Subscribe to the podcast on your mobile device: Apple Podcasts Spotify Stitcher Google Podcasts YouTube iHeartRadio
The purpose of this episode is to dive into the dos and don'ts of conducting research trials on your own farm whether that is for hybrid selection, rate determination, product effectiveness, or machinery value. Laura Thompson, co-coordinator of the Nebraska On-Farm Research Network, joins the FarmBits podcast to discuss this important topic. For those of you who have been following along with all of the episodes of the FarmBits podcast, you may remember that Laura Thompson was a guest on the first podcast episode providing an overview of the importance of digital agriculture and the future of agriculture technology. Laura's experience with on-farm research and digital technologies in agriculture uniquely equip her to discuss how to properly conduct on farm research as well as how to leverage digital tools to execute research and measure research outcomes more effectively. While this is an important topic at any time of year, thoroughly considering on-farm research implementation prior to planting is a great first step toward gaining valuable insights for your operation this growing season.
Opinions expressed on FarmBits are solely those of the guest(s) or host(s) and not the University of Nebraska-Lincoln.
On this episode
Nebraska On-Farm Research Network:
Digital Ag Online Course: https://cropwatch.unl.edu/digital-ag-online-course
FarmBits Team Contact Info:
Samantha's Twitter: https://twitter.com/SamanthaTeten
Jackson's Twitter: https://twitter.com/jstansell87
Jackson: Welcome to the FarmBits podcast, a product of Nebraska Extension digital agriculture. I'm Jackson Stansell
Samantha: And I'm Samantha Teten, and we come to you each week to discuss the trends, the realities and the value of digital agriculture.
Jackson: Through interviews and panels with experts producers, and innovators from all sectors of digital technology, we hope that you step away from each episode with new practical knowledge of digital agriculture technology. Sam: Hello FarmBits followers and welcome to the 24th episode of the FarmBits podcast.
Jack: Planting season is right around the corner and we will be getting into planting technology soon, but before we launch in that next series we wanted to bring you an episode focused on setting up on-farm research experiments.
Sam: With the growing season getting underway shortly now is the time to be planning the execution. of any on-farm research styles that you hope to run this year.
Jack: Designing trials properly and having a plan in place early for data collection is crucial to learning something valuable from any on-farm research trial.
Sam: On this episode of the FarmBits podcast, we welcome back Laura Thompson, co-coordinator of the Nebraska On-farm Research Network.
Jack: Laura has significant experience with guiding growers through the on-farm research process and has a lot of good information offered, particularly regarding how digital tools and precision agriculture technology can be leveraged for effective on-farm research.
Sam: Now that you have a bit of background on this episode, let's get to our interview with Laura.
Jack: So, would you mind telling us a little bit more about the on-farm research network and exactly what the value it is that it brings to producers.
Laura: Yeah so, we're really fortunate to have such a strong on-farm research network in Nebraska and that got started long before I joined the university. So, the kind of formal Nebraska extension on farm research efforts began in about 1989, so quite a track record dating back that far and so the program really started out with kind of a pilot group of growers. I believe there were 12 producers in a county that signed up to participate and they each were going to do a study for three years in a row, and so each of them worked with a crop advisor and the extension educator in charge and determined what question would be of value for their operation, so what they thought would impact their productivity, their profitability for the long term and so they identified some practice and at the time some of those were things like going to no-till and things like that so kind of comparing their standard traditional practice to a new practice. So, over the years the program has really grown. We now are a statewide program. We have support from the Nebraska Corn Board, Nebraska Corn Growers Association, Nebraska Soybean Board and Nebraska Dry Bean Commission, so that really allows us to make it a statewide program and cover a lot of the major commodities that are grown in Nebraska. We typically have about 100 on-farm research studies that are completed each year and that's done by usually about 70 different operations, so some operations are doing multiple studies each year and the topics a huge variety that we get to see. So, those topics could be different things that producers are interested in looking at or that they maybe have heard about maybe new products that are being promoted in their area maybe just different technologies that they've heard about or things they visited with other farmers about so the variety of topics is you know as varied as the different producers that come up with the ideas and then we of course have some more coordinated efforts where we've got different projects like those that you're involved with where we've got kind of a set practice or technology that we're trying to evaluate and work with farmers to kind of conduct the same trial on multiple different locations. So, kind of some different different ways that the network can operate, kind of flexible in the approach in that way.
Sam: Sure, so for those outside of the state of Nebraska how has the Nebraska on-farm research network influenced the development of other networks outside of Nebraska?
Laura: Yeah so, I know that there's a lot of other states that have networks that are similar. We're working on developing similar networks, I've had the opportunity to visit with several different states over the years or even provinces actually a few times in Canada with different conference calls they're going to conferences looking at how to develop some similar network or programs to what we're doing here and then also opportunity just to collaborate doing some kind of multi-state type of on-farm research efforts where maybe a few different on-farm research programs come together around some developed protocol and try to test something a little more regionally rather than just in one specific state.
Jack: Got you, so how are those networks kind of coming together and you know I guess what do you see that potential eventually being and how that data is going to be like shared because I know that's often kind of a hard thing and research studies when you have things going on a lot of different places. How do we actually get all this data into like a single usable form that everybody can appreciate and find value in?
Laura: Yeah yeah, for sure that can be a challenge and then of course you've got all your other questions related to data privacy and sharing and things like that coming into play especially when we're trying to do multi-state efforts where that data is going for some of our multi-state efforts it's kind of been you know, we've just shared the summarized data or sometimes strip off location data to give some anonymity to that. But you know, I think as we have more and more digital technologies available that really does open up opportunities for sharing the data both in just getting that data transferred from the farms to the researchers as well as just storing even through things like our cloud-based platforms where we can all be in there and collaborating and having a shared file storage system. I mean it's kind of something we take for granted at this point but really does facilitate a lot of that collaboration with datasets.
Sam: So, you've hit on a little bit already the variety of tests that you can do, but can you maybe dive into that a little bit more of the types of studies that can be done and can a farmer come to you with any idea that they want to test and how does that work?
Laura: Yeah definitely, so yeah that's something I really love about the network is that farmers will just come to us or come to their extension educator with an idea and sometimes that's a product that they've had promoted and they might be curious or they might be skeptical you know it's just such a great opportunity to get some data on it test it on some select acres on the farm before just implementing it broadly across all their acres and making a substantial purchase. So, yes we definitely encourage that and I think that gives just such a huge variety of the topics. I can give kind of a range of different studies that have gone on. So, on the soybean side a lot of those have been on planting populations different maturity groups with the idea that if we could plant earlier maturity groups and obtain similar yields we could get in with cover crops earlier things like planting date on soybeans fungicide and insecticide applications, starter fertilizer on the corn side. A lot of our work is on different aspects of nitrogen management or fertility management, so technologies for in-season nitrogen management, different starter fertilizer mixes, lime applications. We've had some really successful long-term- I believe our longest was about a 15-year long term study through the network, which kind of a side-track I guess I'm bringing up with this. But, I think that's a huge benefit of doing this on-farm research that's really driven by farmers, if you're trying to obtain grant funding for a 15-year study, I think it would be very very difficult you know that's not really the time frame that we're generally looking at to be able to keep something going and so by the farmer really leading this effort and being the one that kind of maintains that study over time we're able to get this 15-year data set on the impact of lime on corn and soybeans. Okay, so back to the studies in general- also a lot on cover crops just how to get those implemented we had a handful on interseeding cover crops at V4 in irrigated sites this year just on rye cover crop versus snow cover crop, different mixes trying to look at beyond yield on those as well looking at some of the soil health parameters, infiltration, wet aggregate stabilities and things like that and then on the dry bean side we've had some on planting populations as well as inoculants. So, the planting population and varieties are really important in those studies because they're looking at the opportunity to direct harvest the beans instead of doing a two-pass system and so that planting population is going to determine the height that the pods are setting and how well they'll be able to do direct harvest, so those have been some interesting studies as well and great to see a group of producers out in the panhandle participating in this way.
Sam: And what kind of data are you like really trying to collect across all these sites, like what questions are you answering if that makes sense?
Laura: Yeah, that's a great question. So, of course it's going to vary based on what the study is looking at but in all cases we're taking it to yield. We want to have it you know have a bottom line for the producers. What's the yield impact and then one step further what's the economic impact, so not just looking at if we're seeing a yield response but seeing if that response is profitable based on whatever you know additional costs may be incurred to get that yield, so that's always a big emphasis making sure we have that economic analysis and then beyond that it just depends you know a lot of times we do stand counts and that's just so that we can predominantly just so that we can verify if the treatment's having some other impact that would then impact yield. So, if we're reducing stands in some way and then outside of that it really just depends on the study, so a lot of times that would be soils, soil sampling, sometimes trying to monitor that over time or sometimes just getting a baseline so we know where that field is at in relation to what we're seeing for responses. A good example there would be starter fertilizers, so having a phosphorous soil sample to know what the phosphorus level in the field is and then being able to see you know at what soil test, P levels we're seeing a response to starter fertilizer when we aggregate all the data. So, a variety. I don't know if that answers the question well, but it can go much more in depth particularly on some of the like cover crop studies where we're trying to look at soil health parameters and things like that too.
Sam: Absolutely, and just to add on to that one thing you know from the on-farm research meetings that we've been able to go to it's so cool to hear the farmers talk about like oh this was really easy to use or I really liked how that worked or you know you just hear kind of their personal what they thought of it as well, which I thought was a really cool value of those meetings that you have.
Laura: Yeah, I really like that about the meetings. I think it's a really unique part of the program just that it's so the information that's presented is very farmer presented or farmer driven rather than just standing up there and kind of giving a lecture or something about it. I think it's more interesting to hear about- I love hearing the personal stories I think it's really well received by others as well. It's great to hear from a peer or someone who's actually you know implementing it and kind of in your same situation or in your shoes and see what their experience was. So yeah that's one of my favorite aspects for sure.
Jack: Yeah, those meetings are always fun. You know, I know we had to do them virtually this year, but hopefully they'll be in in person before too long. You've kind of mentioned the farmer- driven aspect of research a few times now and obviously there are a lot of farmers out there in Nebraska that are doing this through the on-farm research network but there are also probably a bunch more farmers that are just running research trials out there on their own farms that are not you know supported through the on-farm research network. What is some advice that you have for those farmers who are laying out these studies on their fields as far as how they should lay out their treatments how many replications they should have you know how many different products or you know range of products they should have in the field can you just give us kind of those basic parameters that a farmer should kind of follow for good research?
Laura: Yeah, sure that's a great question, so our guideline when we're working with producers is really to try to get four replications of whatever they're doing for example, if we're doing with and without starter fertilizer for kind of an easy example here, we'd do with and without and we'd do that four different times so we'd have eight potential strips in the field there and that allows us to have a lot more confidence in the results, so we know that there's a lot of variability in fields if you're going to split a field half and half there's a lot of variability even if you don't impose any treatments or do anything different and just look at the yield from a uniformly managed field on half one half versus the other half you're probably going to have some differences and they're probably going to be in the range of a few bushels which is you know what you're trying to detect if you're imposing a treatment or trying something out, so having that replication is really important generally we'll take down to three replications but we would like to start at four as a minimum because sometimes things happen, you know there's ways things can go wrong, so that gives us a little bit of leeway and a little room there. So, that would be kind of our minimum and we have a lot of producers that honestly if it gets set up well then it's fairly straightforward maybe something with a split planter or something like that once they get going they just keep going and we had a couple producers that often have 20 or more replications in a field because once they get testing it, they're set up to do it and they might as well you know put it in different fields just keep those tests going. So, that kind of answers I think the question about replication. The other question then was about I believe how many different things you could test at a time so that that can kind of vary. I would say generally we're looking at not more than four different things and that might be like four different seeding rates, four different nitrogen rates it could be different for different starter products or something like that. Once you're getting beyond that you're starting to take up a lot of real estate in the field. Once you get those four different treatments times your four replications it starts taking up some space. Also, I guess kind of thinking about that the treatment selection I guess one thing that we always encourage is to try to separate the treatments as much as possible. So, it's going to be more informative for the farmer at the end of the day, if they know what say a check with no starter versus a 10-340 is instead of having maybe a check versus 10-340 and two other additives in there and then at the end of the day maybe we have a yield response for that that product or that that treatment that has the products. But, we don't know if it's because of you know which of those products that was in the mix. So, we try to encourage people to separate things out as much as possible or as much as make sense and try to understand where the responses are coming from, so they can really detect which products or which practices are going to be the ones that they want to continue with or adopt.
Sam: Yeah absolutely, and you know you've mentioned having a control is really important and I think that's kind of a challenge with on-farm research and so can you maybe talk about you know why is it so important to have that check? I know that's difficult in some things like herbicides where you don't want to have a part that's untreated or why it's so important to have like that range. We work in nitrogen, farmers don't want to have a really low nitrogen rate in their field can you maybe talk about the value and why it's important to have that anyway?
Laura: Yeah for sure, yeah so in some cases you know it really varies I guess based on if we're looking at a lower nitrogen rate or slower seeding rate versus like a no product type of thing and sometimes you know we aren't able to do a no product or no treatment check and that can be informative too. I would say things like nitrogen, which you guys are well aware of how we set these up. But, in a lot of cases with the opportunity to use variable rate technology we can get those checks in without taking up too much space. So, we might be able to do some lower nitrogen rates maybe we have some spaces in the field that just have 50 pounds but because we're setting those up in a prescription, we're not constrained to have to do a full fielding strip of this really low nitrogen rate. We can just set up a small area in that prescription and we're looking at maybe an acre or so that gets impacted in that way. So, it really can be a fairly minimal amount of the field that gets this less than ideal treatment that it's receiving but can be really informative. So, with something like nitrogen that can really help us understand what the optimum nitrogen rate in the field is. We have to get some of those lower rates to be able to understand how that field is responding. And things like soybean seeding rates, you know those lower populations help us know how low we can go and in a lot of cases I will say I think that the check is not as much of a disadvantage as people are anticipating. So, for an example this year we had a study that the producer had three different nitrogen rates 100, 150 and 200 and they were really nervous about going to that 100 pound rate from their standard which was more around 150 or 200. At the end of the day the lowest economic response was for the 200 not the 100. So, that 100 pound nitrogen treatment actually was close to the 150 and one that you know he could have been concerned about from an economic standpoint was actually the 200 and not the 100. So, you know when we start looking at it economically that also kind of helps us see the that the impact may be not what we had anticipated by implementing a check. But no, I think we're just really able to get some good information whenever we can get those checks in there. I guess back to the starter example that really allows us to you know a lot of times people maybe want to use their check as the 10340 and then try some additional products well that may give them some really good information, but it also would be great to know if they're even getting benefit from the 10340. So, you know it all depends on the question. But, I think there's a lot of benefit when we can add that that control.
Jack: Sure yeah, and I think another important thing that you brought up there is looking at things kind of from that net return aspect or the profit aspect, so everybody kind of knows about randomized complete block designs. I think those have you know long time been the standard in ag and still are and for good reason. But, are you aware of any like other I guess innovative experimental designs that are helping to get more information out of you know to answer complex questions that have been enabled maybe by precision ag technologies?
Laura: I would say the one that I've seen most used is where we use more smaller blocks within the field that are implemented with our variable rate technology, and so those may actually be considered randomized complete block designs depending on how you set those up they may still be randomized complete block designs but and the other opportunity though with that type of technology is to place those kind of blocks strategically in different zones so kind of a similar approach but maybe being a little bit more strategic about how we're placing those using what we know about the underlying field variability and you know the soils, the previous yield history and trying to place those in strategically in different zones. Again, though could still be that kind of typical randomized complete design complete block design. I think the other example of a unique setup is when you're working on Jackson with the sectors and that's I guess just kind of a unique take on the layout to make it work with a center pivot irrigation system where you're looking at your experimental plots being those kind of pie-shaped wedges in the field. So, I think that's a really neat approach that really is taking advantage of the technology that's there.
sure yeah it's it's uh it's really neat but it's also really challenging with those giant sectors out there in the field you you do capture a lot of that variability and i think what you were just talking about there being able to target where we know that variability is in the field and really go out and test these different practices in those areas you know even though it is still a randomized complete block design it's it seems like to me so much it seems like a better way to do that randomized complete block because you're already accounting for you know some of that variability that's out there in the field so yeah and then at the end of the day we're able to kind of look at responses by those different zones so like on our soybean seeding rate studies the goal there really is to say okay is the optimal seeding rate in this lower heavier textured soil different than on this sandy hill you know and likely the answer is yes, so being able to differentiate some of those responses is really powerful.
Sam: Yeah, and thinking about unique designs um you know in southeast Nebraska there's a lot of terraces and you know if you find the flat part of your field to run a study it's not representative of the rest of your farm are there any tips to get around some of that?
Laura: Yeah so, I've had lots of opportunity to play with some of those since our farm is primarily in contoured hilly terraced fields so yeah this is another opportunity where ag technology has really enabled us to do studies in areas that it just wouldn't have been an option before. I remember when I first started visiting with some farmers from the southeast part of the state with similar areas that they farm and they said they would love to participate but they just didn't know how it would ever work, and so I'm really glad that we have some opportunities to make that work now. So, some of the studies that I've set up in those type of areas have been where we're just primarily nitrogen rates looking at multiple nitrogen rates and we're able to lay those different rate blocks in to follow the curve follow that terrace keep them in a similar landscape position, so maybe on the back side of the terrace or whichever side and allow that to follow the curvature that all the equipment paths will be following. So, it takes a little bit of work to get it set up, it's not one of those where it's a really cookie cutter where a tool that's been developed is going to work really easily for you it takes a little bit of manual work and using that previous data to get that layout right. But, I've seen it work really well and you know I completely agree Sam, there's just not going to be that representative of data if most of your field is in this terraced, hilly environment and you're only placing studies in the one like flat spot that you have. So, so yeah I'm really glad that we're able to do that and I've been doing that for a few years here and hopefully we get some more farmers interested in doing that as well.
Jack: Yeah, I think that's really important and you also brought up another you know I think critical aspect in this and that's tools to set these experiments up because it's hard to set them up and while there may not be tools necessarily for terraced operations. What tools are out there for you know more standard operations with flatter fields or you know people aren't working with terraces?
Laura: Yeah, so a lot of times we're using some more robust or type open source type of softwares to set up the trials, but recently there have been a lot of technologies or a lot of software that's kind of catering more towards this. That's for farmer-oriented type of software so some of the tools that we use like ag leader SMS I believe Climate Field View all these tools are starting to or all these softwares are starting to include tools that allow producers to kind of set up trials within their field and try to help get those placed. So, I don't use them extensively because I'm using kind of an open source type of setup but some of the features in those look really nice, trying to help get those placed using your guidance lines that you already have to get the direction right trying to help get them placed in representative areas of the field by using what we know like the soil series maps to try to keep those blocks within a soil series instead of kind of crossing over where that line is, so using what's available to try to do a better job of getting them placed and some of those are you know a lot more user friendly for farmers to be able to take advantage of that and start placing some trials
Sam: And beyond just the software there's a lot of other challenges well sometimes like the expense of the equipment or the new technology. How does the on-farm research network overcome that expense challenge and or maybe some other opportunities to overcome that?
Laura: Sure yeah, that's a good question so sometimes what you're wanting to test would require a very substantial investment, so some of the projects that we're working on like the one you work with predominantly, Sam, with project sense we actually have the technology the crop canopy sensors and the sprayer within our research team to be able to go out and implement that and let the producers see how that's going to work, and I think that's a really great opportunity to get to try that a different approach that we've been using. In a newer precision nitrogen management project that started this past year is providing some compensation that allows producers to get reimbursed essentially for some of those technology costs they may occur to try something, so whether that's you know the technology itself which in some of those cases is something like a subscription to a service or hiring some custom work to do some variable rate application. We're trying to be able to help compensate them such that they could try it out and give it a try without having that additional cost, so that's another good opportunity or method. Some others that I can think of though in our network you know sometimes producers have made the investment and they still want to know you know they went ahead and pulled the plug and made the investment and they're probably going to keep using it, but they still want to know how it's working for them and how it's going to pay off. You know, they're curious people and we all get to benefit from that then because they've set up that study. They've already made the investment, so that's a great opportunity. A couple that I can think of that we had this past year we're looking at active hydraulic downforce systems, so the two producers that were using that they already made the investment, but they were willing to try it and wanted to try it because they wanted to know and so now other producers who are interested in that technology can look at their results and that's a great way to learn and see how that might work or visit with them you know and see what they observed. So, we get to benefit from their investment there.
Jackson: Yeah, it's kind of a nice piggybacking way of doing that right there. Yeah, so we've kind of talked about setting up a trial and what goes in you know maybe what expenses are associated with doing these on farm research trials, what are some of the most common mistakes that people make that you've seen when setting up these research trials?
Laura: That's a good question. Yeah, there's always a handful every year that that don't quite make it to our publication for various reasons. One challenge I would say that happens probably fairly frequently is not treating the area uniformly in other respects so maybe you were trying to look at soybean seeding rate but you also use two different varieties in the field or something like that or we have a variable rate corn planting prescription with some really big variations and then we're not sure how the product that you're testing is responding within those different seeding rates, so just kind of that extra variability that sometimes gets introduced sometimes you know accidentally if you forgot and just did a split planter of hybrids or something like that. So, I'd say that's one of the common ways that it can go wrong. Other errors, I'd say a lot of the other errors that cause us to lose a study come at harvest and just unfortunately what happens is that the producer doesn't combine solely within the treatment strip. So, they might be getting header widths that are a mix of two treatments and this happens more often in soybeans where we've got a header widths that are generally not matching up with planter widths and things just get off or things aren't communicated completely well between whoever was planting and who's harvesting and so that's another way where things get can get lost or kind of messed up. You know, that can happen I think it happened with the corn study this year too where it was at night and it was dark and they just, it just happens you know. So, we do the best we can to avoid that and in some cases you know with some really detailed analysis and really careful looking at the data points we are able to recover some of that data, so that's not a total loss for the producer who's invested their time in it.
Sam: When you do that many studies though as you guys do that's bound to happen sometimes. I think you hit on this already earlier but what in-season checks or data is helpful to collect you know just beyond the yield. So, I know you have like a team there's a team of extension people that help with some of this. What kind of checks are they doing?
Laura: Yeah, so one of them that I mentioned previously was stand counts and that's just kind of a good basic piece of information to have on the studies, just allows us to know if whatever we're testing has also impacted the stand and if that's potentially why we're seeing the yield difference at the end of the year. So, that's one that we often do and then soil samples. I also mentioned just getting some idea of the baseline fertility on a field at a minimum can really be helpful. We've been over the past couple years able to use aerial imagery also on the studies and that's been really beneficial in a variety of ways. In one kind of the big picture way it just allows us to see if there's things going on in the field that we maybe weren't aware of you know maybe there was a wet spot that kind of got drowned out and we really shouldn't consider that when we're doing our yield analysis and then in more detail kind of sense with the research study we're sometimes able to use that imagery to look at the differences in the treatment so maybe the treatment is causing a greater plant, bigger greater biomass, more greenness to the plants or different you know differences in the plant. We've been using these also in our cover crop studies, seeing if a preceding cover crop maybe you know if that cover crop is taking up more moisture in the soil that could impact say the soybean crop that's following it and we can see those differences in the aerial imagery there as well. Maybe the one crop is drier or maybe it's delayed maybe it emerged slower so being able to pick up some of those differences in the imagery I think is really powerful. It's also great that it's a lot denser sample than if you're out there walking through the field you know, you're only going to take so many measurements or be able to make so many observations on foot in a canopy corn field or a you know a tangled soybean field you know you're just not going to cover it that well and so we're able to get really good coverage using the aerial imagery in that way.
Sam: Yeah, so now moving more towards the end of the season and looking at those results. O think there's still probably lots of farmers who get out there and they get their printed yield map at the end and if they see the difference then it worked and if they don't see the difference it didn't work but how do we look at the data analysis procedures and like can you walk us through that to know that our results are accurate and that there's a difference in treatments?
Laura: Sure, so our first step is getting the data from the farmer and not in that kind of printed or PDF version where we're trying to get that raw yield file from the producer. We then go through a cleaning process with the data that's sometimes done with the USDA yield editor tool or it can be done more manually and the goal there is to remove erroneous data points, so there's places in the field where maybe a full swath width wasn't harvested maybe the operator had to speed up or slow down rapidly for some reason in the field and we have erroneous data points there you know there's a variety of ways that that our data can become unrepresentative of what was really going on in the field. So, our first step is to just try to do some pre-process or post-processing and cleaning that yield data after that we're then looking at I'll just say for kind of a general study we're looking at summarizing the yield data points within those different treatments and so we do that using different GIS tools and it could also be done in farmer-oriented data management software, but we're just looking at what the yields were in each of those strips. So, not at this point not overall the strips but in each strip individually or each treatment area individually. On more complicated studies, we might look at those within specific zones then as well based on maybe productivity zones or soil type zones or something like that at that point we would then have the yields for each of these different treatments say a treatment or and a check or a control and we then put that into statistical analysis software in order to look at what our variability is across that study and how much confidence we can have that those differences in yields that we're seeing across the strips are really due to that treatment and not the underlying variability that's going on in the field. So, we might end up with you know eight yield data points. I'll go back to my original example, I think that was starter fertilizer versus a control with no starter and we have four replications of it. So, we've got eight strips say we're just summarizing those yield data points within each of those eight strips so we've got eight values and then we would also have a total you know yield for the check versus the starter but we want to know if there's that two bushel increase. For starters, was that occurring on all four of those replications or you know was that a big swing you know in some of those replications we saw a 10 bushel increase and in some ways an 8 bushel decrease you know where's that coming from so that's really what the purpose of that statistical analysis is trying to tell us how much confidence to have in that yield difference, if it's really due to the treatment that we've imposed or if we need to say well there's a lot of variability going on maybe we don't have a lot of confidence that this yield difference that we're seeing is really due to the starter fertilizer or whatever treatment.
Jackson: Sure, so what digital tools are kind of available out there to help people perform that statistical analysis is this something that's kind of built into like farm management software in some cases, or I know we use ARC, but I don't know how many people are going to go out there and you know ?
Laura: Yeah, I mean the software we're using is primarily Open Source that people could use but it's definitely not on the user-friendly or it requires kind of a different understanding or some coding skills. So yeah, one of our goals in Jackson, I feel like you could talk about this here for us but one of our goals is to provide a tool that would allow producers to do this or agronomists or whoever's working on these trials because I think a lot of producers are really you know they understand this concept they understand that there's variability and they understand what we're trying to do with the stats but there's not really been a very convenient user-friendly tool that's just ready to go for them to have this analysis for themselves. So Jackson, I don't know if you want to like share a little bit about what we've been developing.
Jackson: Yeah, yeah I mean so FarmState is a program that was just released in late February by the Nebraska On-Farm Research network that basically allows growers to input data from studies that they've conducted that's you know been formatted in Excel spreadsheets or you know they've just collected even on you know written down on a piece of paper however they have that data but enter it into the program and basically run what's called an analysis of variants which at the end of the day is just like Laura was talking about. Basically a procedure that is able to identify whether the differences that you're seeing in the study are coming from the treatments or if they're from random error out there in the field and at the end of the day it puts out a PDF report that basically says okay these treatments were statistically significantly different from these other treatments in a very you know simple way that doesn't necessarily require you to be able to read you know statistical output. It just says okay these are these are real differences that we saw in this study, this is how you can kind of proceed with your operation, so that's the goal. Laura: That's something I really like about the way that this is being developed is that all that statistical output is there if that's meaningful to you or you want to read it you have all that detail but also there's just kind of a bottom line sentence that says treatment A was not statistically different than treatment B you know or whatever, and so I think that's really helpful to make it useful for a broad range of people. So, yeah I'm really excited that we're going to be able to have this tool for people and it's just going to be a web tool so it's really easy to access and Jackson's really been leading the effort on getting that that accomplished and created so really appreciate that.
Jackson: Of course, I'm just hoping it brings value at the end of the day like that's I think that's what we all want out of it.
Sam: So, yeah yeah and thinking about sharing those results and making them easy to understand you know you talked about this long process but then a huge part that you play at the end Laura is getting all these results summarized and put into some resources can you talk about those resources that people can access?
Laura: Sure, so after we have all of the data analyzed and the stats done another piece is that we collect you know just some good background information on each site, what hybrid was planted, when was it planted, when was it harvested you know just some of these basic questions and we compile all this data into a report for each site, each study, each of those is then put into our annual report book and that book is then published as a PDF on our website it's also printed as hard copies and made available at our meetings that we visited about earlier. And then another way that we make it available, that I'm really excited that we have this this tool going now is our online results finder database, so this is a place where each study lives as its own PDF and it's searchable or filterable so you can search by keyword or maybe if you're interested in a specific product in a specific area of the state you can filter it down and you know say I want to see in the southeast portion of the state what's been done with this nitrogen stabilizer product and some questions like that and then you're going to get a list of all the studies that have been looking at that and be able to see each of those PDFs each of those reports, so hopefully a better way for people to really access this big repository of data that we've collected and that actually includes data clear back from, we've gone back and digitized some of that clear back to the 1989-1990 results so it has I think close to a thousand research summaries in there and hopefully a much more accessible way than like coming through you know tons of books or PDFs or something that's a giant repository.
Jackson: That’s a giant repository. Yeah, so if you have any last piece of advice for any anybody who's looking to experiment on their own farms whether through the on-farm research network or out on their own would you mind providing that to us?
Laura: Yeah sure, well my advice would be to just try something you know if you have a question about something it's a great way to try to start getting some answers and my second piece of advice would be if you want help with it ,if you want assistance to reach out because that's what we’re here for. We want to help make sure that at the end of the day you're getting reliable results that you can have confidence in and that are going to be informative, help you make more productive or more profitable decisions going forward. So, that's what our goal is to help you get that set up in a way that is going to be the most beneficial for you.
Jackson: It was great having Laura back on the FarmBits podcast to talk about a topic that is important to her and that she has a lot of expertise in and we really appreciate her taking the time to join us today. Now one thing that Laura you know really talked about in the interview even though she gave us a lot of great information was kind of a way of modifying the randomized complete block design that you find in a lot of agricultural experimental trials to account for known spatial variability in the field right, so if you can break a field into management zones and treat those homogenously and you really feel like those are kind of an accurate measurement of the spatial variability in your field it doesn't just make sense to kind of run trials within those different zones and at least see what the different zone responses are instead of just setting up a trial that runs across those zones and may be affected by that spatial variability, and I just think that's a really cool step that we're taking it seems like something simple but I think it really is kind of an innovative way to make use of these precision ag prescriptions to put trials where they need to be in an actual field.
Sam: Yeah, and it's great how you and Laura are working on that FarmStat tool or getting that published soon to make all this a lot easier for farmers to run those designs and have accurate results, and I think my favorite part you know we work with the on-farm research network in our research and we've known a lot about it and we know that value so we can take it for granted and so you know hearing all of this back it was really cool to hear. But, I actually really liked when you talked about the history like how long it's been going in Nebraska and all the value that it's brought to certain producers as we hear them talk about their stories that was fun for me because we see more of the technical side all the time, and so great episode to talk to Laura about all of that so thank you for joining us on the episode of FarmBits and we hope you'll join us again next week you won't want to miss it. Thank you for taking the time to join us today on the
FarmBits podcast, if you enjoyed this episode please subscribe to the podcast on Spotify, Apple Podcast, YouTube or wherever you listen to podcasts to be informed about the latest content each week.
Jackson: We welcome your feedback, so if you have comments or questions for us please reach out to us over email, on Twitter or in the reviews section of your favorite podcast platform our contact information can be found in the show notes.
Sam: We'd like to thank Nebraska Extension for their support of this podcast and their commitment to providing high quality informational material to members of the agricultural community in Nebraska and beyond.
Jackson: The opinions expressed by the hosts and guests on this podcast are solely their own and do not reflect reviews of Nebraska Extension or the University of Nebraska-Lincoln.
Sam: We look forward to joining us next week for another episode of FarmBits.
Transcripts are generated using a combination of speech recognition software and human transcribers, and may contain errors. Please check the audio before quoting in print and write email@example.com to report any errors.