Assessing and Advancing Digital Agriculture in Nebraska

Guillermo Balboa, Research Assistant Professor in Nutrient Management and Digital Agriculture, Department of Agronomy and Horticulture, University of Nebraska Author

01/03/2025 Added

3 Plays

Description

"Nebraska is a leader in adopting Digital Agriculture (DA) tools across the U.S. This seminar will assess the state of DA in Nebraska and highlight ongoing efforts to develop, evaluate, and promote tools and technologies that address the state's agricultural challenges."

Searchable Transcript

Search:

[00:00:00.640]The following presentation is part of the Agronomy and Horticulture Seminar Series
[00:00:05.760]at the University of Nebraska-Lincoln. So thank you all for attending today's seminar for those
[00:00:15.120]who are in person and for those who joined us online as well. So there will be 45 to
[00:00:22.000]15 minutes presentation followed by 10 to 15 minutes of question and answer session.
[00:00:28.080]So for those who are within the room, I will pass on this mic for whosoever want to ask a question.
[00:00:35.120]And for those online, please put in your question in chat box and I will try to get an answer
[00:00:41.200]at the end. So our guest speaker is Dr. Guillermo Balboa. He's a research assistant professor in
[00:00:50.000]nutrient management and digital agriculture at UNL. So he did his master's and bachelor's from
[00:00:57.680]Rio Carto National University in Argentina. Then he did his PhD from Kansas State University in
[00:01:05.280]site-specific nutrient management with Fulbright Scholarship. Then his journey took him to
[00:01:12.800]Australia for a little bit to Commonwealth Science Industries and Research Organization where he
[00:01:18.960]worked to integrate digital agriculture in Australian farming systems. Then he joined UNL three years
[00:01:27.280]back in 2021 in Precision Ag Lab and in this January he started his own farming digital lab
[00:01:35.920]and his vision is to integrate digital agriculture in our traditional cropping system approaches
[00:01:42.640]in Nebraska and beyond. So today he will share his part of his program in this
[00:01:49.760]presentation about the state of the art of digital agriculture in Nebraska. Dr. Babbo.
[00:01:56.880]Well thank you, Mandeep, for the introduction. Thank you for the committee to invite me
[00:02:01.760]to present at the seminar series. It's an honor, it's a great opportunity to have these sessions
[00:02:07.840]at the department level to interact and also for external people to join. So as Mandeep mentioned,
[00:02:15.360]the talk of my presentation will be presenting basically the first half of an assessment on
[00:02:21.600]digital agriculture that we conducted last year and the second half will be more focusing on
[00:02:26.480]things on digital agriculture that we are doing some of the projects and we have a complete list
[00:02:33.440]at the end. So basically that is kind of the organization of my talk and my goals for today
[00:02:41.200]when you leave this room will be to answer some of these questions. For example, what is digital
[00:02:46.320]act to have a more complete understanding of the concept and to have some idea of what is
[00:02:52.400]the status of digital agriculture here in Nebraska.
[00:02:56.080]With this assessment that we did and then some of the things that we are doing and some of the
[00:03:01.920]projects that we're leading and others are leading here to tackle some of the Nebraska challenges like
[00:03:07.360]the nutrient management of other crop management strategies and to start you know when I was
[00:03:14.240]preparing this presentation I was thinking should I go deep and focus in the full research
[00:03:20.960]and in people that is on the topic or should I start with something super simple and try to
[00:03:25.680]get more in deep in some of the slides so I will decide just to put a satellite imagery from a
[00:03:31.860]field that we have an experiment this year and a lot of things that you can see in this imagery
[00:03:36.440]20 years ago we don't have the capability of having this in our computers just log in into
[00:03:43.200]a system and you can basically have a satellite imagery of a field any part of the world and we
[00:03:49.500]can see that this field for example is under irrigation and we can see that there is some variability in
[00:03:55.280]color there are some tracks for the irrigation circle and then we have today the technology
[00:04:01.280]available to put for example a yield map and a yield map I will say is one of the most known
[00:04:07.540]tools of precision agriculture that was born many years ago that is available today that
[00:04:14.060]pretty much all the combine has this capability of producing a yield map a yield map for those
[00:04:20.660]that are not specifically in the area is a map where it's showing the productivity
[00:04:24.880]of different sites so it's basically a sensor that is saying how much bushels you're having
[00:04:31.980]a specific place and it's tied to a GPS so you're giving a location and basically putting
[00:04:37.320]that information in the software it allows us to make different categories this map for
[00:04:43.100]example we received yesterday so it's not clean but you can tell that there are differences
[00:04:47.740]in yields so the green is associated with higher yields the red is associated with lower
[00:04:52.400]yields and then you have some interviews.
[00:04:54.480]Intermediate categories some of the variability that we were seeing before is being captured
[00:04:59.460]with this type of sensors as you see here that lighter color in the soil is associated
[00:05:05.640]with you know less yield and some other areas are associated with higher yields and then
[00:05:12.040]there's some other variability that is showing in that map that is basically generated by the
[00:05:16.920]practice of having that irrigation circle so today basically we have the capability of using
[00:05:24.080]technologies to generate and collect information collect data I will say because it's not
[00:05:29.060]information yet from different fields this is the most well-known tool that is a yield map but then
[00:05:34.400]you can collect satellite imagery so you can use a drone as a platform to use a camera multi-spectral
[00:05:41.300]camera RGB camera to collect information you can use a differential GPS to make an elevation map
[00:05:46.880]and then all that sources of information are available you know to of data to process
[00:05:53.680]to generate information and to make a decision so if you are seeing this and you're seeing that you
[00:05:58.840]have a lot of variability within the field the question will be why most of the farmers in
[00:06:04.600]Nebraska in the U.S. and in the rest of the world are pretty much doing the same management across
[00:06:10.600]the field if you have different if you have variability means that we have different
[00:06:14.840]requirements of resources and if you are applying the same I will say management or inputs to the
[00:06:23.280]the consequence will be that in some areas we might be more efficient and in some areas we have less efficiency.
[00:06:30.040]Okay so these tools have been around for many years and a lot of research have been done but the adoption
[00:06:35.360]is remains overall low so the idea was to make an assessment at the state of Nebraska and to see what
[00:06:44.560]was the adoption of these technologies to use as a baseline to move forward the research. I said before
[00:06:52.880]that the adoption of this technology was low but in general terms this was a survey
[00:06:59.200]and a study was conducted by USDA and showing that Nebraska is one of the state leaders with
[00:07:06.160]pretty much almost 50 percent of the farms using any type of these technologies in their crop on
[00:07:12.080]livestock operations. To me this is kind of an overestimation but then the general confusion that
[00:07:22.480]we have here is that people confuse what is precision ag with digital ag and I will say
[00:07:29.120]that as I mentioned before we have technologies to produce this data related to a GPS pointing
[00:07:34.560]and spatial and temporal variability but then throughout the years we have been adding different
[00:07:38.880]sensors like IoT sensors equipment that we put into the field that sends something in the field
[00:07:44.720]that can be for example water or it can be temperature or something and it sends that
[00:07:48.880]to a cloud and then we receive that information
[00:07:52.080]somewhere in the world and we analyze it or we can for example collect all that information
[00:07:57.520]and process in in other platforms and you have different levels of connectivity
[00:08:01.280]so IoT is a technology that is not considered precision act but it will be
[00:08:06.640]more towards what we call digital act right so what we decided then is to do a survey across
[00:08:13.360]the state and what you are seeing there is the state with different counties and
[00:08:16.880]the different calls are the number of responses that we got we sent 2500
[00:08:21.680]surveys and we work with the Bureau of Sociological Research to make sure that
[00:08:25.600]we're sending and the number of surveys to each of the countries to reflect the
[00:08:30.560]amount of farmers that we have in each of those so that will be different
[00:08:34.300]colors that you see and those are specifically the responses that we got
[00:08:37.120]and we asked 40 different questions about digital agriculture mainly and
[00:08:42.460]then since we are working with a lot of nitrogen management projects the second
[00:08:47.020]part of the question of the questionnaire was in relation of the use of digital
[00:08:51.280]lag to manage nitrogen in Nebraska. So in the next couple of slides I will be
[00:08:55.920]showing you some of the results that we got. The first question that we asked them
[00:08:59.980]is did you hear about digital lag? Do you know what is digital lag? And if you can
[00:09:04.160]see more than 50% they don't really know they don't have a clear definition of
[00:09:08.620]what is digital lag. Only 23% of the farmers say that yeah they know the
[00:09:12.880]concept or they know what we are talking about and 22% of the farmers they say I
[00:09:17.200]don't I never hear that word or I don't really know what it is right.
[00:09:20.880]So we asked them after that to make a list of five words to define that
[00:09:28.360]concept if they hear it before and that's the word cloud that we have. The
[00:09:33.180]bigger the size of the word is the more times that they repeat it. Most of the
[00:09:37.980]farmers they were only able to put one or two words and the most mentioned or
[00:09:42.960]the highly rated what computer GPS data precision technology variable rate so
[00:09:48.000]and then there are some
[00:09:50.480]the rest of the words are in kind of always in kind of mentioning the
[00:09:56.720]technology and not focusing in the negative side I will say of the
[00:10:02.420]technology. So to bring a definition and that this is kind of the most known
[00:10:09.960]definition of this lag is the use of technologies that combines multiple data
[00:10:13.880]sources and analytics methods to integrate into a system that allows
[00:10:18.680]farmers but also
[00:10:20.080]stakeholders in the agricultural value chains to improve food production.
[00:10:23.740]Improve food production is kind of a super huge topic but it can be improved
[00:10:27.700]productivity, improve efficiency and also reduce the environmental impact.
[00:10:33.540]And if you see here I just put in different in the different figures I
[00:10:37.480]will say the different branches of digital lag and as mentioned before
[00:10:41.260]precision lag is one of the biggest one but then you have IOT, big data, the use
[00:10:46.120]of digital platform for example, the use of mobile applications,
[00:10:49.680]or technologies like blockchain that allows us to track all our production
[00:10:54.660]throughout the different systems as allows us to calculate different
[00:10:59.640]metrics or trace like the carbon footprint. So basically it's getting
[00:11:04.200]information making the right analysis of that information to be able to make a
[00:11:09.660]decision and to have an impact on on the system so we can record what is the
[00:11:15.180]value of that information right. So if this process
[00:11:19.280]is not complete then we are not fulfilling I will say the goal of the
[00:11:24.560]digital lag that is making an impact in the agricultural value change. Having the
[00:11:28.660]technology is not itself considered to be digital lag and I just put this
[00:11:33.220]because sometimes you hear that people talk about digital lag and precision lag
[00:11:37.220]and they use the terms referring to the same thing that I will
[00:11:41.600]use this analogy. Digital lag will be kind of an
[00:11:44.240]umbrella and then precision IoT and big data can be the different sections or
[00:11:48.880]other than all the ones that are here right so it's a kind of a larger concept.
[00:11:53.600]So in that survey we just put together, sorry about the size of the
[00:11:59.620]different categories, we put a list of 33 technologies that we consider digital
[00:12:04.780]agriculture technologies that we surveyed before in Latin America
[00:12:08.860]and we basically asked the farmers if they knew their technology, that is
[00:12:14.120]the blue bars, and then if they use the technology, those are the green bars, and
[00:12:18.480]then I just added like a pin indicating the technologies that usually are
[00:12:25.200]related with the use of a GPS so those can be categorized as a precision
[00:12:29.760]technologies and it's just put a line in the 50% just to show that usually the 50%
[00:12:35.080]when we have more than 50% of adoption for technology we can say
[00:12:39.060]that the technology was fully adopted and you can sell that the technologies
[00:12:43.140]that most farmers are using is the phone applications that are related to
[00:12:48.080]precision to management of some aspect of the fields following by GPS yield
[00:12:53.600]monitor solid examples so a little bit variable of the precision acting
[00:12:58.560]noise are in LA at the most adopted but you can see in the top robotics or the
[00:13:04.640]use of blockchains or beautiful fencing or automatic feelings there are a big
[00:13:09.140]list of of technologies for example the use of and in vegetation indexes that
[00:13:14.880]has less than 10 percent of adoption even though
[00:13:17.680]they have been a lot of research around just this give us a huge quick picture
[00:13:22.020]that beside what this initial service said there was a huge adoption or
[00:13:26.920]Nebraska was leading with 50% there the number of technologies adoption is not
[00:13:33.600]super high just focusing on the last category that is basically phone
[00:13:38.680]application is the first one in adoption which is classified in the survey and we
[00:13:43.220]asked them what they look when they're looking for these applications and we're
[00:13:46.800]seeing that the use of the phone application is more toward getting basic
[00:13:51.720]information like weather markets and commodity prices but then there are other
[00:13:55.560]applications that are very useful and helpful like applications to manage
[00:14:00.080]nutrients for seed selection diseases scouting you know that are very super
[00:14:05.380]valuable but only have 20% so these are apps that I would say they're more
[00:14:09.940]complex and can have a huge impact but the adoption of those are low and some
[00:14:15.920]analysis that we did in the past is to classify those applications in in four
[00:14:21.440]different categories going from according to the complexity and the
[00:14:25.280]value that they are bringing with the data they provide us and basically
[00:14:28.760]answering four type of questions one one the applications that can be
[00:14:33.160]descriptive or can just provide an answer to a question of what what is the
[00:14:38.180]weather what is the price of X Y and C the second category is a diagnostic
[00:14:43.360]analysis type of thing so
[00:14:45.820]allows us to to basically make a diagnostic for example it can be for a
[00:14:50.560]disease or for a nutrient deficiency the third category there is more complex and
[00:14:54.640]can give us a lot of values any type of platform application that can help can
[00:14:59.260]make a prediction on what will happen and the last one are more complex is all
[00:15:03.980]those type of applications that or platforms that can give us a
[00:15:07.580]prescription about something it can be a viral rate application or a predicting
[00:15:12.220]the weather in the future like we will see in some slides in the future
[00:15:15.400]these are survey something that we did in Latin America with this initial
[00:15:21.700]categories we went through four different countries in Latin America and
[00:15:25.540]we were benchmarking for these applications how many were descriptive
[00:15:30.640]for diagnostic predicting and prescription and you can see the size of
[00:15:34.240]this donuts reflects the amount of apps in each of the groups and you can see
[00:15:39.760]that the most useful the ones that can predict or make a prescription has a
[00:15:44.160]lower number
[00:15:44.980]right in terms of numbers in Latin America and Brazil and Argentina are the
[00:15:50.440]countries leading just to show you that you know how you can classify the
[00:15:55.060]different applications that you can use in terms of continue with the survey and
[00:16:00.340]just a couple more slides we ask the farmers if they record the value of
[00:16:05.500]these like the benefit and we put a list of benefits like reduction of inputs
[00:16:09.640]increase of profits increase in yields most of them have more than 80% of
[00:16:14.020]selection
[00:16:14.560]they were able to select multiple transfers but only three out of ten
[00:16:17.860]almost three out of ten farmers they basically know the value of the things
[00:16:21.400]that they are using
[00:16:22.220]ok and this is very important and we'll see in the future why because is one of
[00:16:26.380]the main barriers to adoption where some farmers are using the technology but
[00:16:30.280]they're not very sure on what is this we talk about reduction production of
[00:16:34.060]inputs and you know they cannot quantify how much was that reduction and that is
[00:16:38.860]also is that and it's not allowing for the technology to move forward
[00:16:44.140]faster there's a lot of information in this but in this slide but it's
[00:16:48.400]basically a summary of the barriers for technology adoption so from those 33
[00:16:53.740]technologies that we show before we ask them what is different levels of no
[00:16:58.480]barrier moderate barrier high barrier or if they're not sure if that is a
[00:17:02.140]barrier I will put different categories and maybe 10 or 15 years ago technology
[00:17:06.820]cost was the main response to this question but you can see that the main
[00:17:10.900]one here with 70% condensing
[00:17:13.720]the moderate and the high barrier is a lack of information about the value of
[00:17:17.800]that technology so farmers they don't really know if the technology is what is
[00:17:24.340]the benefit of using that technology pretty much all of them are high but we
[00:17:28.100]can identify also the lack of qualified labor to manage that technology again
[00:17:32.860]new barriers are rising and this is something to take in account in that a
[00:17:37.820]review that we did a couple of years ago we identify some economical and social
[00:17:41.620]environmental indicators
[00:17:43.300]that we can track to basically show the benefit so in any technologies are using
[00:17:48.020]in the field you can basically pick some of these some of these indicators and
[00:17:54.040]use them to characterize this just a list of examples to put numbers into
[00:17:59.140]different technologies so we are able to show the real value of different
[00:18:03.220]technologies the second part of that survey that I mentioned before was more
[00:18:07.280]relating to having a screenshot of what was the the the logic behind the
[00:18:12.880]nitrogen management in in Nebraska and we try to classify those and do some
[00:18:18.580]analysis by by different ages or by different site farm farm size and then
[00:18:24.940]after that we ask them what are the technologies that these are like
[00:18:27.860]technology that they used to make those decisions this is just basic information
[00:18:32.080]this kind of it's kind of similar is there's no statistical difference a
[00:18:37.200]trend on the younger the generation in the found they tend to apply more
[00:18:42.460]nitrogen these are statewide they are not separated by county but we have that
[00:18:47.020]information by NRDS as well then we also separate these by farm size and
[00:18:54.340]specifically this was for corn so we tend to see that smaller farms they tend
[00:18:58.940]to apply more and then there is a medium side farm I will say in between a
[00:19:03.100]thousand and two thousand acres are the ones are applying and kind of more
[00:19:07.140]nitrogen right this is just a characterization and to see observe some trends
[00:19:12.040]to see if we can analyze the results with more in deep so we asked them about
[00:19:17.780]the four rates the right the four hours right rate right timing right placement
[00:19:22.040]and right source this is just for example the right timing and you can see
[00:19:26.440]the younger the farmer there are more towards making applications early spring
[00:19:32.140]or split applications and when you know the older farmers that have a different
[00:19:37.320]thinking on how how they can they are thinking the splitting
[00:19:41.620]or the application nitrogen there is a limitation the split there because I
[00:19:45.340]can include fall winter applications that we are seeing that younger farmers
[00:19:49.600]are kind of and not to all that technology over there I would just skip
[00:19:54.280]these ones but basically we can also analyze and the timing and this in this
[00:19:59.980]case we just separated them by NRD and you can see this is interesting because
[00:20:04.540]allows you to see statewide what is the the timing of the nitrogen application
[00:20:09.340]and they were this can help us to target
[00:20:11.200]studies for example for for fall winter application we can see that this is the
[00:20:16.660]area the southeast area of the state is where it's more focusing and then when
[00:20:21.040]we are trying to to take a look to what is the adoption of split
[00:20:27.680]application we can we can see that in the different areas or NRDs of the
[00:20:32.740]of the state and and that's I think is very useful information I wanted to
[00:20:38.860]arrive to this section that is one of the
[00:20:40.780]methods that they use and farmers apparently they trust in a lot they use
[00:20:46.020]and they could select here multiple options so lab recommendations but also
[00:20:50.440]the second one the intuition traditional rate or own experience is having a huge
[00:20:55.660]impact of how they define that natural rate and all the I would say newer
[00:20:59.620]technologies or models or digital like things like chrome models platforms
[00:21:03.600]sensor base or even a calculator has less than than 25% of adoption and this
[00:21:09.500]is a statewide
[00:21:10.360]we run this survey as well through the Nebraska on for research network all
[00:21:14.500]these last values are going up and also run that through the closed group that
[00:21:19.300]we work together in nitrogen projects and all these numbers are going up I
[00:21:23.320]just wanted to keep simple in this slide but you know they're more advanced
[00:21:27.460]method that that can be beneficial for farmers of core taking in taking account
[00:21:34.240]all the the traditional experience that they have but put in numbers in order to
[00:21:38.500]fine-tune those rates
[00:21:39.940]this is the last one just to characterize on the characterization and
[00:21:45.880]we ask them why you know some of these technologies we are showing that there
[00:21:50.020]are there are some there's some room to reduce basically the rate without
[00:21:55.000]penalizing yields and basically what farmers are worried is that they they
[00:21:58.600]reduce and the rate they can the yields can be penalized or the profit can be
[00:22:05.440]reduced or if they when we talk about for example splitting application that
[00:22:09.340]can be
[00:22:09.520]a better strategy they can be worried about potential wet weather that
[00:22:15.100]don't allow them to get into the field to do that application if they
[00:22:18.880]if they're in a dryland condition they don't have a fertigation system for
[00:22:22.720]example and then again financial stress learning curves and then distrust in
[00:22:27.640]this case of precision AG is kind of lower one question is what is
[00:22:33.580]natural conservation districts and those are the one that that's good
[00:22:39.100]life questions and so in the second part of the presentation this is just to to
[00:22:46.240]show you some of the the barriers when we are talking about applying some of
[00:22:51.320]these technologies basically this is what they are mentioning that is is not
[00:22:56.060]allowing them to to go for them and the barriers that they identify so the second
[00:23:01.840]part of this talk will be kind of how we use these DALAC tools to contribute to
[00:23:08.680]adoption of more of a larger number of these DALAC as management technologies
[00:23:15.100]and and to start with that the first question will be if farmers know what is
[00:23:19.840]the efficiency level that they are having in their that they're having in
[00:23:23.660]their own fields and for that I was just presenting one one of the projects that
[00:23:29.180]is a benchmark dashboard that this idea was initially brought by Dr. Patricio
[00:23:36.060]Grassini so there's no
[00:23:38.260]accessible tool that helps growers today to know the current efficiency and to
[00:23:45.860]track that among the years in specifically in corn production. The University for
[00:23:50.320]example is doing a lot of research and that information is not being
[00:23:54.040]systematized to provide a specific answer to the farmers in general so the
[00:23:59.040]idea was to develop a dashboard that can help farmers with their own data to see
[00:24:03.220]what are the levels of efficiency and to compare these levels with other farmers
[00:24:07.140]and
[00:24:07.840]here you can see this is a figure with how much nitrogen is being uptake and how
[00:24:12.100]much grain you can produce so you can have the same yield with three different
[00:24:15.640]levels of efficiency of course nitrogen is a super complex nutrient that's why
[00:24:20.900]we have this heterogeneity so basically by farmers knowing these they can know
[00:24:26.920]if they are in a good path or if they can improve and basically
[00:24:32.160]this dashboard is ready to be entered into the testing phase what is basically
[00:24:37.420]farmers can do they can select the legal description of the field and then
[00:24:40.960]they need to enter just the yield how much nitrogen they apply and if they
[00:24:46.000]have irrigation how much how many inches of water and if they have nitrates in
[00:24:49.780]the irrigation and that value into the equation to account for that and then
[00:24:54.940]the system will retrieve information so data and other parameters and basically
[00:24:59.500]this is a test user for example I have 12 fields and 110 acres and basically what this
[00:25:07.000]dashboard can tell the farmer classifying to into green yellow or red
[00:25:12.700]according to the level of efficiency in this case this means that is screen
[00:25:17.800]means that he's in the top 25% of yielding farmers for that area and we're
[00:25:23.800]calling this we have three different levels of classifications similar fields
[00:25:27.400]so he can compare himself with farmers that has similar soils and
[00:25:33.160]weather characteristics and we're using well I'm not going into details
[00:25:36.580]for that but they can compare with farmers are in the same area or some
[00:25:40.180]farmers are in the same county that's a political region division so they cannot
[00:25:44.380]see the information but they can see the aggregated and they can compare in this
[00:25:47.980]case the this farmer is in green but he could be in yellow or in red if he's in
[00:25:52.520]red means that for the management that he's doing there are farmers are getting
[00:25:56.320]more yield in that case if you select nitrogen you in this case he's in green
[00:26:00.640]for both but he could be green here and red here because maybe he has a good
[00:26:04.820]yield but he's applying
[00:26:06.160]much more nitrogen than farmers in there it's a very basic I would say light
[00:26:10.240]lighting system to classify how they are performing and basically the phase two
[00:26:16.900]of this dashboard will be to combine all university research to say okay if you
[00:26:22.420]are in this area you are doing good you can just improve by using A B and C if
[00:26:26.980]you're in this area by seeing the management you can try the A B and C
[00:26:30.880]technology or if you're in this case you can try all these other sites
[00:26:35.740]sets of technologies. So basically this is a very simplistic
[00:26:41.200]approach it's a benchmark that's why it's not giving any specific
[00:26:44.740]recommendation it's just telling the farmer it's recording information so
[00:26:48.580]they can select the year they can select if they have multiple fields they can
[00:26:52.860]filter different fields but it allows them to see for the different years how
[00:26:57.740]they were performing and that will only work if other farmers get the data input or
[00:27:05.320]if we make agreements with NRDS to put that information into the system
[00:27:10.680]another dashboard that we are working is a dashboard for agricultural water use and
[00:27:15.580]nutrient management specifically here so the previous one was a benchmark thing
[00:27:20.080]in this case what we are using this national project we are participating in
[00:27:24.760]with different universities University of Maryland developed I will say and I
[00:27:29.780]will show you here in this slide a forecast system and basically
[00:27:34.900]is this one over here that if you see it's called Seaworth and what this
[00:27:40.600]theory is showing you is error in the prediction for example for summer
[00:27:43.580]temperatures and precipitation and if we look to the Corn Belt the more white
[00:27:48.700]means that the prediction is has more accuracy the more blue or green is that
[00:27:54.040]you are having more just just as a picture to compare that this is what
[00:27:58.120]NOAA is using to give us all the prediction nowadays this is a new system
[00:28:02.740]that it basically is
[00:28:04.480]using artificial intelligence to focus the prediction into the corn belt so
[00:28:09.860]it has by reducing the scale of the of the model because NOAA is a
[00:28:17.040]regional scale it will improve accuracy so we are using this input
[00:28:23.380]so it's an input of weather prediction basically to three to six
[00:28:28.520]months to develop to use it as an input for different tools we have a growing degree
[00:28:34.060]data tool we have a dry down calculator we have a data viewer that basically
[00:28:38.200]allows you to see predictions for temperature and precipitations for the
[00:28:42.700]future and give you the probability of a currency what is the difference between
[00:28:46.480]this you can say oh there are other growing degree tools out
[00:28:52.420]there but basically those tools are based on on average for weather and
[00:28:57.040]they are not using the prediction they're using basically normal data in
[00:29:01.780]this case for example this is the
[00:29:03.640]normal data for last year and this orange was the prediction of the model
[00:29:07.420]and it can help you to better manage your night your your your crops and to
[00:29:12.500]see you know the the probability of a currency in this case of the different
[00:29:17.460]phenological stages so basically this platform right now is giving you a
[00:29:23.940]forecast so you basically select any part of the Corn Belt and you can access
[00:29:28.220]a forecast is publicly available and in December there will be a next
[00:29:33.220]phase three release where we are incorporating a nitrogen sorry an
[00:29:37.960]irrigation tool and then it will be incorporated also a corn management tool
[00:29:43.280]that will include nutrients and and some other
[00:29:47.920]management practices, like cover crops,
[00:29:49.600]that will allow you to select any place in the Corn Belt
[00:29:53.460]and basically play with different strategies.
[00:29:56.200]And they will run them all, and they
[00:29:57.620]give you a prediction on yields and other parameters.
[00:30:01.640]So again, this is just-- the first one
[00:30:03.720]was a benchmarking dashboard.
[00:30:05.660]This is something that allows you
[00:30:07.520]to make different decisions, right?
[00:30:09.580]And this is already available online.
[00:30:13.320]Going to the last part of the presentation,
[00:30:16.680]I will just focus on how we can use crop models as a tool.
[00:30:21.580]And I will include crop models as a digital act tool,
[00:30:24.120]even if they have been around for many years,
[00:30:27.180]the capability that they have to allow us to make decisions.
[00:30:31.420]And this is more for the research side.
[00:30:33.440]What we did here in a study, we used apps in here.
[00:30:37.980]We basically were trying to explore--
[00:30:40.560]and this was in Australia--
[00:30:42.620]if we can incorporate new crops in different areas.
[00:30:45.320]If you have a problem with--
[00:30:46.520]let's say, corn in Nebraska, what
[00:30:47.960]are options we have in terms of crops?
[00:30:50.020]So we selected, in this case, was sorghum, different locations
[00:30:52.680]with different soil types.
[00:30:54.240]We set it up, the model, for different initial water
[00:30:56.560]conditions.
[00:30:57.280]And we ran historical databases from 1940 to 2015.
[00:31:03.320]And basically then, we constructed probability
[00:31:07.340]categories for the different scenarios,
[00:31:09.620]in this case, of initial water holding capacity.
[00:31:13.280]And we produced these for each of the combination
[00:31:15.320]of the planting dates.
[00:31:16.360]So basically, what we can get here after, I think,
[00:31:19.960]this was 10,000 simulations, we can basically
[00:31:22.900]predict for this specific location,
[00:31:25.140]in this case, for a full profile,
[00:31:27.120]sown in August 1 with the crop, there
[00:31:30.940]is a 25% probability of having, for example, 5 tons of yield.
[00:31:36.800]So basically, we did this for many crops,
[00:31:40.540]for many locations, and for many planting days.
[00:31:43.360]And then we selected 10 or 12.
[00:31:46.200]And we met with the farmers to check
[00:31:49.000]if those scenarios were realistic to them,
[00:31:52.480]having in account their experience.
[00:31:54.440]And this was early adopters.
[00:31:56.120]This is an example of how we can use
[00:31:58.480]what we call big data generated by models
[00:32:00.760]to make decisions in the future at a regional level.
[00:32:04.160]In this case, it can allow us to play with different crops
[00:32:07.040]to incorporate in a specific area.
[00:32:10.640]This study also included, for example,
[00:32:12.460]tracking the variability in the weather.
[00:32:14.320]So usually, we have huge databases
[00:32:16.040]and this case was 90,000 records for this area.
[00:32:20.420]And by doing this type of analysis, for example,
[00:32:22.620]we were calculating different traits.
[00:32:25.260]And this case was the numbers of days
[00:32:27.780]with temperatures in the summer above 35 Celsius.
[00:32:31.360]And we are seeing that from 1940s, value was close to 48.
[00:32:36.940]But in the last normal that we ran was close to almost 65,
[00:32:42.100]I think.
[00:32:42.800]So there is a trend that we can calculate by using
[00:32:45.880]and this was done, again, with R.
[00:32:48.320]And then you can basically get big databases
[00:32:50.800]and do some of this type of analysis
[00:32:52.360]to get quick information on how weather, for example,
[00:32:55.640]is impacting in a specific area and getting more value out
[00:33:00.260]of the data that is out there.
[00:33:01.660]And these are some skills that, for example, any student
[00:33:05.140]graduating could have.
[00:33:06.940]And besides producing the traditional metrics,
[00:33:10.680]you can start doing this analysis
[00:33:12.100]and see how weather is changing throughout the time.
[00:33:15.720]In this case, this was part of a paper that we published again.
[00:33:22.340]And again, it's using crop models in long-term simulations
[00:33:27.360]to evaluate different management practices
[00:33:29.520]and what we call common practices.
[00:33:31.680]This was what farmers are doing.
[00:33:33.960]This was around Kansas.
[00:33:35.480]And we compared to ecological intensification.
[00:33:38.140]And then we ran this simulation from a set of numbers.
[00:33:42.120]And we calculate these means, and we separate them
[00:33:45.560]by different weather scenarios.
[00:33:47.540]The weather is changing, so in cool and wet environments,
[00:33:50.000]we are seeing that when you intensify,
[00:33:52.060]you can decrease that gap between the potential
[00:33:56.060]by 2.5 times.
[00:33:57.140]So a combination of management practices
[00:33:59.500]can help to reduce that gap.
[00:34:01.140]But we also have scenarios, like in warm and dry years,
[00:34:03.860]where no matter what you do, you can reduce that gap,
[00:34:06.860]but the impact will be smaller.
[00:34:09.140]So anyway, this is kind of a more high-level thing,
[00:34:12.020]but it's how we can use crop models to guide our--
[00:34:15.400]our decision-- our cropping system decisions
[00:34:19.680]in a regional approach, and to see
[00:34:21.820]how the different weather scenarios can
[00:34:23.980]impact the benefits of different combination
[00:34:26.780]of different management practices.
[00:34:30.720]Going to what we are doing here in Nebraska now,
[00:34:32.880]we are kind of using the same approach
[00:34:34.860]to try to calibrate, in these cases,
[00:34:38.000]again, APSIM model with different management
[00:34:40.540]practices.
[00:34:41.120]These are some of the experiments
[00:34:44.280]that we are evaluating.
[00:34:45.240]We are evaluating here at N-rate with different timings
[00:34:47.640]and different rates, use of cover crops,
[00:34:50.000]and enhanced efficient fertilizers.
[00:34:51.860]So we're trying to use the model in order
[00:34:56.560]to kind of simulate what is happening in reality.
[00:35:00.280]But also, in this specific case, my goal
[00:35:02.760]is to combine crop models with information
[00:35:06.000]that we can get during the season.
[00:35:08.020]This is the flies that we have been doing
[00:35:12.440]over the experiment.
[00:35:13.280]Each of these plots are--
[00:35:15.080]of the experiments, so those differences
[00:35:16.800]that you are seeing there are the color crop strips
[00:35:19.460]that we have, and basically how we
[00:35:21.180]can couple the benefits of using crop models
[00:35:24.320]with the benefits of using any in-season index that
[00:35:27.460]can help us to correct that model in the prediction
[00:35:31.040]and to be more accurate.
[00:35:32.000]Instead of saying, let's go with crop models,
[00:35:34.480]let's use any sensing technology,
[00:35:36.840]we can couple the benefits of both.
[00:35:38.660]When I was doing this project, we found out that this is
[00:35:44.920]a specific model that we're using was not capturing very
[00:35:47.920]well what was happening under the surface and the root.
[00:35:51.520]So we started-- it was not the initial plan,
[00:35:54.280]but it went that way to see how we can explore the root system
[00:35:58.440]in all the experiments that we are doing in relation
[00:36:01.120]to nitrogen management.
[00:36:02.420]So basically, we set up a root washing facility
[00:36:05.160]that allows us today to process a hundred samples that we
[00:36:09.840]collect from different depths in different experiments
[00:36:13.460]in our project.
[00:36:14.760]And we can basically analyze them, scan them,
[00:36:18.240]not only take a look at the root length,
[00:36:19.740]but also to the biomass, and see how we can relate
[00:36:22.980]those traits to irrigation levels
[00:36:25.920]and to nitrogen parameters that we are measuring.
[00:36:29.580]This is just last year data, but we
[00:36:31.760]are collecting this every year.
[00:36:33.180]And we can see how the irrigation is impacting
[00:36:36.840]the root depth and biomass.
[00:36:39.260]And basically, we're using that to calculate--
[00:36:43.880]go to the--
[00:36:44.600]this one-- to improve this ratio that is part of the model.
[00:36:48.860]That is the root-to-shoot ratio.
[00:36:50.220]This is how much growth we have per unit of aerial biomass.
[00:36:53.700]So basically, all this work with roots
[00:36:55.800]is meant to fine-tune this specific indicator,
[00:36:58.720]to fine-tune that model that we are
[00:37:00.360]using to predict yields and to better manage natural.
[00:37:03.740]Just to see how deep we can go sometimes
[00:37:06.260]if we are trying to see why we are not capturing
[00:37:08.560]the variability in temperature and other parameters
[00:37:11.440]that this can be the option.
[00:37:14.440]Just moving forward, and the last part
[00:37:18.500]is just to show how we are analyzing different experiments
[00:37:24.220]that we are conducting in our team.
[00:37:26.980]And this is part of the CIG project.
[00:37:30.960]It's conservation innovation grants
[00:37:32.420]where we are testing different nitrogen practices.
[00:37:35.900]And the options that farmer has to test
[00:37:37.620]are crop models, remote sensing, biologicals,
[00:37:40.200]and enhanced efficient fertilizers.
[00:37:42.340]And basically what we are doing is trying to put
[00:37:44.280]together historical data that we have in this case,
[00:37:47.920]for example, for crop models where we are seeing
[00:37:49.780]that there's not too much impact in the yield,
[00:37:52.620]but we are reducing the amount of nitrogen
[00:37:55.920]that we are applying.
[00:37:57.400]And again, beside the overall response in the field,
[00:38:00.720]we are seeing what is happening in different levels
[00:38:03.360]of productivity, in high productivity environments.
[00:38:06.560]For example, crop models in this case,
[00:38:08.500]this particular model for these are nine sites
[00:38:11.200]were showing more response in high yield
[00:38:14.120]in areas within the same field, and these medium yielding areas
[00:38:18.000]and these high yielding areas.
[00:38:19.820]And then we are going to each of these areas of the field,
[00:38:23.340]and we are calculating different traits,
[00:38:25.100]like topographic traits or indexes.
[00:38:27.680]And basically, this is a real plot
[00:38:30.440]that is showing you that when the variable is farther
[00:38:32.880]from the center, it means that that variable is explaining
[00:38:35.220]the response of that trait.
[00:38:37.400]In this case, for example, if you go to adapt n,
[00:38:40.740]it's showing, for example, for the nitrogen model,
[00:38:42.900]that the topographic, when it's
[00:38:43.960]indexed, so how the water moves into the system,
[00:38:47.500]is showing the larger response for that specific model.
[00:38:51.760]So the idea is not just to test, do experiments
[00:38:55.700]at an on-farm research level, but also
[00:38:58.200]to start taking a close look from the research side
[00:39:01.200]to what are these things that are explaining
[00:39:03.480]the different responses that we are having when we compare,
[00:39:06.700]in this case, crop models.
[00:39:08.620]In this case, what's adapt n with what the grower was doing
[00:39:12.680]in this particular case.
[00:39:13.800]This is another example for sensor base.
[00:39:16.360]And here you have 60 different experiments
[00:39:19.760]that were conducted between 2015 and 2023.
[00:39:23.600]And basically, what we found here
[00:39:25.720]is that there is an increase in natural use efficiency
[00:39:29.860]when you use any remote sensing technology.
[00:39:33.740]The natural rate was decreased around 40 pounds.
[00:39:37.400]And there was, I would say, a yield penalty
[00:39:39.680]around four bushels.
[00:39:41.920]But overall, putting everything together
[00:39:43.640]and these are different sensing technologies,
[00:39:46.400]there is a benefit in the efficiency of the system.
[00:39:50.420]There is a reduction in the rate that the technology is
[00:39:54.460]advising compared to the farmers.
[00:39:56.540]And you can see here in the probabilities here,
[00:39:59.360]for example, in 50% of the cases,
[00:40:02.960]the model was giving a lower rate compared
[00:40:06.560]to the blue that is the farmer.
[00:40:08.240]So there's a probability accumulated.
[00:40:11.060]So here in 50%, you're seeing that farmers
[00:40:13.480]are always above compared to the growers.
[00:40:16.520]And this is specifically for this sixth experiment
[00:40:19.780]that are more than 100,000 points analyzed
[00:40:22.720]behind this technology.
[00:40:24.800]Some of the other things that we are developing,
[00:40:27.960]and I think I need to go a little bit closing now.
[00:40:32.540]This is something that we are developing,
[00:40:34.300]taking account all the things that we mentioned before.
[00:40:37.300]This is a tool that we are developing
[00:40:39.480]with anyone can just basically pick a field,
[00:40:43.320]any part in the US and basically can retrieve
[00:40:46.840]sadly information, calculate different indices,
[00:40:50.220]and basically make a cluster analysis
[00:40:52.840]and show us the variability that you have.
[00:40:55.100]The idea behind this and all that worldwide
[00:40:58.900]is being done by our postdoc, Pablo.
[00:41:01.620]The idea behind this is to see what are the different levels
[00:41:04.500]of variability that we have within the field
[00:41:06.280]and in an easy way and a free tool that anyone can access
[00:41:10.300]in order to see the results in productivity
[00:41:13.160]and this is something that we are currently working on.
[00:41:17.500]Another project that we're working is a project
[00:41:21.120]we're trying to estimate soybean protein and oil content.
[00:41:24.420]This is maybe something that will be more important
[00:41:27.100]in the future but basically we have 15 locations.
[00:41:29.680]This is a national project coordinated by Kansas State.
[00:41:32.720]We have 233 farmers fields in 13 states until last year
[00:41:36.760]and we added 15 in Nebraska.
[00:41:39.260]Basically we go to each of these places and we collect
[00:41:43.000]samples, seed samples and soil samples in each of these places
[00:41:47.180]and then that they were selected by a cluster analysis before
[00:41:51.380]and then the idea is to do an analysis and to predict
[00:41:55.940]the protein and oil variability within field scale.
[00:42:00.220]So this is a tool that is coming in the future.
[00:42:04.020]The last one and this is something just a little bit more
[00:42:06.620]fancy for you.
[00:42:07.560]Basically something that everyone measures is a seeding
[00:42:12.840]calculation, how many plants we have and how well distributed
[00:42:16.780]are and this is for corn and this is basically an application
[00:42:20.080]that we are developing to count, sorry that it's in Spanish
[00:42:24.220]but this is a collaboration that we have with a group in Spain.
[00:42:27.280]So basically what you have here is two different ways of
[00:42:30.460]measuring the seeding, the distance among plants.
[00:42:34.860]They're both accurate but this is more accurate because you
[00:42:37.960]can tell that it's identified the right center of the plants
[00:42:40.900]and here there are some mistakes you can see.
[00:42:42.680]There is a difference there.
[00:42:45.580]So basically we measure the difference between the real
[00:42:47.480]distance and what we measure with pixels and basically this
[00:42:51.920]application that can be used in any phone can give you an
[00:42:55.780]estimation of the population and how well or how wrong you
[00:42:58.940]planted your corn.
[00:43:02.280]Okay.
[00:43:05.480]The last one is just a message to say that digital ag and
[00:43:09.800]precision ag and all these tools, you know, we are
[00:43:12.520]-- I have the pleasure to work in Australia, in Argentina and
[00:43:17.780]here in the U.S., but these are also impacting countries in
[00:43:23.280]Africa, for example.
[00:43:24.280]This is a report that was put together where they were
[00:43:26.880]providing digital advisory service through the phones
[00:43:29.420]applications or linking farmers to different markets to get
[00:43:33.280]better prices in their operations or they were putting
[00:43:35.980]digital banks for farmers and this is the impact that it has
[00:43:40.280]in income and productivity.
[00:43:42.360]If we measure this, you are seeing that also in developing
[00:43:46.060]countries, this can have a huge impact.
[00:43:48.900]So summarizing, overall adoption of digital act will remain
[00:43:54.640]low in Nebraska.
[00:43:56.220]I think that using benchmark indicators to show the value
[00:44:00.240]of each of the technologies that we use is important.
[00:44:03.640]There are many dashboards that can be used as a decision
[00:44:06.260]support tool to increase these adoptions.
[00:44:09.260]Each dashboard has a different application and it can
[00:44:12.200]be reached for different target farmers.
[00:44:14.780]And then using data analytics, that was the last portion
[00:44:17.120]of the presentation, can help us to understand what is
[00:44:20.320]the potential of these technologies from the research side of things.
[00:44:26.520]With that, I would like to thank you all, the team.
[00:44:28.760]This is 24 members.
[00:44:31.700]We have Pablo, Julio, and Bruno that are basically helping me to conduct all the experiments
[00:44:37.580]and visiting scholars that were also here in July for the DOM project.
[00:44:42.040]Also, thanks to the Nebraska Research Project, also Glenn for helping always in the field,
[00:44:51.060]and of course, Laila Puntel that gave me the opportunity to start these positions.
[00:44:56.080]We are continuing collaboration with her and also with Pablo to get all this data that
[00:45:01.040]we collected in the last three years into publications that can be used to move forward
[00:45:06.200]all the management.
[00:45:08.040]And of course, funding agencies with all their support.
[00:45:11.880]This will not be possible.
[00:45:13.400]So thank you.
[00:45:21.940]Questions for Dr. Baboha, please.
[00:45:25.640]Thank you.
[00:45:28.100]I appreciate your experience in the CSIRO, that is the largest research organization
[00:45:34.260]in the Southern Hemisphere.
[00:45:36.760]And for the audience, I would like to let you know that the Southern Hemisphere is a
[00:45:39.740]little larger than the Northern Hemisphere.
[00:45:41.720]One thing you mentioned in the beginning of the presentation, that 75% of the farmers
[00:45:49.360]realized, agreed, expressed, and reported that they are not using digital agriculture
[00:45:57.000]technology because of the lack of information.
[00:46:02.500]They don't have information, they don't have information reached out to them.
[00:46:08.840]That's very good.
[00:46:09.840]Do you have any information?
[00:46:11.560]Any information about what kind of sources farmers would like to use or want to use?
[00:46:20.360]Given that I recently had a live strategic planning focus, farmers realized that many
[00:46:27.260]of the times our website is not accessible for them.
[00:46:32.040]They type something, it doesn't come up, it goes to either South Dakota or Iowa State
[00:46:36.460]when they type something.
[00:46:38.480]So what kind of information sources?
[00:46:41.400]You believe in your personal opinion with experience in the southern hemisphere and
[00:46:46.000]here that easily accessible to the farmer and farmer rely on.
[00:46:51.240]Thank you.
[00:46:52.240]Back to you.
[00:46:53.240]Well, thank you now for the question.
[00:46:58.120]We did ask that in the survey related to nitrogen, so I can reply from that.
[00:47:04.140]And one thing that I want to highlight is this was a statewide survey, so it was surveying
[00:47:08.740]farmers that are in connection with UNL.
[00:47:11.240]Farmers that were not.
[00:47:12.800]When you analyze a RAS Confirm research survey that we did, all these metrics are going up
[00:47:17.620]because it's a huge impact.
[00:47:19.940]I don't have a response on what will be the, I would say, the right or what they prefer.
[00:47:26.040]We are running a, now we're finishing the CHG project and we're asking them that question,
[00:47:32.520]what will be the, I would say, the approach that they want.
[00:47:36.580]From my thinking and from my experience these three years here, I think that UNL is still
[00:47:41.080]doing great in terms of extension events, but as you know, we might need to focus more
[00:47:50.980]maybe in those events to show those metrics and to start trying to convince them.
[00:47:58.060]And I think the on-farm research network is doing a great job with that because those
[00:48:01.980]are the ones measuring as well.
[00:48:03.280]So I think that the system is in place, we just need to maybe focus more on which is
[00:48:10.920]the best metric to convince them.
[00:48:12.820]But yeah, I don't have an answer, that's just personal opinion.
[00:48:17.680]It's not a follow-up question, it's just one suggestion that we used to have the app three
[00:48:25.560]or four years back and it was very effective for the farmer.
[00:48:28.920]They download the app, the phone, open, connect everything with the university, whatever they
[00:48:34.160]like.
[00:48:35.160]With this nitrogen as one of the challenging things in Nebraska.
[00:48:40.760]One reason could be have app that's continuously with the evolving technology.
[00:48:48.260]So people look at everything nitrogen research, what is going on from the university.
[00:48:56.000]That could be one option.
[00:48:57.840]That's the solution.
[00:48:58.840]Yeah.
[00:48:59.840]That can be enough.
[00:49:00.840]There are too many apps going around, but yeah, I mean, any, anything that can help
[00:49:04.000]to reduce that gap is welcome, I think.
[00:49:07.460]Thank you, Nat, for the question.
[00:49:08.600]Thanks, Gabriel.
[00:49:09.600]Thanks for the presentation.
[00:49:10.600]I've got a kind of question interested in one of your earlier slides regarding adoption.
[00:49:15.720]I think you had some blue and green bars where you were surveying growers, whether they are
[00:49:21.820]aware of a technology versus whether they use it.
[00:49:28.040]The continuum of innovation also includes this adoption and you had this really nice
[00:49:33.940]chart of complexity as well, where you scale up with complexity and value, you increase
[00:49:40.440]the risk of lack of adoption and you addressed some barriers to adoption that you're considering.
[00:49:46.340]I'm just curious to know if you, through some of that data, if you derived any disadoption
[00:49:51.860]trends.
[00:49:52.860]So did you ask the question, this is, I know about it, I use it.
[00:49:57.880]The other question is, have I used it and stopped using it?
[00:50:01.280]And then, yeah, obviously the question is then why did they stop using it and what can
[00:50:05.180]you learn from that to inform the complexity relationship that you have?
[00:50:10.280]Thank you for the question.
[00:50:13.720]Yeah, we're not asking this question, and that's more associated, and again, I think
[00:50:19.060]that something that was lacking in that survey was more the kind of the social aspect in
[00:50:25.180]many of these things.
[00:50:26.960]And this adoption will be kind of trying to measure something that can be considered negative.
[00:50:31.340]So that was something of the feedback that we got from the sociologist saying that you're
[00:50:34.160]not digging into that specific question that you did.
[00:50:37.000]That is coming next Monday.
[00:50:38.940]We are launching a national survey.
[00:50:40.120]We have 10 different universities, and there is a question regarding did you start using
[00:50:48.400]something and you stopped and why.
[00:50:51.540]We didn't have that in our survey.
[00:50:53.900]That's a great point.
[00:50:55.680]Hi, Guillermo.
[00:50:59.040]With all this, the historic data you're collecting from farmers and continue to collect with
[00:51:04.400]these tools, is there any accession or the variety?
[00:51:09.960]The information being put in and being considered when they're giving these recommendations,
[00:51:14.720]nitrogen recommendations.
[00:51:15.720]I'm thinking of both farmer, like what you're actually recommending to the farmer concerning
[00:51:20.780]what variety they're planting, but also there's a lot of questions in just historic G by E
[00:51:25.800]by M that could be asked with the dataset like that.
[00:51:30.540]Yeah.
[00:51:31.540]Thank you for the question.
[00:51:33.340]Now, there is not that, I mean, we record what is the, what they have been, what is
[00:51:39.800]the genetic that we're using in a specific study, but we are not conducting studies on
[00:51:44.600]that type of, or how, I know that other universities are doing that, you know, studying, you know,
[00:51:51.920]how the, how is the changes in natural use efficiency, for example, or in productivity
[00:51:56.340]against the changes in the genetics that we're using, but we are not specifically doing an
[00:52:01.280]analysis.
[00:52:02.280]So, not you, but with the dataset, that those varieties do exist with some of this data?
[00:52:09.640]Like your cloud source data that they are putting the varieties in, or no?
[00:52:15.060]No.
[00:52:16.060]I mean, in all our experiments, we do that, but in our dashboard, for example, there is
[00:52:19.980]no requirement for that.
[00:52:20.980]There is a requirement, for example, on the growing degree days requirement, but not specifically
[00:52:26.540]on the genetics.
[00:52:27.540]Okay.
[00:52:28.540]Thanks.
[00:52:29.540]Yeah.
[00:52:30.540]Thank you for the presentation.
[00:52:35.340]You mentioned that in terms of adoption.
[00:52:39.480]Farmers age, young, older farmers are more inclined to adopting these digital technologies
[00:52:47.600]in agriculture than younger farmers.
[00:52:51.120]Is there a reason for that?
[00:52:52.920]Was your survey able to throw a light on that?
[00:52:56.600]Thank you.
[00:52:58.220]So the question is, if we identify any reason of that difference, well, not yet.
[00:53:07.220]We don't have, basically.
[00:53:09.320]Any questions asked to identify that?
[00:53:14.940]I can infer and I can tell.
[00:53:17.420]So I mentioned before that we ran this survey three different times.
[00:53:20.860]This is statewide.
[00:53:22.160]And then basically statewide, more than 60% of farmers has more than 55-year-olds.
[00:53:29.000]So the average age is higher.
[00:53:34.820]If we move to what we did in the on-farm research, that drops down to 50%.
[00:53:39.160]To 46%.
[00:53:40.600]And if we compare the mean time of the farmers that are closely working in early adoption
[00:53:45.420]of nitrogen technologies, that drops to 38%.
[00:53:48.860]So there is a trend of, I would say, younger generations.
[00:53:54.240]But still, the amount of younger generations that we have is very small.
[00:53:58.580]We have some open questions saying, why you're not adopting?
[00:54:01.240]And we have an open-- beside all this closed question, there was an open section, why you're
[00:54:05.600]not using this.
[00:54:06.700]And they say, I'm close to retirement.
[00:54:09.000]This is too much for me.
[00:54:10.000]My kids are not anymore in the farm.
[00:54:12.640]So all that aspect.
[00:54:15.140]So the social aspect, I think, is huge and needs to be investigated.
[00:54:22.300]Because the research, I will say, there are many technologies, the approach of using a
[00:54:27.400]sensor to see how the crop is and to deliver an action rate has a lot of logic.
[00:54:32.900]But the social aspect of that is something that needs to be studied.
[00:54:38.840]We are starting some collaborations on that path.
[00:54:43.520]But we are super early with the economic department and some of our professors.
[00:54:51.640]But there's no data yet.
[00:54:53.840]Thank you.
[00:54:55.720]Any more questions?
[00:54:59.220]So we have two questions in the chat.
[00:55:02.880]One was answered real time by you and Dr. Memo.
[00:55:05.840]And the other is from Brazil.
[00:55:08.680]And he wants to know that a point has been made that digital egg can be used for predictive
[00:55:15.680]analysis.
[00:55:16.680]How precise or accurate those predictions are as the season goes by?
[00:55:24.080]And he said, for example, they mostly rely on a preventive approach in their corn and
[00:55:29.800]soybean production in Brazil.
[00:55:32.620]And the second part to the question is how affordable digital agriculture is to growers
[00:55:38.160]in the U.S.?
[00:55:38.520]Okay.
[00:55:39.520]Well, so thank you for the question.
[00:55:42.560]The first part I will say talks about how accurate are those predictions, if I understand
[00:55:48.300]correct.
[00:55:50.880]So basically, I will say one of the maybe it's not a shift in the approach, but usually
[00:55:59.780]in terms of nitrogen management, I think there is some more focus in research in the U.S.
[00:56:08.360]is not into determining specific rate, because it's very hard to determine a specific rate,
[00:56:15.380]but to reduce uncertainty of that prediction.
[00:56:18.280]So the error, basically.
[00:56:19.760]And basically, if you have the correct approach and the correct variables under analysis and
[00:56:25.420]good data, you can basically reduce error in your prediction.
[00:56:29.080]I think the fear is not going into develop something that tells you apply 50, 80, or
[00:56:37.200]100.
[00:56:38.200]You can give a range, and you can have a probability of occurrence.
[00:56:41.460]That is one of the figures that I show.
[00:56:43.540]So if you can simulate and you can use a lot of analysis to generate that probability,
[00:56:48.960]you can give a number with a probability associated to that.
[00:56:53.380]We are more getting used to, I want a number.
[00:56:56.100]But I think that a probability approach is something that-- it was discussed, for example,
[00:57:02.060]in our natural use efficiency workshop a couple of years ago.
[00:57:05.760]The second portion is how accessible it is.
[00:57:08.040]So, well, I don't have a-- I'm not an expert in cost, but, for example, when we are approaching
[00:57:14.100]farmers in these projects to test these technologies, the projects are coming with funding to cover
[00:57:20.420]that aspect.
[00:57:21.420]But, yeah, I don't have a specific analysis or general comment on that.
[00:57:29.280]There's one more question from Dr. Memo.
[00:57:32.980]You shared that lack of information and know-how are one barrier to adoption.
[00:57:37.880]How can we be effective in translating these tools to end users?
[00:57:43.320]What partnership and incentives may be needed?
[00:57:45.820]Well, I think the university has-- I mean, this is great information for the university
[00:57:49.900]because our graduates need to focus on how to use this tool, how to use satellite images,
[00:57:57.440]how to process information that is freely available if you want.
[00:58:00.800]I mean, there are companies that sell you better images for resolution, but you have
[00:58:04.720]the technology.
[00:58:05.720]So we can improve our images.
[00:58:07.720]Our training for undergraduates and graduates, we can also have specific training for certificates,
[00:58:14.080]I will say, for people to manage this technology or to test this technology.
[00:58:22.000]I think it's not only about what the people that we are educating at the university, but
[00:58:27.920]also having these type of certificates to educate the driver.
[00:58:32.060]Today, 20 years ago, somebody that was driving a tractor, today you are getting to a tractor
[00:58:37.560]is basically you have three or four computers.
[00:58:39.960]So you need to be able to manage that and to know.
[00:58:44.260]I would say certificates for different stakeholders, but also continue growing the programs, the
[00:58:53.300]Digital Act program at UNL, the Precision Act program at UNL, with basically positions
[00:58:59.760]and funding to support this, because Digital Act can be applied to any of the areas of
[00:59:06.400]the people that are here.
[00:59:07.400]To horticulture, to crop production, to cattle production.
[00:59:13.720]So it's something that can be used in all the areas of the missions of the university.
[00:59:18.940]Thanks for the answer and great presentation.
[00:59:22.780]It's almost time.
[00:59:24.220]Please join me once again thanking our speaker.
[00:59:33.760]Thank you.
[00:59:33.800]Thank you.

The screen size you are trying to search captions on is too small!

You can always jump over to MediaHub and check it out there.

Comments

0 Comments

Assessing and Advancing Digital Agriculture in Nebraska

Description

Searchable Transcript

Comments icon comment

Related Channels

Comments