Water vapor fluxes are extremely useful to potato growers looking to irrigate their crop efficiently. This episode brings together Beau, a farm manager and Ammara, an evapotranspiration flux modeler, in a discussion of how flux science is used by growers. From soil moisture measurements, to camera-based analysis of pest infestation, we cover the pros and cons of different technologies used in agriculture.
.
In this episode of Meet the Fluxers, we learn about the intersection of agricultural irrigation practices and scientific research through a conversation between Ammara and Beau. The discussion reveals how farmers like Beau rely on a combination of moisture probes, weather forecasts, and personal field observations to make irrigation decisions, while scientists like Ammara are developing models like the Wisconsin Irrigation Scheduling Program (WISP) by integrating actual field evaporation measurements from flux towers. The episode highlights challenges in current irrigation models, including their tendency to overestimate water needs and their inability to account for practical farming constraints like irrigation system rotation times. It also explores the growing role of technology in farming (including AI, remote sensing, and variable rate applications), the importance of effective communication between scientists and farmers, and how climate change is creating new challenges that require adaptation in both farming practices and scientific models. A key takeaway is that collaboration between researchers and farmers is essential for developing practical, user-friendly tools that balance productivity with sustainability.
Transcript
Jess: Ammara and Beau, both of you have an interest in flux data and your lives are influenced by it, but you both have completely different professional backgrounds. Beau, what is your experience using irrigation data for models or online tools?
Beau: A lot of the online tools that I currently use are more moisture-probe based. I subscribe to the UW Ag weather, so every day I get an email that is generated saying what estimated ET is, predicted rainfall, what the previous day’s weather was, with the weather forecast. I have done some work with WISP. Purdue maybe had one a while ago. I haven't looked at it that recently, but most of my work with WISP was back in 2014 or 2015, kind of right after I got my job. I was just looking and noticed that it's been updated along the way since then. Right now, we're mainly going down the moisture probe way of looking at soil moisture.
Jess: Were there any trends that surprised you while working with WISP?
Ammara: Yeah, absolutely. So, yeah, WISP is the Wisconsin Irrigation Scheduling Program model that was made by a soil scientist in a Wisconsin university. The WISP model is based on remote sensing data. Remote sensing data can have a coarse resolution. So even before comparing it based on rationale and literature review, I thought that there might be some mismatch. But I didn't know that…if WISP is telling us that farmers should irrigate crops using more water or less water. So, I didn't know in which direction the results would go. So, what we did…we used observation data to see how accurate the models are. We set a flux tower on a farm, and we had data for five years from that flux tower. I saw that WISP, for example, it's saying that farmers should irrigate water more because the magnitude of evapotranspiration that I was getting from WISP was higher than the observations, which means that then farmers probably would be over-irrigating water. I was surprised because of that. That was the reason that we decided, let's look at each component of the model.
If a model is producing some results that are different from observations, there could be two reasons. There might be some inaccuracy or error in the input data or there might be some inaccuracy in the assumption of the model. The model is based on a bunch of equations, right? So, it's possible that one equation might not be capturing the trend of water fluxes properly. Those were our two hypotheses. And then we decided to test those.
Jess: Beau, you mentioned that one of your issues on the farm is that sometimes the irrigation system leaks.
Beau: Yeah. I guess that's…that's one thing that's hard. And, I would probably say it's more of how long it takes an irrigation system to get around. A lot of times, what we found kind of our sweet spot for, for irrigation for potatoes is about four tenths per revolution. In some fields, you can do that in 12 hours. Some fields, it takes two days, literally forty-eight hours, to make a revolution. And so, it's hard if you're trying to go off a model to tell you, “Hey maybe you don't have to turn on irrigation today! Turn it on tomorrow.” Well, it takes two days to get around. So, if it's withdrawing half an inch every day or every two days, you've got to keep it going non-stop. The other thing is, what if it breaks down? Then you get behind and you can't get caught up, particularly in years like last year [when it is] so dry in Wisconsin. That's something that I just don't think a lot of the online tools are accounting for, as of yet.
Jess: Yeah, those are both things that it's difficult for models to reflect because of unpredictable things. Your eyes are more reliable as sensors sometimes, and your personal experience, than a faulty sensor.
Beau: Yeah. For sure. The other thing with probes is, a lot of times, they can be cost prohibitive and if you've got one probe and… yeah, okay, you're supposed to put it in an ideal spot in the field, but man, there's a lot of fields where there is no ideal spot. How do you go off just one probe in 160 acres to tell you what to do? There is a lot of that. You have to go out to the field. You have to see it with your own eyes. You have to dig in the root zone, see what you’re…what you're finding. I do have a lot of hope I guess, and I think there are a lot of positive things that will come out of, just like Ammara said, whenever they went out and kind of proof the data. That's awesome. That's hopefully helping to rethink the formulas and everything that comes up with stuff. At the end of the day, going out to the field is the best thing you can do. But I think there's a lot of good stuff coming down the pipeline to help better manage online.
Ammara: Like work in progress, yeah.
Beau: Yeah.
Jess: Do the probes usually reach as far down as the crops? Is that difficulty?
Beau: Yeah, so we've tried different probes over the years. We've actually had some AquaSpy probes and even I think Sentek probes that were about four feet long that would measure EC (electrical conductivity) data every four foot. Right now, we're kind of focused on two probes. One is a CropX probe. I’m wanting to say they're about two feet. The other one is called Soiltech, which is made to kind of be buried in the hill of potatoes. That's kind of where they focus: potatoes. That, you're just getting maybe six to eight inches down. So that's a very good point, you're not getting a lot deeper than that.
Jess: What were your opportunities to communicate with scientists then? And what are they like now?
Beau: From my point of view, I haven't worked with them as much as I would want to in the future. I would definitely be open to working with them or any other flux scientist. One of our farms is about an hour away from Madison, from UW. Reach out to us, we would be more than willing to work with you.
Jess: Is it often that you hear from extension researchers?
Beau: Yeah. Particularly in the Horticulture Department. We do have different trials working with them. UW has such an awesome extension network. The Hort. Department, they do a really good job of on-farm trials, trying to get real-world experience, trialing different stuff. So, I can't say enough good things about them.
Ammara: Growers also have an association which is called Wisconsin Potato and Vegetable Growers Association, so they have their annual and quarterly meetings too, I believe. I used to meet with them once every four months, and then they have an annual conference that happens usually the first week of February. And then, there is a show where the growers come and they have different products. I think that is one way by which scientists and growers can communicate with each other. But yeah, as Beau said, there are a lot of professors who are doing amazing things at UW-Madison and it's not just the water side. People are also working on organic matter, carbon cycle, pesticides, and GMOs. Stuff like that. So, I think it would be good if we can have a more diverse annual conference in Madison or somewhere near Hartland farms, I think in the future. So that could be one way. I know WPVGA have some people in their mailing lists and they send emails, but maybe they could make effort to include more people, new growers, or new scientists.
Jess: Ammara, when you were using flux data to build upon WISP, with how often did you speak with people who were working around the tower, or did you communicate with them about the progress on the model?
Ammara: For each growing season I would make plots and then compare each growing season with the previous one, and how this growing season is different. I would show them, is precipitation higher in 2021 versus 2020? And then, I would show them temperature, precipitation, evapotranspiration. I would compare evapotranspiration from forests versus crops. So yeah, once every four months, and then I would present once a year.
Jess: Did you get a lot of feedback when you went to the annual meetings?
Ammara: Yeah. I think the first time I showed them that error, that the model is telling us to use more water, actually we don't need to use more water. So, then a grower asked, “I know, I get an email. Let's say I get an email that ET (evapotranspiration) was two millimeters. So, how much should I reduce?” I like those questions. They were very direct, like, more applicable. “You know that we are over-irrigating. How much should I reduce?” And of course, it will vary between if you have a dry year versus a wet year. They had questions like that. So, I think WISP needs to be integrated in a bigger model, in a real-time model. Then hopefully it can solve that problem that Beau was talking about that it will take two days to take that… to make that round. You cannot just stop something.
Beau: I remember you guys presenting on what, like, a pine plantation, the ET on that, as well. As a grower that was kind of interesting, just to see, “Okay, maybe potatoes aren't as bad as they're let out to be.” Granted, pine is totally different, but some of that work that you guys presented, I thought was really cool too.
Ammara: That was one of the motivations behind this work. Growers were interested in, and DNR (Wisconsin Department of Natural Resources) was interested in seeing that, how forest is different than agricultural land. I think in Central Sands Wisconsin, some growers are interested in expanding their agricultural area to the south of the Central Sands, but we have a lot of forests there. So, their hypothesis is that plants are also evaporating a lot of water. So, if we have to clear, if we can clear that land, we will be doing a favor. But then DNR is a government institute. They need to have data or evidence before they go ahead and cut the crops. So, our project was partially funded by DNR too. They wanted to see that comparison between crops and forests. Even though pines also evaporate a lot of water, if it's a dry year, probably the magnitude of water that is used by crops will become much higher compared to a normal year. But for pines, it might not be that different.
After my presentations, usually, I got emails from the scientists too that, “Oh, you need to be careful what you are saying! You have to explain properly.” So, I know I made a lot of mistakes when I started working. I learned a lot of things. I wasn't supposed to be using SI units (The International System of Units) because growers are more comfortable with units of inches. So, my first presentation was a bit of a disaster, I think. I used a very controversial line that one professor from the engineering school, he said, “Ammara! You cannot use lines like that. You are in a land of plenty water. I mean, it's already a contested area.” Yes, it was a lot of learning experience for me too.
But the growers, I was really impressed with how knowledgeable they are. They come to all these meetings and all these conferences, and they sit and listen and then their questions are very insightful. They ask, “Okay, how much should I reduce?” Questions about clouds too. That, “Okay, WISP is… probably there is a mismatch between observation and model output. So, what are other reasons?” They would also give me some of the hypotheses too. [They] said, “Oh, do you think it could be because of that? It could be because of this.” Talking with them was really helpful.
Note: Forests use more water than grasslands and short rotation crops due to larger above-ground biomass and extensive root systems. Clearing is also a restoration technique for tallgrass prairie, an extremely rare but historically common ecosystem in Wisconsin. It can be done through burning, harvesting crops if the area is former farmland, herbicide application, etc. This is typically followed with seeding and maintenance.
Jess: What kinds of science are farmers using to increase productivity?
Beau: It's tough. I think it's on, kind of, on every front that farmers are looking for new science. A lot of new chemistries are kind of, hopefully coming down the pipeline. We tried a new RNA-based insecticide this year. If you guys are familiar with that, it's kind of the first out on the market. That didn't work exactly how we wanted it to, but we're trying to still figure out the best way to apply it, but that's one. A lot of the technology that I see is trying to connect all these different worlds. Someday it would be nice to have our Soiltech sensor talk with our irrigation technology so all of our pivots are remotely managed, I can turn them on and off on my phone, I can change the depth, how much water they're applying. I can have an idea of when they're going to shut off, I can shut them off whenever I want. But it doesn't do me a lot of good to have all these moisture probes if I've got to get somewhere else and look at all my moisture probe data and then go back over and decide whether to turn on that pivot or not.
A lot of the newer technologies are trying to connect some of that stuff. There has just been a mismatch with different programs not talking with each other. That's not necessarily science-based, I guess. But like Ammara said, just kind of improving models I see are helping us out quite a bit as well.
Ammara: I also came across a similar type of technology that growers are interested in. They are using GPS and GIS systems and drones and satellite imagery. They are interested in real-time data and we can see that it will be very helpful. Science has taught us that it's more helpful to use fertilizer and water at a variable rate because different areas might need different amounts of fertilizer. I have seen that growers have learned that, and now they use variable rate technology.
Like Beau said, they are also using GMOs. So, genetically modified organisms. You can use this biotechnology to come up with drought-resistant seeds and that can help you to not lose a lot of money if there is going to be a drought. So, CRISPR gene editing is very helpful. So, of course, it's not just hydrologists or soil scientists. We also need people from the genetics department too.
Growers have their own sensors. They go into the field. They're very good with observation. They have learned from the previous generations too, and they look at pH levels, soiI moisture, and then sometimes their own observations can be better than a model. Models can have errors or uncertainty. Then, of course, they are using the ET model, WISP, which is another type of technology.
There is also some research going on about pests, that natural predators are better, and they are also looking at sustainable nutrient management. So, we have people in the Agronomy Department. A lot of professors have done amazing work there. We not only need precision irrigation, we also need precision fertilization. It's not only just quantity, it's also about water quality. If we are not mindful of using our fertilizer, we are going to pollute our water or runoff. It's amazing how Wisconsin-Madison is very diverse. All these people are coming together and using different types of science to make farming more sustainable.
Beau: One of the big areas I think there's a lot of opportunities in, and it kind of goes with what Ammara is saying, is the use of AI in farming. For the past three or four years, we've had these special cameras that are installed on our irrigation system. So, whenever our pivot is walking, it's taking pictures as it walks around and it sends it up to the cloud. Then it sends us the report saying how many beetles we have, what stage, are they larvae, are they adults, what kind of grasses you have, how heavy your infestation of this pest or that pest is. And again, that's all AI too. So, I think that's kind of a promising area.
Jess: How much have pests influenced the productivity?
Beau: Horrible. If you guys have any experience with Colorado potato beetle, it is by far our number one enemy and just a perfect, perfect organism to build resistance. Since I've been here, I would say one of the worst yield-limiting factors that we have as a farm has been the potato beetle. And that specifically is what this RNA product is targeting. What I really like about it is it's designed specifically to shut off something in the Colorado potato beetle. It's not affecting anything else that's out there. I have high hopes for it. It's not perfect quite yet, but it's the start of a lot of stuff coming down the pipeline.
Jess: What happens if too much water is used on the farm? Does that affect productivity?
Beau: It can. It depends on the season, I would say. Early on, like, especially. We're mainly a potato farm, so that's what I'm going to talk about. Early on it's hard to overirrigate and cause any damage. Once you get to where we are now, where the crop is mature and starting to senesce, you can definitely overirrigate. It causes a lot of blemishes on the potato. It can help increase black dots, silver scurf. Lenticels, how potatoes breathe, can expand and become kind of unsightly. A lot of the problems don't cause anything wrong with the potato. It’s just people don't like it, they want a perfect looking potato whenever they're buying it. Later on, you can definitely mess up a crop of potato by overirrigating. Early on, not so much.
Ammara: In the discussion of one of my presentations, one grower mentioned that we might not lose the quality of crop by overirrigating, but we can lose a lot of money if we under-irrigate. Under-irrigation can be a lot of problems for them.
Beau: Yeah, that's right.
Maoya: We all kind of want to combine all the best technology together to make the best decisions. Flux towers are capturing the whole footprint of the field, the cropland. They're sort of our best technology to get the actual ET of that entire area. As we progress and perfect these models with AI, how much are we going to trust those estimates?
Beau: From my point of view, I would say something that's easy to use and kind of talks to the basic farmer, in farm language. I think I would trust it. It's just like anything else. It's a trust but verify. So yeah, it gives you some data, you go out and just kind of see what you think and how it relates. Just like the AI with the camera. We still send out scouts to each field every week, but it just kind of helps guide us. So, maybe the scout doesn't spend as much time at each field or maybe they target where they go in the field. In an ideal world, if you had field-specific flux data, you would use it to help guide you. You wouldn't use it as 100% truth, but maybe eventually you would.
Maoya: Would there be, like, opportunity for test cases and, sort of, collaboration?
Beau: Yeah. On our farm, anybody would be welcome to come out and do that.
Ammara: Growers are very welcoming. If the AmeriFlux team can have some training programs for both stakeholders and growers, I think it would be great. And I understand that AI has faced that criticism, that is data driven. We know that we don't have a lot of data for droughts or extreme events. Even when I was working on a model here in Cambridge, I noticed that my model performance was less accurate for extreme events. Yeah, there will be some limitations, but AmeriFlux has data for decades. So, if we can have all that data, we will have a couple of droughts in that one decade, then we can improve that bias in our AI models. Collecting data is good too because then we can assimilate that data with AI. We can improve our AI models. Right now, if they are not 100% accurate, maybe in the future, I think they will be close to at least 80% good.
Maoya: How do you feel about using big tech models versus academic models? And what could academics do better?
Beau: The challenge for both big tech or academic would be how you communicated to the farmer. I can understand it a little bit of scientific lingo, but not anywhere on the level that you guys can. My wife is a physician, and she talks in all these fancy words. But I know that each one of those words has some basic meaning or some basic word that says the same thing, but for some reason she talks in fancy words. If you can just talk and figure out how normal non-science people talk or how they would read something, I think that would be huge. I don't have a big, huge problem with big tech. Yeah, it would be nice if academia had perfect everything, but I understand it's a money issue. Big tech makes people pay for stuff, where academia usually gives it away. Yeah, I would definitely rather from UW have this nice Wisconsin-based prediction stuff. But, if big tech had it and it seemed reliable, I think we would go for that as well.
Ammara: Just translating these models into user-friendly tools can be very helpful. The growers are the ones who are going to be using those tools, so if those tools are not user-friendly, then I think they will be skeptical about adapting to those models or recommendations from those models. Making them user-friendly and easy to understand would be helpful
Jess: Are there challenges engaging with scientists or stakeholders?
Ammara: Sometimes I think we think that we have conflicting priorities. Scientists might think growers might have…that their focus is cost. The scientists think that they have long-term sustainability and accuracy goals, so that should be their priority. But we do need crops to feed the world too.
If we can just balance these diverse interests, that can build trust between different stakeholders. I was reading an article, I think where DNR and growers, they were contesting about something. And I think a person said, “I cannot put my boat outside the lake!” Lake levels were going down, especially in 2010, when there was that big drought. And then another person, during the meeting he said, “But we are feeding the world, do we have to irrigate our crops?”
Both goals are important. Now, we just have to balance these diverse interests and increase the collaboration between different groups.
Beau: I would add too, if you're trying to get more farmers to sign up, I would reach out to an organization. I'm lucky we're a part of the WPVGA. But if I was trying to get a potato farmer signed up or try to collaborate with on research, I would contact somebody at the WPVGA and they would say, “Hey, contact these three or four farms! Maybe don't contact these three or four other ones that might not be interested.” With most cropping systems, there's definitely state organizations that you could reach out to. Most of those people want research done or are very research-oriented just like WPVGA.
Ammara: And I think that training that we were talking about, it's not just only the training of growers. I think it's the training of new grad students too, like I was. Scientific writing can have a very different style, and we make presentations for scientific conferences in a certain way. But if you are going to a group that is very diverse, PIs (principal investigators) should be very clear with their students. Especially if the PI has been to those meetings that you have. “You have to give me a mock presentation. I want to make sure that what you are saying makes sense to other people too.” So that technical literacy can come. We have to talk about these complex models, but students can do a better job conveying their ideas. They shouldn't just target those meetings just like other academic meetings or conferences. Those meetings are different.
Jess: How do you manage water resources now, that's different from in the past?
Beau: One of the things that we started as a WPVGA initiative in 2013, I think, so now we have 11 years of data. We started doing well depth sampling, just seeing where the well-depth is. For instance, in some of the fields that we farm, the water table is five feet down. Some of them that we farm are 30 feet down. We have a list and every single year, we go to the same fields in the spring and in the fall just to determine, “Okay, did we have an impact?” Also, just to have a long-term basis of what we're using, if any.
It's always interesting because it hasn't changed in 11 years. We're in a different part of the state. We farm within three or four miles of Wisconsin River, which I think has a big effect on stuff, so maybe we're not seeing the whole picture. But, from my experience living in Wisconsin, it seems like it's more of a flooding issue than it is a drought issue or with lakes drying up, or with wells going dry. We're definitely blessed to have too much water more often than not.
Jess: Ammara, do you notice any difficulties with modeling and severe weather trends?
Ammara: The error between observation and model output becomes larger when it's an extreme event, whether it's a dry year or whether it's a wet year, flooding. So yeah, I have seen that. Because of climate change, probably growers are changing some of their strategies.
I was reading a report where, in some part of the country, now they use raised beds to improve their irrigation and there is more controlled water release mechanism to manage excess water. Models might not have all these changes that growers are doing from their side. So like, human impact. We also have a shift in growing season. Growing seasons are getting longer because our winters are getting warmer. We are seeing that here, especially in the East Coast. That will definitely have an implication on the model results. I am thinking again about that drought resistant seed, but of course, flooding is also a problem. When we have an extreme event that is not a mean trend, the error is higher.
Jess: Is that because of sensors and electronic equipment? Or is it the limited amount of data from those extreme events?
Ammara: I think it's both. Yeah. It's both. Sometimes plants also respond to change in climate. For example, if there's too much carbon dioxide in the atmosphere, and we know there is because of methane and greenhouse gasses. So what plants do is they start to close their stomata more, which means there will be less water that will be evaporated. That improves water use efficiency. Models might not be able to capture that. If we don't have that equation in our model, how is it going to capture the water use or change in water use efficiency, which can happen as the climate is changing?
Trevor Keenan from UC Berkeley, he published a paper where he said that there is a forest ecosystem that, their water-use efficiencies are increasing because they don't open their stomata as much because there is a lot of carbon dioxide outside. So, they just close their stomata, but if they close it then there will be less water that will be available. So, those kind of things, we haven't included those equations in a simple WISP model, which was based on precipitation data, which is more like atmospheric demand. It doesn't do the biological mechanism. It hasn’t been incorporated. We can have canopy cover, we can use that, but it's not as efficient. There has to be a balance. If we are going to use a really complicated model, it's going to take a huge effort to run that model and then calibrate it. “Do I have data to show me what my model is calculating that is the actual water use efficiency?”
Jess: What changes have you considered in any due to climate or weather stress? So, Beau would a certain amount of stressors on the potatoes cause the farm to move indoors?
Beau: No, I don't... Not anytime soon by any means. One of the big things, I don't know if you guys know of Dr. Eric Snodgrass is his name, he's a meteorologist out of Illinois who works for Nutrien. Very smart guy and just really does a really good job of kind of looking at ag weather and what ag farmers should be looking out for as far as weather. He gave a talk a couple years ago. He talked about how in some areas, the amount of rainfall isn't going to change, but the frequency of two-inch rains or higher is going to increase. This is a very good year. Since I've been here, since 2013, we've had probably twice as many two inch plus rains this year as we have any other year since I've been here. It's hard, how do you manage for that? One of the things I felt like we already have been doing a pretty good job of is just spacing out our nitrogen inputs. We don't want to pollute the groundwater. We don't want to lose money, but more than anything, we don't want our crop to die prematurely and not size up. I don't have a perfect answer, besides just, we need to split it up. Some varieties, maybe we need to put on a little bit more earlier, some varieties a little bit more later. It's just something that we have to take into consideration. It seems like it's hotter, it is hotter now. That one's a little bit tougher to grapple with. That one is…It may be better to look at different varieties that do better in Southern latitudes. We are trying to look at all of that. I don't have perfect answers for what we need to do. We just know we need to do something.
Ammara: Some change can come from the policy side, too. If they can give incentives to the growers to use more efficient irrigation systems that would be helpful. What Beau was saying, I have also seen the same trend in the rainfall data. I was looking at from 1965 to present that intense rains are more frequent now. The stochasticity or variability is increasing a lot, especially in the past one decade.
Jess: The very equations driving the changes in your models will have to change.
Ammara: Yeah, yeah. Exactly. So, it's like under, like, assumptions. The WISP model, one cause of the error was that the longwave radiation budget was incorrect. The way we corrected it, we used our observation data and we used an inversion method to come up with our own coefficient. The previous WISP model was using a coefficient that was trained on data from Arizona. Arizona is a dry area. We are dealing with a humid region. So, we shouldn't be using the same coefficient. We updated those coefficients. That underlying equation that we were using was not correct for a humid region. It might work for Arizona, but not Wisconsin.
Maoya: To what degree are farmers willing to be a bit experimental adapting to changes?
Beau: It certainly depends on the farm. Another thing too is to what degree. Okay, if minimum you're going to have 80 percent the yield or some type of guarantee, you would be willing to sacrifice x number of acres. If you could have a crop loss or 50% yield or something, maybe you want to do that on five acres. Farmers in general are open to trial and error. Farmers kind of get a bad rep sometimes like we don't care for the environment or don't care for groundwater quality. But that's not true.
Maoya: Would this require, sort of like, partnerships and sort of commitments with scientists to really monitor the ups and downs over a long period of time?
Beau: There's grants that are out there too, like, for different research trial programs that reimburse farmers x number of dollars if they take part of a trial and stuff doesn't work out very good. There's definitely opportunities out there. Something small scale a lot of farmers would be willing to try.
Ammara: If that funding can come from both government sources and some private companies to reimburse growers, that will be great.
Jess: How could flux science better meet the needs of farmers and how could flux science play a role in meeting changes in future water demand?
Ammara: Flux science will provide us those flux towers. Flux towers are important because I am using that data to refine my ET model. If I have a better, improved ET model then irrigation scheduling can be more precise. You can reduce water waste. We can use that flux data to calibrate and validate remote sensing products too. I can have a flux tower on one farm, a potato farm. I can collect data for five years. Now, we should also care about corn too, let’s say soya bean. We can move that tower for a couple of years to the other type of crop. Then we can use that data to correct a remote sensing product. A remote sensing product can cover a bigger region.
Sometimes policies are made on a regional scale, rather than just on a farm scale. But, we have to start from a farm scale, then upscale our model because of the heterogeneity. We can’t just start with the bigger model. We won't be able to recreate from where the error is coming. Flux science is very important, first to make a regional model, and then upscale it to a larger area. For climate change, of course, we need data for many years. AmeriFlux has data for decades. AmeriFlux science is very important in that regard. It can give us those cycles of water level going up, water level going down. That information can help us to predict future changes in water availability and demand. For example, let's say I have data from AmeriFlux in 2010. It was a really dry year. Now, if I also have irrigation data from growers, I can see, “Okay, in a really dry year the irrigation demand went that much higher.” I can just kind of, use that information to make a prediction. What will happen in a future drought or future wet year? How do they need to change the scheduling of their pivot?
Jess: Beau, would that be helpful if flux data or flux models could predict what's going to be a dry year or a wet year? How much advanced notice would you need to decide to plant a different crop?
Beau: It would be tough. You would definitely need almost a yearly, “next year is not going to be very good” to know which crop. Most farmers are buying seed in October the year before, November, December. By January, pretty much, you have everything bought, laid out, and planned. So, you definitely need to know fairly early.
Ammara: A year ahead notice, and then, if it's going to be a dry year, then you can grow a crop that is going to use less water. But yeah, you need to know a year ahead.
Jess: Could you use that information to decide to take more risk or take less risk?
Beau: Yeah, not us, but smaller farms certainly could do that. I mean, think of the farmer's Almanac, if you guys have ever looked at one of those. They still make their money trying to guess what's going to happen in the next year and a lot of people believe it. A lot of farmers, particularly smaller farmers, would make decisions like that.