So there are state boundaries that make changes, but it is more of a regional approach where we are all using the same approach. The reason the data is different between different states is because each state is responsible to collect their own data. And obviously the responses that you get with the different databases make differences. And yes, as Dan mentioned, the big impetus behind this approach was that when you look at the relationship between yield and nitrogen rate needed for the yield, there was no relationship. And so there was data and everybody that was doing work in the different states had their own approach to the yield goal.
Fabian Fernandez:
And those recommendations were certainly based on data, but that's where over time things started to change. I don't know if it's because of the hybrids or different climatic conditions or different management systems that have evolved over time. But the reality is when you look at the yield compared to the N rate needed to get to that yield, there was no relationship whatsoever and so this approach was started. And it was rolled out, I think in 2005. That's when we started with this approach. Before that it was for the most part in the states that are participating in the MRTN approach, they were all using yield goal.
Dan Kaiser:
Well, and I think you're looking at the yield goal based system, a lot of that stuff dates back to a time when fertilizer was relatively cheap. And if you start looking at it now, pricing wise you see a lot of fertilizer pricing follow... I mean, the trend really is to follow crop price. So if it isn't as cheap anymore, really, it just made sense to factor in economics at some point, particularly with nitrogen, since nitrogen we get a clear response curve. We look at other nutrients like phosphorus and potassium, we can put rate studies out with seeing six, eight rates and you get a yield response to that first rate and that's it. So you don't get these clear curves that we see with nitrogen that make it really easy to start factoring in economics.
Dan Kaiser:
As fertilizer prices have shot up, that was just one of the things that really made sense. I think one of the things that a lot of people think about it too is that when you're managing it, if you dial back, if you look at what we call the agronomic optimum, which is the maximum yield produced by the maximum nitrogen rate regardless of economics, is that typically when we're looking at our MRTN values, we're within about 1% of what we have in terms of maximum yield at the given site. So it isn't as if we're giving up a lot, particularly if we're dealing with a 0.1 price ratio. The thing we'll have to see is what the price ratios are in the spring of 2022 compared to... just to see how things are at, because I know when Jeff, you were talking at some point, I know with one of your meetings, you were looking at some of these figures. And we're looking at 0.15, 0.2, the way things are sitting right now, which really starts to dial back on the nitrogen.
Dan Kaiser:
So that's one of the things that we'll see what happens in spring, because looking at it in terms of following it, you start looking at a bigger dial back with the price situation we're in. But normally everything tends to manage or moderate around a 0.1 price ratio if you look at it historically with the price of nitrogen to the price of corn within the last 15, 20 years.
Brad Carlson:
Well, another factor that went into adopting the MRTN approach, if we go back to the time when that came in place was there was still a lot of use of the old formula of 1.2 times your yield goal, minus your nitrogen credits. And I guess a couple of factors, Fabian talked about the increased nitrogen use efficiency of corn hybrids, the whole 1.2 thing really wasn't holding, it wasn't taking that much nitrogen anymore to produce a bushel of corn. But the other aspect is that formula also assumes that you could just keep right on applying nitrogen into infinity and keep raising your yields. And that's really also what Fabian and Dan were referring to related to their not being correlation between the increased use of nitrogen and yield. That you can't just keep putting on nitrogen and see the yield go up. Obviously that's what the response curve is. But it wasn't even predictable from one field to the next. It just simply wasn't working.
Brad Carlson:
And so while most of the land grants had shifted to a yield goal system with categories in it, depending on the needs of each state, the old formula was out there. I know there was at least some seed companies, excuse me, that were still pushing that. And it was probably important for that to go away too.
Paul McDivitt:
What are some of the pros and cons to the MRTN approach?
Fabian Fernandez:
Yeah. So that's an interesting question. And I always say that there is no perfect approach, and every approach has their own set of conditions that are good and others that are not so good. One of the things that I think is very valuable about the MRTN approach is that it's actually very transparent. You can go into the calculator and you can look at the distribution of the response curves that were used to generate the database, not the actual response curves, but you can see the frequency where a certain nitrogen rate would provide the maximum return to nitrogen. So you can see that kind of distribution, you can see the relationship between yield and N rate as we were talking earlier, all of these things that before were really presented. Before the only thing that a farmer had was basically an equation with a factor, typically 1.1 or 1.2 times the yield or something like that. Now farmers have the ability to look at the database to a certain degree.
Fabian Fernandez:
And then the other part too that I think is a benefit is that it allows farmers to deal with risk. Nitrogen management is a risk management. And what we have in the calculator is that economic optimum. But then we added a plus minus $1 per acre on that economic, so that if you are risk avert or risk tolerant, you can figure out what that rate would be. And like I say, everything has a good side and maybe a side that is not so good, because on one hand, this allows farmers more flexibility, but from a regulation standpoint, regulators don't like that because they like to have a specific number, just one number. And having a range creates some difficulties in there.
Fabian Fernandez:
The other thing that I think is interesting is that it needs a lot of data. This database has to be large in order to work. And that again is a pro and a con. If you have a lot of data, it's a benefit, but if you have limited data to make the calculation, then it could be a problem because any new site that you add into the database can skew the data substantially depending on where it is. The other thing that it's, again, pro and con depending on where you're looking at is that it's a generalized approach. For instance for Minnesota, we have the N rate calculator for continuous corn or corn soybean. Those are the only distinctions, but it's for the whole state. And so there are changes differences in the state.
Fabian Fernandez:
Of course, we have added now the sands. They are a separate thing that is not part of the calculator, but we deal with those differently. But it is just one approach. And with the increase of interest in precision ag, that can be a challenge because I always say that the benefit of precision ag is that it's very precise, it's specific, but that's the challenge as well, because then you have an approach like the MRTN database that is general for the state, and you can maybe not use it perfectly for every situation in every field. And so that's one again, pro and con.
Fabian Fernandez:
The other thing is that there is no differences other than continuous corn, or corn soybean. We don't have any other management consideration in the calculator. There is a lot of interest now for instance into split applications or having different rates based on geographic location or different kinds of soils and things like that. And the challenge to do that is that we have looked at it and I like John Sawyer at Iowa State and Emerson Nafziger at Illinois have looked at this in... that they have larger database to see if they can part out some of these things. Separate the database on soil types or regions or things like that. And where that makes sense, they have done some of those changes, but where there is not enough data then you're locked into having to leave it the way it is, because simply you just don't have enough information to determine whether there should be a change in there.
Fabian Fernandez:
And then the other thing that I can think of as a pro and a con is that some people argued that this is purely an economic thing, and it doesn't take into account the environment. And I would argue that that's not the case. When you are using nitrogen effectively so that the plant uses the amount of nitrogen that you're putting in there, the environmental impact is pretty minimal. And we have seen some of that in relationship to nitrate leaching in terms of residual nitrogen in the soil, things like that, where if you're over-applying, if you go above the MRTN, that's where you start to have environmental issues. If you're below that, you are not really improving anything environmentally and you're reducing yield.
Fabian Fernandez:
But those are some of the concerns that people have when they look at these things and say, "Well, it's an economic and so it's all driven by economics and the environment, then it's not part of the equation." But in reality, we need to remember that even though it's maximum return to nitrogen in terms of economics, the data behind those calculations is based on the agronomic response. And again, if the crop uses that nitrogen, then there is no nitrogen loss to the environment at that point.
Jeff Vetsch:
I would add that I think one of the big positives of this MRTN approach is that it is a database and it's updated regularly. So it's not a stagnant model or a equation that was used for 20, 30 years and we can't figure out where the derivation of it came from. The other big positive, I think is the fact that you can calculate these price scenarios, which today's changing prices are interesting. You can put in some different scenarios and see how it changes. And as Fabian said, that helps you think about your risk and what do you feel comfortable with?
Jeff Vetsch:
I think from cons, the fact that we don't currently use it for irrigated sands I think creates some confusion, and that could be a negative. And the uncertainty of how many sites are needed to break off into a separate region or area, that's also something that we have to deal with and we give a lot of thought to. And then from the actual input of the data, we know that what model we choose helps or does influence the economic optimum. And that's something we think about and we look at, and we critique probably more than we should, but it makes us scratch our heads sometimes, and that's on the inside of the database.
Jeff Vetsch:
But as Fabian said, the best, I think positive is that you get a chance to look at that probability distribution of what the optimums were that are in the database. And that really can help you if you're risk averse or willing to take some risk because you're not going to be right all the time, but where you want to be is where you feel comfortable with, well, if I'm right 75% of the time, then this is the distribution, and this is about the rate that I need. And that gets back at that acceptable range.
Brad Carlson:
When we do the Nitrogen Smart program, we will occasionally get a little bit of pushback from farmers just simply related to the economics part. You'll occasionally hear, well, just tell me what the response to nitrogen is and I'll worry about my own economics. And I understand where the farmers who have that opinion are coming from. I think from an educational standpoint though, really the strength of the MRTN approach is that it acknowledges that the response curve gets very flat at the top, just the same way that I just talked about a little bit earlier, how you can't just keep pouring on nitrogen and see your yields go up into infinity. We know at some point that yields just stop going up when you reach a certain amount of fertilization. And in particular we will see yields continue to climb but very slowly to the point where you're adding lots and lots of nitrogen without a lot of yield improvement.
Brad Carlson:
And it gets very difficult to say just exactly where that is, not to mention the fact that, of course, from an environmental standpoint or even an economic standpoint, it just makes no sense to keep pouring on nitrogen to get that last one or two bushels. And then from a research standpoint, of course, it's very difficult to even establish where there's a one or two bushel difference reliably. And so the MRTN approach incorporating the economics into it, and the price of corn, the price of nitrogen just simply allows you to visualize that law of diminishing returns when you're applying nitrogen. And we like to stress when we do the Nitrogen Smart program that that rate recommendation window is plus or minus a dollar. A couple of pounds of nitrogen and really a fraction of a bushel of corn. And that's a pretty wide window.
Brad Carlson:
So from the farmer's standpoint, there's a lot of flexibility there. It really just gets into understanding your own circumstances. And that of course, is a weakness for the individual farmer to have to try and figure out just exactly where they are at in that system.
Dan Kaiser:
Well, and I think Fabian brought up one other key points there too, it's that the variable rate thing is really, I think the bigger issue. Right now since farmers do have that option, really for us trying to figure out how to recommend variable rate and with a system where we have essentially a single recommended amount with a range, we could recommend growers vary the rate within that range. But realistically there's likely going to be more variation within a particular field. So it's difficult.
Dan Kaiser:
The issue though looking at it is how do you predict some of that variability? And we're just not there. I mean, Fabian, I know you've looked a lot at different factors and that nitrogen cycle trying to predict inputs and outputs. And if you could come up with some accuracy of that, I think we could give growers an idea in terms of within maybe this MRTN structure and how to vary similar rates within fields, but it just isn't there right now. So I know that's the biggest probably con I see out of it right now is growers asking us, I have variable rate, how do I use this? And there really isn't a straightforward answer with that. I mean, a lot of it goes back to what we typically would recommend is starting with the MRTN and making adjustments based on knowledge you have within a particular field is really the best thing you can do.
Fabian Fernandez:
Yeah. And I was going to mention that as well, that when you look at this, you could say, well, this is a brute approach to management, because more data is better. But the real challenge is all these variables that impact nitrogen availability. Like Dan mentioned, I've been doing a lot of work trying to understand the different parts of the nitrogen cycle to have better idea of what are the inputs and the outputs and things like that. And our ability to predict these things is very limited.
Fabian Fernandez:
And we have looked at this in many different ways, putting a lot of different variable into the equation; weather conditions, soil types, all sorts of variables. And at best, we are able to predict at about 0.57. So if one is perfect relationship, about half of the time we are able to predict it using all the information that we can possibly think of; early season conditions, mid season conditions, precipitation, temperature, the soil type, how much nitrogen is mineralized. All those things and we are still not really able to predict these things very well, partly because we are not very good at predicting weather or forecasting weather in a way that we can use that information. By the time we apply a hundred percent of our nitrogen, even if we are split applying, we still have like two month ahead of us where we have no idea what is going to happen.
Fabian Fernandez:
And so having this approach where you have a large database, and that's where I think there is the benefit is that allows you to say, "Okay, based on all the potential variabilities that are out there that we can really predict very well, what is the chance that I will be approximately right most of the time?" And so that's, I think the real benefit of having this calculator. But again, it requires a large database in order to start to have some confidence that the values that we have there will be representative.
Fabian Fernandez:
And one thing that I would say in terms of progression of approaches, we start talking about why this approach came about in about 2005. If you look back, I look at the history of recommendations from the university, and we went from these ivory tower approach where the university will say, "Well, this is the right number," and everybody follows it, to understanding that there is a lot of variability, that there are a lot of unknowns. And again, as I mentioned, the transparency in these approach is I think very valuable because people can see that it's not a perfect approach.
Fabian Fernandez:
We never come across saying, "Well, the MRTN calculator is the approach, and you should use it." We recognize that there are a lot of differences for specific fields. And the thing that I always invite farmers, especially with all the tools that are available out there, is to start with the MRTN calculator and do trials of their own in their farms over the years. Because again, one year data might be completely different than the next year in the same field, but doing these things over time to really adjust the MRTN to their specific conditions. And if they do that kind of research and they will have the confidence to say, "Well, in this particular field, I need to go up or down relative to the MRTN because of the history that I have collected over the years."
Paul McDivitt:
How is the MRTN approach performing?
Jeff Vetsch:
I think in the sites that I've looked at over the last, oh, I don't know, five to 10 years, it still does a pretty good job in corn after soybeans. And it's frequently plus or minus the acceptable range. It's within 60 to 70% of the sites, it's still getting them right. Or they're just at the edge of either the lower side of the acceptable range or the high side of the acceptable range.
Jeff Vetsch:
I think where we're seeing more challenges is in our corn on corn sites, especially on our poorly drained soils. And it gets at the amount of N that has to be in that plant, more of it has to come from fertilizer. And the soil's ability to provide some of that N is more unpredictable in continuous corn or corn after corn. And that's where I've seen more challenges and wide ranges in optimums across a 10, 20 site database. But there's a lot more sites in the database than 10 or 20.
Fabian Fernandez:
I agree with what Jeff mentioned, and I think the reason the approach is still performing well is because of what I think Jeff mentioned earlier that the fact that we are able to continue to add data to this database, because with these guidelines, there is absolutely nothing that is static about it. Everything's changing all the time. The hybrids are changing, climate is changing, the way that we do agriculture with tillage and other things is changing. And the fact that we are able to continue to add new data into that database, it allows us to maintain a tool that doesn't become obsolete. And I think that that's a huge benefit compared to earlier approaches where it was set, and then it would just stay in stone for years before somebody, again, did some research to revisit those questions. So that's, I think part of the reason we feel that the approach is still very good. And we do see and we've seen it here, we see it in other states as well, where the N rates seem to be creeping up. And that's all related to all these variables that impact corn nitrogen response.
Dan Kaiser:
And Fabian, on that creep that you're talking about, that's one of the things we've seen consistently in the database, particularly for continuous corn, and I know Jeff brought that up. And you look at the data. So the MRTN started, we said around 2005. Our database stretch back, I think down to 1990 when the initial database started. And if you looked at from about 1990 to 2000, corn, corn was pretty consistent at about 160 across all the sites. Then you look at around 2000, then we start seeing almost a linear increase just yearly in terms of the optimal N rates at that given point in time. And I've really been looking at that quite a bit and then trying to break our database down looking at it in 10 year chunks, just to see what's happening. And corn soybeans been relatively flat.
Dan Kaiser:
I mean, certainly the last, we'll just say five, 10 years, we're just a little bit of a jump up and we've seen it climb a little bit here, I think in the last five years. Just at the time we're recording this, I went through and I was looking at the 2020 data, just my data and Jeff's data added in, and we're looking at MRTN jump of about seven pounds on that. So we're seeing that with some of these wetter years where that... just to give a little jump for the corn soybean. But the continuous corn, again, it's been a steady climb and that's one of the things I'm struggling with, because I know there's some spots that are likely higher than what we have recommended now. The question though is, and I think this was brought up, is really do you start with the MRTN and make adjustments, or do we recommend more nitrogen?
Dan Kaiser:
And I think we need to be really certain that that recommendation is consistent enough for additional nitrogen across those sites before we do that, because the other option is just saying, if given circumstances are present within a given year, then apply more N. And we've got some options, I think, to potentially do that because we really just want to try not to have just consistently too much nitrogen applied just for insurance purposes. That type of strategy is really not easily justifiable right now, just with a lot of the issues we have environmentally going on. So that's the thing. I mean, that's what we'll be looking at here moving forward is some of these sites, because we know again, there's some sites like as Jeff said, continuous corn that's been more of a challenge.
Jeff Vetsch:
Yeah. To follow up with what Dan mentioned. I'd say from 2016 or 2014 through 2020 is that trend where we saw a wetter than normal period across much of the states corn growing regions and that creep of those N rates going up and up. But interestingly enough, the last couple of days, I've been calculating our results for this or tabulating our results for 2021. And I would say in general, all but one of our sites actually had economic optimum calculated values that were considerably lower than they have been in previous years, and pretty much in line with the middle or the lower side of the acceptable range of the MRTN for both corn on corn and corn after soybeans. And it does make some sense because it was a year that did not have a lot of nitrogen stress, and that aligns with this. It's not going to be right every time, but if you can adjust it on the fly or in season, that's better than putting on that insurance and as Dan said, up front.
Paul McDivitt:
Can you talk about alternative approaches to the MRTN?
Brad Carlson:
Well, Fabian alluded to the issues with MRTN and variable rate nitrogen and that gets down to the... Maybe instead of saying, gets down to, it gets out to the big picture of where we have issues with it. And that just simply gets into what is specifically going on right at the individual plant level. Now nobody's going to advocate getting precision to the level of single plants. But the question becomes, in smaller areas in the field that we are seeing variability from one area to the next, and is MRTN just simply washing it all out.
Brad Carlson:
It's interesting that for years farmers would say to us university folks that we want to see more large field research because your small plots are not representative of whole fields. And now we're doing precision ag work and you're saying, well, we do whole field work. You're washing out all the differences across the field and we're not finding where there's differences from one area to the next. And so some of the variable rate nitrogen work that I've worked with and processing yield maps over the last few years, we've found some really incredible yields with some fairly low nitrogen rates. It really gets into what Dan was talking about, being that when we see some variation from what the optimum rate is in a site or in a year, or in a field or whatever that is, can you explain that? And was that something that was predictable? Obviously predicting the weather as has already been alluded to is not easy.
Brad Carlson:
But beyond that, there are soil conditions, there's general weather trends. We know that for instance we're coming into this year with a soil moisture deficit. We're not likely to see a lot of pressure related to denitrification coming into this growing season. Jeff talked about the MRTN for last year being a little bit lower. We would expect that under dry conditions and so forth. And so realistically it becomes more of, I guess, what we talk about with Nitrogen Smart is, an educated guess. It's not just simply guessing, but it's taking the information that you have available and making adjustments. It's not assuming that for instance, MRTN is be all and end all. But if you're going to flex your rate higher or lower, having some justification and some reasoning behind that.
Brad Carlson:
Ultimately speaking, I think some of the crop models that are available commercially will perform this function. They actually record what really happened on the field as well as what all of the variables that went into that, whether it be from different hybrids and different crop protection standpoints and the fertilization and so forth. And long term big data mining of some of these data sets may actually become very useful and predictive. But at this point, I don't think we're quite there yet. And so we get back to like I said, just doing a lot of educated guessing if we're going to start flexing our nitrogen rates beyond where we're at with general MRTN rates.
Fabian Fernandez:
And not feel complacent about the approach because we certainly are not. We are always trying to figure better ways to get this tool to improve so that we have a good, useful tool for farmers. But one thing that you notice is that every competitor, if you will, if you call them competitors, they are comparing their approach to the MRTN. And so that, I think it's some evidence that this approach is as robust as it gets at the moment. When you look at top brands in any kind of market, everybody; the competitor, the generic brand always puts compared to, and puts the top brand. And that's certainly what's happening with the MRTN approach. So I think there is a lot of value. It's not perfect, but it is probably the best that we have right now.
Fabian Fernandez:
And Brad alluded to this with the models, there are different models out there that are used to predict nitrogen. And the big difference between the models and the MRTN approach is that the MRTN approach integrates all the variables that impact nitrogen response into that value that we provide as the MRTN value. Because it comes directly from data collected in fields, in real fields, with real weather, with real nitrogen dynamics. And the models, what they do is they predict some of these things that are happening in the field and put it into a specific thing. So where you have to input how much rain is happening or has happened, and what is the temperature and all of these different things, and then it estimates based on all of that. So it's a differences in approach. One integrates everything and says, okay, this is what happened this year, and there are all of these conditions, and this is where the MRTN is.
Fabian Fernandez:
The other approach is more of, okay, splitting up every single component and predicting what may happen or may not happen given the conditions that are inputted into it. And just like we talked earlier with the MRTN that it's generalized to a certain degree. These things with the models, this is the same thing, because the response to mineralization or denitrification or leaching or all of these things are parameters that are estimated, they are not focused specifically on the measure impact. And I think farmers sometimes feel like... they get this false sense of security with a model where you can see, I don't know, a needle moving forward or backwards or whatever based on the conditions that happen every day to determine what the nitrogen rate is. But the important things to recognize is that behind that number, there are a lot of assumptions that are estimated guess, best possible guess, but they are still an estimation. It's not the actual.
Brad Carlson:
Yeah. I think the one thing that's going to be interesting with the crop models as we go on, Fabian, as you mentioned, they're estimating a lot of factors like mineralization, leaching, denitrification and so forth is that then they are obviously collecting or inputting the data at the end of the season on yield. And so they are able to look at themselves and decide how they performed. And so as I mentioned over time, there will be a big data feature of these, where they have self-corrected based on their own performance. And we would anticipate them getting better and better as the years go by.
Fabian Fernandez:
For sure. And the reason I become skeptical about some of these things is because I actually have seen some of these assumptions. And I have seen these assumptions in my own work where, let's say that I am trying to really understand mineralization, and I do studies should really focus on that and everything as possible in a field that I can control, I am controlling or measuring. And then I try to predict based on all of that information what will happen next year. And I looked at what actually happened the next year, and I get completely different results. And that's where my skepticism starts, because even in a specific field under specific conditions, we are not able to predict things very well. What is the chance that a model that obviously is using much more generalized data to predict these things will be able to make a better estimate? It's very limited.
Paul McDivitt:
What changes, if any, do you see for the MRTN approach in the future?
Dan Kaiser:
Well, one of the things we've been looking at, and I think we've alluded to this, some of the things that we've seen in different areas of the state, I think do bear looking at more regionally splitting the database out and looking at whether or not we have a different set of guidelines we can use. I mean, I did this back in 2015 when I took over work on the MRTN database looking at south... I wanted to really look at southeast versus south central, particularly for continuous corn, because we knew at that point in time that the numbers were going up. I wanted to see if there was any differences. And at that point in time, there really wasn't any data that says that the two were any different.
Dan Kaiser:
I think when I looked at them, the MRTN values for continuous corn, for loam soils in the southeast versus more poorly drained soils in south central and southwest were within a couple pounds. So I need to look at that again. I mean, that'll be the big thing. Brad, I think brought up some of the... or Jeff brought up the irrigated sands. The recommendations we have right now in the irrigated corn publication are based on MRTN. It's just a very small database. And I think one of the things that we need to be looking at doing now that Fabian and I have more data is looking at putting some of that data into the actual calculator that growers can use it.
Dan Kaiser:
The problem at that point in time, I remember John Lamb was working on this, he really had was the fact that the most consistent data he had for irrigated sands was continuous corn. So that's why you see some crediting still for some of the other crops off of the continuous corn recommendation for say corn soybean, because either there wasn't enough locations or the data wasn't consistent enough to come up with a recommendation based on the locations you have.
Dan Kaiser:
So I think that's going to be the main thing that you'll see splitting off potentially the northwest part of Minnesota, which if you're using our MRTN approach, I'd really look at the lower end of the profitable range for northwest Minnesota, because that seems to be an area that's different. The rest of the state is just going to depend on what the data tells us. Again, the thing I like about this approach is that we can let the data tell us what we should be doing moving forward. So that'll be the big thing with it.
Dan Kaiser:
In terms of anything else, certainly the thing about the MRTN is we can factor other costs into say nitrates and drainage water and those type of things. I mean, there's no plan for that. But the nice thing about the approach is you can factor in other things. And as Jeff, I think mentioned before, you could look at different fertilizer prices. There's just some flexibility with it that is nice for that. And then we could always build some other cost factors in if we ever have to. Again, I don't see that happening, but there is some flexibility there.
Dan Kaiser:
So that's the thing I think over the winter here I'll be looking at is now I think we have enough data to look at petitioning some of this areas out. The bigger question on my mind, and this was brought up before is, if we see differences, what do we do? And that's where I really want to see this 2021 and even the 2022 data before I really make a large wholesale recommendation in terms of recommending more nitrogen in some areas, just to see if we get back closer to those MRTN rates, particularly with continuous corn, because we talk a lot about the MRTN being a starting point and making adjustments from there. And that's really where I look at this as, and I want to make sure that the recommendations that we don't go and well over recommend nitrogen in some of these areas if we're making changes.
Dan Kaiser:
So again, this is something that it isn't where on a whim we make changes with this. There's a lot of thought process that goes behind what to use in terms of the data set and what to do. And so it's something that we do spend some time on, and that's one thing we'll be doing again. I think this winner, since we're talking about this more, since we're about 15 years in, it's doing a reevaluation, particularly within the state.
Fabian Fernandez:
And the other thing that I would mention that I get this question asked a lot is about split nitrogen applications. And right now we don't have that. And I don't foresee making any changes in that regard because simply we see that there is no much difference between a pre-plant application versus split application, if we are doing a pre-plant application at the appropriate time, close to the time of planting, around that time. And the only times where we see a difference between a pre-plant and a split application is in the sands, which that's part of what we have in the publication for the sands. But for finer texture soil, which you simply don't see much of a difference. The only time we see those differences is when we have extremely wet conditions. And so again, we go back to this thing of not being able to really predict things very well for those situations specifically. And so it doesn't warrant right now to make a different set of recommendations for a split application compared to a pre-plant application.
Brad Carlson:
Yeah. We get that quite a bit with Nitrogen Smart. We'll have guys ask us if they can lower their nitrogen rates if they're split applying. And the research does not necessarily bear that out. The one thing though that I would say is a caveat to that though is the extent to which maybe farmers are over applying or applying a little higher rate or going higher in the rate window or so forth that can adjust themselves back down or if they're not comfortable with recommendations, and now they're willing to be comfortable with the recommendations because they're split applying. Then the overall effect is you are lowering your N rate, but we're not necessarily lowering our recommendations. And as far as being comfortable with actually coming out with recommendations to apply less, because you're split applying, I have not seen that we've got data that we can really comfortably go on that. I guess individual farmers may have enough experience in their own fields that they can feel comfortable with this. I happen to know farmers who do do that and are comfortable with it, but that's not for everybody.
Jeff Vetsch:
I have a few site years that do show that that was possible. But as Fabian said, it's not consistent enough or enough site years to justify a separate recommendation in reducing the rate.
Paul McDivitt:
Any last words from the group?
Dan Kaiser:
Well, one of the things that I would just want to say is that we talk a lot about the MRTN and rate is one of the four Rs. And a lot of our rate recommendations really are contingent that essentially all the other Rs are being... the source, placement, timing are being followed that are more optimal for your soil to prevent loss. So obviously when you see some year to year variation of that, but again, it's only one of them. And it gets focused on a lot, particularly by environmental or people that are worried about nitrates in the surface water, but it isn't the only practice out there that we need to be worried about. So it's one of the things to bear in mind that I think it's the easiest one to pick on because it's the easiest one to quantify sometimes, but it's not the only R when it comes to nitrogen management that needs to be focused on when you're looking at what's the best option for a field.
Jeff Vetsch:
I would add that I'd love to see farmers if they haven't looked at it and don't know what those MRTN rates are to take time to look at it. And there are programs available, the Nutrient Management Initiative, and some others that will actually provide financial support for them to try their rate versus an MRTN rate and a side by side strip trial across the field. Do it for two or three years with that financial support, if you can ask around and find a program that's in your area to do that. There's very little risk involved if you can do that and see how it compares.
Paul McDivitt:
All right. That about does it for the podcast this week. We'd like to thank the Agricultural Fertilizer Research and Education Council, AFREC, for supporting the podcast. Thanks for listening.
(Music)