Transcript:

 

CHRIS CHAMBERS 0:00
My name is Chris chambers. I’m a Client Support Manager at METER Group and the TEROS product manager.

Jeff Ritter 0:06
My name is Jeff Ritter. I’m the product manager for our plant canopy and atmospheric monitoring instrumentation.

CHRIS CHAMBERS 0:14
So we’re here to answer a few questions today. Hopefully we can help out. And if you have any future questions, be sure to send them to us. Okay, so our first question is kind of a kind of a transmission option. Question. Can ZENTRA data loggers transmit on a LoRaWAN network? Or when will this be available?

Jeff Ritter 0:39
Yeah, so our data loggers are currently transmit over a cellular network. So they’re not currently enabled for LoRaWAN. We’re working on some other technology to to expand that those capabilities over over WiFi, or maybe even in the future over a radio mesh network, but and there’s

CHRIS CHAMBERS 1:01
a lot of ins and outs and what have yous with the way you deliver data. Right? And generally, we our philosophy is, you know, however, you can get it to the cloud. But LoRaWAN does have some data limitations. It’s generally short snippets of data that you’re bursting, right? So maybe it’s something we’ll have for the future. Right now. It’s not one of our one of our priorities.

Jeff Ritter 1:29
Right. Especially when you’re talking about really data rich instrumentation like a weather station, and ATMOS 41. Exactly. And ATMOS 41. The more sensors you add on there, the more limitations you run into with with the low data networks.

CHRIS CHAMBERS 1:43
All right. Okay, how about installing power sensors? Do you had? Do you have any suggestions for power sensor installations?

Jeff Ritter 1:53
You know, the biggest thing with with any sort of sensor that’s measuring radiation is to get it level. incident radiation is going to be hugely impacted by the the angle of incident light. So if you don’t have horizontal to the surface, you’re not going to be measuring accurately.

CHRIS CHAMBERS 2:10
And it just says, Do you have a general rule? Like how far like what’s what’s good enough? As far as leveling?

Jeff Ritter 2:17
It typically within three degrees of dead level is what we would suggest for those. But then again, you know, closer to zero is better, more accuracy,

CHRIS CHAMBERS 2:27
More level is better. That’s right. Okay, moving on, how high? Can a chain link and close exclosure be? And I think, the better question, or maybe the more relevant question is, how high should a chain link exclosure be, I’m gonna I’m gonna just inserts. Sure top down process,

Jeff Ritter 2:52
there’s, there’s a couple of things to consider. What what is the exclosure trying to exclude, you know, if you are trying to exclude deer, for example, they can potentially jump a six foot fence. So if it’s for cattle, they’re probably not going to be able to get over a six foot fence. But the height of that exclosure can potentially also impact your weather station. So you need to keep in mind what your application is for. If you were trying to adhere to World Meteorological Organization standards. It weather shape is gonna have to be pretty high compared to any obstructions that are around there. So you need to keep keep that in mind what you’re trying to keep out, as well as what you’re trying to measure.

CHRIS CHAMBERS 3:40
What purpose how high as a general rule, say if it’s an ATMOS 41, how high above the exclosure Do you think it should be?

Jeff Ritter 3:50
That’s a good question. Typically, you want it if you’re measuring ET, you want two meters above the canopy, so keep that in mind with with your exclosure how high that needs to be. So it’s not impacting what you’re measuring there. If it’s a WMO station, I think the recommendation is 10 times the height of the nearby obstruction.

Oh, goodness. That’s a pretty high if you’ve got to need a big chain link fence. Let’s skip over to sensor recalibrations. As a general rule, and let’s let’s focus on ATMOS 41s. Since that’s what we do. How often should you recalibrate sensors?

Yeah, so there’s there’s recalibration and there’s maintenance, right. So if we’re talking about just recalibration, the sensors that do need to be calibrated should be probably every two years. Okay, we typically recommend every two to three years. And those are the sensors that that can drift. So specifically something like a relative humidity sensor, or the pyranometer. Those can drift over time. So to make sure that those are within the published accuracy spec we say every two to three years for those. Now maintenance a little bit different, where that’s going to be site specific. If you’ve got this station out somewhere where there’s a lot of detritus from the atmosphere, a lot of leaves, or if there’s a lot of bird activity, and you have a lot of bird fouling on the station that can prevent precipitation flow, or it can you can foul your pyranometer and then degrade that measurement. So maintenance should potentially be at least yearly. But depending on the depending on the site, you know, every couple of months even.

CHRIS CHAMBERS 5:32
And nice. So if you suspect your data’s is isn’t great, you might need calibration. But it’s not two years yet. Do we have any, do we have any tests or anything that we can, that we can recommend client do to check on it or standards?

Jeff Ritter 5:46
Absolutely. And again, there’s there’s a lot of measurements being made on the weather station. So each one is going to be be unique each verification, the biggest thing is getting out doing a site visit and making sure that it’s clean and is working properly. But we do have methods for verifying the accuracy of your precipitation gauge, for example, and the other other sensors on there as well,

CHRIS CHAMBERS 6:11
Right. And there’s always that bird that wants to just perch on the funnel. And poop right down in there.

Jeff Ritter 6:16
If a bird has made a nest in the funnel, you’re gonna have skewed data. So bird spikes. Use your bird spike kit.

CHRIS CHAMBERS 6:21
Here’s a tricky one, if I have no other option, other than mounting my wind sensor on a roof, what do I have to consider in particular to get acceptable data?

Jeff Ritter 6:33
Well, I mean, in general, you’re not going to be getting the same wind speed as you would on the ground. And so you just have to accept that there’s not a lot you can do. If you’re mounting this way up in the air, the the environment is very different up there can be some turbulence and that you wouldn’t normally see in a field or something like that. That’s right, the conditions on the ground are going to be very different. And so if that’s your only option, you know, it’s better than the not having those data, but you just have to live with that, that error.

CHRIS CHAMBERS 7:07
But get it I assume get it up a certain height above the roof, probably on the peak, and…

Jeff Ritter 7:13
And make sure that there’s not any obstructions around, you know, on top of a roof, it’s very easy to have things in the way that you might not think impact the wind speed because you think it’s far enough away.

CHRIS CHAMBERS 7:23
But chimneys or HVACs.

Jeff Ritter 7:25
That’s right, all that stuff can can play a big role in what you’re seeing. So again, even if you add on top of a roof, you want it several meters above the above the rooftop,

CHRIS CHAMBERS 7:34
Right. So the question from our live studio audience is to take, should we take a reference reading, like on the ground or something like that to compare to the roof?

Jeff Ritter 7:45
I think that’s a good idea to get a general sense. I don’t think with the variability of wind speed, I don’t think you’ll ever be able to say calculate correction factor doing that, I don’t think you’ll get that level of consistency and precision. You have to have it there all the time. That’s right. And in that case, you know, they’re saying that’s not an option in this case. So it’s not a bad test, though, to if you’re measuring wind speed on the ground, and you you’re measuring it over the course of a couple of days, and then compare it to what looks like on the roof, you can at least say you know, we’re at least this far off.

CHRIS CHAMBERS 8:16
Right. Right. And it probably depends on the questions you’re asking as well. Absolutely. Okay, how do you level an anemometer? And each anemometer is going to be slightly different. The ATMOS 41 has a couple of options, doesn’t it?

Jeff Ritter 8:32
That’s right. Yeah. So the easiest way and any high quality instrument is going to have some sort of built in way to level it because of how important that is, especially if you’re using you know something that that expects you to be horizontal. A cup anemometer or a two dimensional sonic anemometer like what’s on the ATMOS 41, the expectation is that your level. So you can introduce errors if you’re not. So that was 41. There’s a built in bubble level. There’s also an accelerometer built into the instrument. So then you can check all the time you’ve got a data readout exactly of how level you are. And so you can actually see when your instrument has fallen over or just tilted out of spec, which would you prefer? Personally, when I’m installing it, the bubble level works the best just right there. But then I also keep a readout typically in my remote data of what the accelerometer is outputting. So I can be alerted anytime that the station might tilt outside that specs I can flag those data for, for exclusion.

CHRIS CHAMBERS 9:32
Great. Oh, the ever classic site selection when faced with a less than optimum location, like, say, urban settings, is it better to cite the anemometer five to six feet above a rooftop peak or at the top of a five to six foot ground pole? We’ve addressed this a little bit in a previous question about the about roof mounting. And this really depends on the questions that you want to ask.

Jeff Ritter 9:59
It dependson what you’re wanting to look at. So it’s hard to say in general, but if you’re in an urban setting, that typically means you’re going to be looking at some sort of impact on human comfort.

CHRIS CHAMBERS 10:13
Or ecosystem services possibly.

Jeff Ritter 10:15
Right. And so I would lend more towards having it closer to the ground. Because that’s that’s the microclimate that you’re interested in. But again, it’s going to-

CHRIS CHAMBERS 10:26
sometimes the roof is the best you can do.

Jeff Ritter 10:28
Yep. Absolutely.

CHRIS CHAMBERS 10:30
The next question, kind of, kind of harks back to our exclosure. Question previously, and crop production inside cages, presumably to protect them from from browsers or birds or something. How would you position, how would you position the device to minimize the interference from that?

Jeff Ritter 10:57
Right, and I think as far as like evapotranspiration is concerned. So the thing to keep in mind when you are measuring evapo transpiration, or in this case, when you’re modeling evapotranspiration, and that’s what we’re doing with with most weather stations, right? It’s not a direct measurement of evapotranspiration, but it’s a modeled value. So you’re already are simplifying fairly complex process, right? The specific implementation of that model, what we call FAO 56, in the way that ATMOS 41 does it is a further simplification of that model. So if you were to take that and you start playing with some of the other parameters, like putting this under your canopy or under a cage, you’re going to introduce more error into that measurement. So one of the beauties of that model, evapotranspiration measurement is that it’s actually a model of the reference or the potential evapotranspiration, okay, not directly what that plant is doing right next to it, right. So in order to understand what your plant is, is doing as far as evapotranspiration goes, you need to understand a little bit about your plant and how its interacting with the environment. And that’s done through what’s called a crop coefficient. So if you can get a sense of what the crop coefficient is for your plant in cages or out of cages, however, you can you can apply that to your reference evapotranspiration value, and then adjust it that way.

CHRIS CHAMBERS 12:24
And would it be, would it be just as well to use like a reference, like not that crop a reference crop nearby? Like, if you’ve got a grassy area or, or another crop that’s not covered? Would that be helpful at all or…

Jeff Ritter 12:39
Using the crop coefficient from it depends crop coefficients can be very dynamic from one crop to another, but even over the course of a growing season, they can change. So you do need to have some information about your specific crop and how it grows. And in that case, you would want your your weather station out, as we would typically recommend out in a clearing.

CHRIS CHAMBERS 13:01
Gotcha. How to correct the air temperature measurement. And let’s knock out the… How do you correct the air temperature measurement due to solar heating from sunshine?

Jeff Ritter 13:14
Right? So this is a good question in that the ATMOS 41 does this automatically already. Because anything you put out to measure temperature is going to be impacted also by the solar radiation load.

CHRIS CHAMBERS 13:30
That’s why we see radiation shields on passive devices.

Jeff Ritter 13:33
That’s right. So you need to have a measurement not only of the solar radiation, but also how much heat is then lost back to the environment through cooling. So you need a measurement of the wind speed, right. So if you measure both solar radiation, and you understand a bit about the solar absorbtivity of your temperature sensor, and you’re measuring wind speed, you can actually come up with a correction factor to correct your air temperature. So that’s what we do on the ATMOS 41. And what we actually see is, it is closer to what we would call it an aspirated shield in a air temperature sensor that has a air current flowing over it.

CHRIS CHAMBERS 14:07
Which is typically the gold standard for air temperature measurements.

Jeff Ritter 14:10
Right. It is closer to that than say a passively radiated, a passively shielded sensor. Because those those are also suffering from solar radiation causing temperature increase and the ATMOS 41 measures everything that you need to do the correction. So that’s right. That’s all built in right into the data output already.

CHRIS CHAMBERS 14:30
So this is kind of crop specific. Our next question is crop specific. I’m not a banana crop, agronomy expert. So maybe we’ll just stick to the biophysics of this question here. If you’re very limited in space, and the only area available to install the station will have a 30 meter diameter, canopy around it. But it’s an obstruction on obstruct Is that going to be enough to give you good data? For your say you’re doing? E- a reference E T measurement?

Jeff Ritter 15:10
It depends on I don’t know the height of a banana plant. But yeah, if you can get it high enough, it should be able to get two meters above the canopy.

CHRIS CHAMBERS 15:19
What’s the minimum that we recommend have like, a space like a fetch for or an area of influence for the ATMOS 41?

Jeff Ritter 15:29
Typically, WMO standards, say 100 meters away from from obstructions, okay. But again, a lot of ET measurements, we aren’t necessarily adhering to WMO standards in the first place. That’s just kind of best practices if you can. So it trying to avoid the obstruction as much as possible. And even if it’s not a perfect measurement, it’s better than just guesswork.

CHRIS CHAMBERS 15:55
Okay. So just to boil it down 100 meters would be ideal, but 30 meters is probably enough to get you some some good information.

Jeff Ritter 16:05
I would imagine so without knowing specifics of bananas.

CHRIS CHAMBERS 16:09
Okay, great. Can the ATMOS 41 fail to record rainfall data? Anything can fail to record rainfall data.

Jeff Ritter 16:17
Yeah, and there are two main methods that it can cause problems there. Aside from some sort of a catastrophic failure, like your weather stations struck by lightning and circuitry is all fried. The main modes that it can cause, really, almost any weather station to fail to record weather- precipitation data are if, if the pathways clogged, you know, if a bird has built a nest in your funnel, right, you’re not getting rain through there to measure in the first place. But also, there are limitations on on if the station isn’t fully level, you know, if rain isn’t getting into the funnel, or or if if rain is isn’t flowing to the correct pathway because the station is off-level.

CHRIS CHAMBERS 17:03
This isn’t this isn’t unique to the ATMOS 41. If it’s missing the drip counter, or if it’s missing the bucket.

Jeff Ritter 17:08
Yep, I think half of the water flows can can impact.

CHRIS CHAMBERS 17:12
That’s right. And here’s the tricky thing. Because rainfall, precipitation, in general is crazily, maybe not stochastic, but it’s highly variable spatially, right. And so, to, you know, just as an anecdote, you know, we had a softball game on one end of town, and it dumped on us here, the softball field was only a mile away, right? And they hadn’t had a single drop. So here I am just taking it easy, assuming the games gonna get canceled, and they were gonna play and I was late to the game. Everyone was mad at me. But…

Jeff Ritter 17:51
And we see that a lot with people who, who see a rainfall event and you want to compare one station to another one that’s miles away and ranges is not that homogenous and how it falls over an area.

CHRIS CHAMBERS 18:03
Even over a couple hundred meters we can get large differences.

Jeff Ritter 18:05
We can get huge differences in that rainfall event.

CHRIS CHAMBERS 18:09
So but we do have a method to double check a rain gauge, right?

Jeff Ritter 18:13
Yeah. And it’s something that I suggest people to do all the time is check your rainfall gauge, we’ve got a published method for how to do that. It’s fairly simple. You just essentially pour unknown amount of water through. And that’s a great way to have confidence in your measurements.

CHRIS CHAMBERS 18:30
Will ZENTRA cloud eventually facilitate, facilitate additional metadata. Yes.

Jeff Ritter 18:37
Sure hope so.

CHRIS CHAMBERS 18:38
It has to someday. It’s on the list.

Jeff Ritter 18:40
Yep.

CHRIS CHAMBERS 18:43
Which sensors, and where should they be installed to complement hydric measurements at tree level?

Jeff Ritter 18:50
Yeah, so there’s, you’ve got to understand that there is a continuum of how that that plant exists from the soil all the way up into the atmosphere, right. And so you want to make sure that you are able to monitor not only what’s happening in the atmosphere around that plant, but also what’s happening in the soil. So understanding the water content of the soil with soil water, potential sensors, or water content sensors is important and for where to install those, you want to have them at at the root zone. So you need to understand how deep your your plant, or, if it’s a tree, how deep do the roots of that tree grow?

CHRIS CHAMBERS 19:32
Now, what about something like a vapor pressure deficit? And suppose you’re doing a sap flux study? That’s going to be one of the large- one of the major variables if your for your for your evapo transpiration. Any thoughts on that?

Jeff Ritter 19:51
Yeah, so you can get vapor pressure deficit directly from the ATMOS 41 Where it’s measuring vapor pressure and temperature. If you want to get vapor pressure deficit specific to the leaf, you’ll need to measure leaf temperature. So you can do that with something like an infrared thermometer, measuring canopy temperature, and you can get VPD that’s specific to the leaf itself.

CHRIS CHAMBERS 20:12
Great. Any tricks to keep stations running when they’re experienced power, power drops? Let’s take a moment on this one. Because power management of data loggers, you’ve got a ZL6 Data Logger, self contained solar, rechargeable batteries. And, you know, you’ve got a bunch of sensors on it. And what what really throws a wrench in the works frequently is cellular communications, right? Any thoughts on when you should watch out or take steps? Or there’s got to be some balance of how much power is being used by the cell modem versus how much light you can harvest? Especially if you’ve got your canopy your solar panel shaded.

Jeff Ritter 21:01
Yeah, you’re right. So the first thing and the easiest thing is to just make sure you’re in an area that’s got good solar exposure. If you do that you should be okay. If that’s not possible, you’re right that the cellular communications is the biggest power draw, on most most loggers, it takes a lot of power to communicate. So making sure that you have a an area with a good connection. And you can test that before you finalize your installation. Right. And potentially using some sort of an antenna on there to, to boost that signal. That can be a big help to make sure that you’re not spending excess power trying to send data remotely.

CHRIS CHAMBERS 21:49
And if you are experiencing power problems, contact support, we’re here to help.

Jeff Ritter 21:55
That’s right.

CHRIS CHAMBERS 21:57
Does ET does reference at work in ZENTRA cloud now? Yes, yes, it does. What kind of care must be taken into his installed station on a glacier surface?

Jeff Ritter 22:10
Well, first off, you’ve got to realize the limitations of your of your weather station. You know, if you’re using a low power weather station, it probably does not have an onboard heater, because then that’s a big power hog. So if you don’t have an onboard heater, and you’re putting this someplace where conditions are consistently below freezing, some of the measurements aren’t going to work. Precipitation, precipitation, if you get ice buildup on some of the sensors that can that can include your measurement measurement pathway.

CHRIS CHAMBERS 22:39
Also, your glacier surface is always moving.

Jeff Ritter 22:41
Your glacier surfaces always moving. So you need to make sure that this is this is secured in a way that you are comfortable with not falling over.

CHRIS CHAMBERS 22:50
You might want to get some alerts for can we set up alerts for say the onboard accelerometer so that if it gets out of level gives us…

Jeff Ritter 22:59
Yup. You can set up alerts so that you get a text message or an email. If your station ever tilts.

CHRIS CHAMBERS 23:06
Assuming you have cell coverage in cloud access.

Jeff Ritter 23:09
That’s right, good thing to consider is that this might be a smaller issue. But saturating vapor pressure as calculated over ice actually uses different coefficients in the Tetens formula than it does over water. And so you might need to do some some post hoc data calculations to correct for that, but…

CHRIS CHAMBERS 23:28
Cool. Is the bug in your API solved? Yes. Let’s journey into our virtual weather portion of today’s broadcast. So come with us gentle listener. And first off with it. What do you think about virtual weather? First of all, what is it?

Jeff Ritter 23:50
Sure. So it’s a virtual weather is, you know, we were talking about models a little bit earlier. And virtual weather is essentially taking broad scale weather modeling and and reducing it down to a grid level. So you can get essentially predicted weather in your location using some some pretty fancy statistics stacked on top of those models. So it’s, you know, it’s really powerful. It’s great for looking at, at broad scale questions, if you don’t need great resolution, or extremely high accuracy at those resolutions. And like anything, it depends on what question you’re, you’re asking to see how useful it’s going to be. I’m a big fan of, of using satellite and remote sensing applications using model data when you can, as long as you understand the limitations and how important it is to so ground truth those.

CHRIS CHAMBERS 24:42
So let me get this straight. Say you’ve got a nearby regional weather station that’s like five miles away that you’re getting precipitation from for your farm. The idea is that you’re kind of you’re trying to find some way to estimate whether your your specific location is always like three millimeters behind that one or half a millimeter, or you get more rain in your specific locale…

Jeff Ritter 25:11
essentially, yeah, it’s trying to create grids at those scales from, from stations or farther away, or just from overall weather mapping models.

CHRIS CHAMBERS 25:24
How well does it work?

Jeff Ritter 25:27
You know, I’ve seen some that say, on certain days, they’re within two degrees, sometimes I’ve seen them as far off as four or five degrees, as far as temperature goes. So, you know, for the, for the scales that a lot of people are looking at, that’s pretty good.

CHRIS CHAMBERS 25:42
But you’re talking about averages, then Right?

Jeff Ritter 25:45
Right, so, so once you start getting down to really complex topography, or if you get into applications where it really matters, you know, even a couple degrees off can can cause huge implications. If you are, if you are grower, and you’re trying to look at weather inversions, or colder drainage, you know, it’s not going to work for that,

CHRIS CHAMBERS 26:04
Or if your max is actually five degrees off your vapor pressure deficit is going to be, you’re not going to capture that in the average.

Jeff Ritter 26:10
That’s right, so we’ve got a weather station here at our building that’s kind of at the top of this little hill, and I’ve got one down, you know, just a few 100 meters away, and the bottom of this hill near near a little river, that we get some cold air drainage down there that, you know, even into the summer, I was seeing freezing temperatures down there, that would not be captured in in a virtual weather forecast. And that can impact you know, not only the fauna that wants to move through there, but you know, the sort of plants that are going to grow there and, and how they grow there. How happy they’re going to be it’s going to impact pest management, disease management, etc.

CHRIS CHAMBERS 26:49
Right. Right. And finally, for the segment, current state of the science. Do we think virtual weather will improve substantially? Or will we always need local weather stations for accurate forecasts?

Jeff Ritter 27:07
Yes to both. I think that absolutely. You know, models are continuously improving. The more data we get, the more computational resources we have devoted to those, those are going to improve, but I don’t foresee ever having the sort of coverage that we would need for for complex landscapes, especially with how dynamic they are to not have, you know, measuring the actual microclimate that you’re interested in.

CHRIS CHAMBERS 27:11
And having the local the local weather stations, you can actually use the data to train your virtual, virtual weather station.

So it does matter what you install your ATMOS 41 over right, the roughness of the surface, say if you’ve got a canopy, you might see more more if you’ve got a rough surface, you’ll probably see more turbulence right? You’ll probably see some different canopy, some different dynamics at it. You have any thoughts?

Jeff Ritter 28:15
You know, in general, it’s gonna depend on what you’re you’re asking if you are wanting some measurement of you know, the model values we’ve we’ve spoken about earlier, evapotranspiration, typically want it higher above your canopy than that, because of those, you know, because of the impact that the surface can have. So I would say up to two meters,

CHRIS CHAMBERS 28:36
Minimum two meters…

Jeff Ritter 28:37
Minimum of two meters typically, is what you want. Those are that’s what the evapotranspiration model actually uses. And you’ve got to correct the model if you use something other than two meters.

But in general, it comes to the question that you’re asking, you know, if you want a measurement in the canopy or close to the canopy, then you know, that’s where you should put it and get your sensor down there. That’s right.

CHRIS CHAMBERS 29:01
The WMO and NWS have standards for the location and distances for weather stations and their sensor locations, right? Are there any standards for METEONET or microclimate stations?

Jeff Ritter 29:15
There are the one that specifically comes to mind was published back in 2019 by the American Association of State Climatologists specific to exactly this citing standards for musinet weather stations now a lot of that refers back to WMO guidelines. So if you are already using WMO, that should be fine for for musinet scale work, but there are specific exceptions for the musinet approach that that are laid out in there. So yeah, if you look for that, for the American Association of State Climatologist there’s a good document on that.

CHRIS CHAMBERS 29:57
Now some of these like the WMO specifically said As that you have to keep your wind speed sensor, like so far away from your station, right? But all that kind of breaks down with, with all in one stations that weren’t even around when the standards were written. Right, right. So how do you feel about people looking for WMO standards where the technology has kind of gotten ahead of the standard?

Jeff Ritter 30:25
Well, it’s great to have as a reference, you know, to look back on these stations that do adhere to that and say, Okay, we know how good those measurements are. So we can approach that with the stations that are currently slightly bending those standards a little bit in some way. And we can use that to, to make an argument that potentially in the, in the future, you know, we can modify those those standards. And as long as we have the data to back that up, I think that’s a good way to approach it.

CHRIS CHAMBERS 30:53
Would it be possible through modeling to obtain values of physical parameters such as temperature and humidity, from inside the canopy using data from a station outside the canopy 300 meters away?

Jeff Ritter 31:04
Yeah, yeah, there are. For sure biophysical models that you can use to to model for example, wind speed inside the canopy, if you’re measuring it above. I’m not as familiar with with looking at temperature or humidity models, though I’m sure that they exist.

CHRIS CHAMBERS 31:24
Probably depends on the canopy structure.

Jeff Ritter 31:26
Probably Yeah, you will, you’ll definitely need to know something about the canopy architecture around the leaf area index is going to play in there to know how dense that canopy is. But you know, if you’ve got the right parameters, you can you can model just about anything.

CHRIS CHAMBERS 31:40
Since ETis driven largely by radiation, is ET prediction closer using virtual data?

Jeff Ritter 31:47
No. So we were talking earlier about how ET is already modeled by these weather stations. And one of the what are the variables and that we’re actually measuring for ET? Solar radiation, wind speed, relative humidity and air temperature, net radiation? Right? net radiation, right. So it’s right, yeah, that’s a good point where the ET model is using net radiation, which we are already assuming a value of that from solar radiation in shortwave. So if we then start breaking that down even further and not using measured parameters, but using them taken from some other other modeled source, you know, you’re going to introduce more error than if you’re just measuring it locally, right?

CHRIS CHAMBERS 32:34
And it is such a driver of ET. In the system that you…

Jeff Ritter 32:34
That’s the one you really want to nail

CHRIS CHAMBERS 32:41
For an outside for an outside station used to estimate reference ET, do you get concerned about the local site dryness, if different from a well watered grass, etc, on the ET reference estimates?

Jeff Ritter 32:54
Yeah, that needs to be taken into account. What some further context for this is for reference, evapotranspiration, that for that model, it’s assuming a closed canopy, and a well watered crop, right? So if your crop is not those, you know, we already expect that reference ET is likely higher than what your plant is actually seeing. You take this these things into account.

CHRIS CHAMBERS 33:18
And since we’re not actually measuring the the eddy covariance, then violations in the assumption are going to are going to really undermine the data.

Jeff Ritter 33:26
That’s right. You can take that into account however, and a lot of people will, you know, look at ET and know how much they need to adjust it again, using like a crop coefficient, you can you can apply.

CHRIS CHAMBERS 33:39
These are always fun, what is the best weather station density for 100,000 hectares? 100,000 or a million?

Jeff Ritter 33:47
You know, honestly, it doesn’t matter if it’s 100,000 or a million or even one hectare, the my answer is always gonna be the same. And that I can’t tell you, it’s gonna be entirely application specific to where you’re putting it. If I was to put this out on my grandparents farm in rural Minnesota, I could probably put one of them on their 2000 acre farm. And that probably describes the weather there pretty well.

CHRIS CHAMBERS 34:12
But it alsocomes back to just good sampling procedures, right? You need a representative, you need to be representative of what you’re sampling, you need to have a large enough sample size to capture the variability. So…

Jeff Ritter 34:25
But but even that the the area of measurement of one of these stations is also dependent on the topography around and on the on the wet weather conditions. So Right. You know, if you’ve got a very complex topography, you want to be capturing that you want to measure all those different microclimates. So it really comes down to what are the prevailing weather conditions and and the topography of your site

CHRIS CHAMBERS 34:51
And what level of detail you want. That’s right. Let’s spend a minute on metadata because it’s easy to overlook. On the importance of metadata, and just for an example of some metadata, that it’s important, it’s a little, it’s a little off topic, but definitely related for soil moisture sensors, right? If you’re looking at volumetric water content, you really can’t interpret those data unless you have some soil type. So that’s one example where metadata is extremely important. And so the same things apply for weather stations. We’ve already talked about crop co- coefficients a little bit. So the crop that you’ve got one of them metadata should people collect specific to weather stations? And what are what do you see are some of the main main things that get left behind that that impact your data?

Jeff Ritter 35:20
So one of the first things that comes to mind is having some sort of qualitative approach to, to looking at especially in this case, an orchard, whether that’s some sort of having a PhenoCam where you can draw information from, or some other way to look at the phase of what your your orchard is. And so having that I think, is really valuable, too, in order to tie it to certain phases.

CHRIS CHAMBERS 36:30
will probably slow aspect.

Jeff Ritter 36:34
That’s right, getting back to you know, having complex topography. It’s one thing to just plunk your weather station down. Where do you think the topography is interesting to actually quantify that and be able to have those as reference points? So you can say that I have this here, and it’s, you know, facing this aspect of the hill, you know, that’s going to impact that? Absolutely. Right.

CHRIS CHAMBERS 36:56
So can we say with enough confidence that local measurements are most relevant for modeling pests, diseases, etc? And this probably depends on the model that you’re using. For for pest risk, right?

Jeff Ritter 37:12
I would say generally, yes, but you’re right, it depends on the model that you’re using and what parameters specifically you’re looking at. But for the most part, you know, the ones that are very important in pest management and disease management are going to be air temperature, relative humidity, relative humidity, and those those sorts of things are definitely better with from local measurements. Absolutely. Right.

CHRIS CHAMBERS 37:34
And, microclimate is going to play a big role in this as as well as the overall weather patterns.

Jeff Ritter 37:40
That’s right. Yeah, microclimates are predominantly what matters to, you know, living creatures in in those places, pests included,

CHRIS CHAMBERS 37:49
and in that case, you’re going to be more interested in what’s happening, like right down in your canopy, than you are two meters above the canopy. Yeah, those

Jeff Ritter 37:56
pests don’t care what’s happening above the canopy, they care about what’s happening right there, where they’re where they’re eating your crop.

CHRIS CHAMBERS 38:02
So let’s, let’s pause and just look at the location of your station for a moment. And this is pretty specific. Is it better to leave a station inside a fruit orchard for more reliable growing degree days? Or evapo transpiration data?

Jeff Ritter 38:17
No. So typically with with the model, evapotranspiration, there’s other models, there are like we’ve discussed some assumptions involved. And so you want those out seeing the total solar radiation that that’s incoming, otherwise, you’re going to introduce extra error to those

CHRIS CHAMBERS 38:36
and your evapotranspiration needs to be positioned over a reference crop.

Speaker 1 38:41
So if you are measuring, if you’re able to measure evapotranspiration, directly through, say, an Eddy Covariance setup, then yeah, you want those in there in your crop. But measuring evapotranspiration or getting modeled values of evapotranspiration, you definitely want to adhere to those recommendations.

CHRIS CHAMBERS 38:56
And it depends on your model as well. You know, if you need to have the microclimate, then it’s definitely going to be best to be right down in your canopy. That’s right. If you’re looking at larger landscape processes, and you’re probably going to be up and away from it.

Jeff Ritter 39:10
That’s right. Very application specific for your research.

CHRIS CHAMBERS 39:15
So know what questions you want to ask, what is the advantage besides using just one port of the all in one ATMOS 41 versus separate sensors like the 22 plus the 14 the pyranometer ECRN. And especially for ET assessment.

Speaker 1 39:30
Right. So, I mean, you hit on one and a lot of those play into that same, same reason using one port is is for convenience. And all in one sensor. You can set those up in minutes versus versus when you’re setting all of those up that at that time adds up. And we have the data management is a big, big deal when you are pulling data from multiple sensors and you are dealing with it that way. Also having all those sensors co located allows you to, when you are measuring ET and you say that these are all two meters above the canopy. They’re all right there, you don’t have to worry about any of them getting in the way of each other having them all air temperature correction, your temperature is corrected off of wind speed, that’s the exact same location as the as the air temperature sensor, right. So, you know, it allows for a level of convenience and data management throughput that you don’t get when you have your separate sensors,

CHRIS CHAMBERS 40:34
right. Could the Tama weather station provide ET naught? Yes, yes, it can. It seems that all in one systems like the ATMOS 41 can be teamed with ancillary power heated rain gauges or unpowered chemical snow melting gauges, right? Suppose you want total precipitation out there. But that is a separate sensor. The ATMOS 41 currently doesn’t have heating. It’s just a it’s just liquid precipitation, right? Just measures rain. Yep. So yeah, there could be other instruments used with the ATMOS 41 to get you that that snowfall precipitation or the total precipitation?

Jeff Ritter 41:20
Absolutely. For for precipitation, you still could run into a limitation if your pyranometers covered on the ATMOS 41. Right. But if that freezing weather precipitation is important, then you will need to pair it with something else that has a heater on board.

CHRIS CHAMBERS 41:35
Right. So our next question is about is kind of a precision irrigation question. And can you for like high value crops? Can you a couple of weather monitoring with soil moisture monitoring to backcheck evapotranspiration. And there’s some quite sophisticated techniques to do that. Particularly with the TEROS 21 and TEROS 12 together where you’re measuring matric potential and water content, you can set your set points with a TEROS 21 matric potential, you know, with a TEROS 12 what your soil storage is. And so then it’s not a very large jump to monitor your evapo transpiration and just refill the soil with what you’ve what you’ve transpired over the previous day. Right? What What does someone need to watch out for for that? What are some pitfalls they might run into?

Jeff Ritter 42:34
So the- I mean, that’s a great way to do it, if you’re measuring the pool of water in the soil, and how available that water is, you know, that’s a great way to, to look at the water loss through the system. The biggest pitfall is that you’re only looking at one very discreet location. So if you had a million soil sensors out over your field, you know, you could you could monitor that. And so this is kind of where we get back to that question of scale. You know, are you using a satellite? Are you using a weather station are using ground sensors,

CHRIS CHAMBERS 43:09
A million sensors, I don’t think you need a million sensors.

Jeff Ritter 43:10
You probably don’t need a million.

CHRIS CHAMBERS 43:12
I think you could do it with less

Jeff Ritter 43:14
You can do it- you can do with less, but there is a lot of heterogeneity in capturing may be able to integrate that into a more usable signal for your irrigation system. And so back checking that with soil sensors is is a great way to see what’s actually in the soil at that moment. And then you can use ATMOS 41 modelled ET for that integrated signal,

CHRIS CHAMBERS 43:44
The soil moisture sensors can corroborate your your readings.

Brad Newbold 43:44
All right. Well, that’s gonna wrap it up for us. Thank you again for joining us today. We hope you enjoyed this discussion. And thank you again for everybody who has sent in such great questions. Also, if you have any questions that we didn’t get to please contact us via our website at METER group.com. Finally, you can subscribe to the METER Group YouTube channel and accept notifications to see previous episodes of office hours and to get notified when future videos are available. So thanks again. Stay safe, and have a great day.