In this 30-minute webinar, METER plant, canopy, and atmospheric monitoring instrumentation Product Manager Jeff Ritter provides tips you need when choosing, installing, and maintaining a weather station to ensure you’ve got the accurate data your project requires. You will learn:

• How to ensure you’ve got all the measurements you need—enough to address concerns for now and later
• The best locations and practices to collect weather data
• How to keep your data reliable and continuous
• Choosing a system robust enough for your application
• How to handle weather data once the station is installed
• And more

Next steps

• Explore the climate and weather solutions hub
• Link to slideshare
• Listen to We Measure the World Podcast Episode 43: On-site weather data paving the way for better road construction
• Watch webinar: Weather data: Why accuracy is more complicated than you think
• Learn more about the ATMOS 41 Gen 2 and ATMOS 41W weather stations
• Follow us on LinkedIn: https://www.linkedin.com/company/meter-group
• Questions?

Our scientists have decades of experience helping researchers and growers measure the soil-plant-atmosphere continuum.

• Talk to an expert
• Request a quote

Transcript:

BRAD NEWBOLD 0:00
Hello everyone, and welcome to “Top mistakes that impact your weather data quality”. Today’s presentation will be about 30 minutes, followed by about 10 minutes of Q and A with our presenter, Jeff Ritter, whom I will introduce in just a moment. But before we start, we’ve got a couple of housekeeping items. First, we want this webinar to be interactive, so we encourage you to submit any and all questions in the Questions pane, and then we’ll be keeping track of those for the Q and A session toward the end second, if you want us to go back or repeat something you missed, don’t worry. We will be sending out a recording of the webinar via email within the next three to five business days. All right, with that out of the way, let’s get started. Today. We’ll hear from meters ATMOS Product Manager Jeff Ritter, who will discuss how to avoid the biggest mistakes that impact the quality of your weather data. Jeff is the product manager for plant canopy and atmospheric monitoring instrumentation here at meter. He earned his master’s degree in plant physiology from Washington State University, where his research focused on leaf level gas exchange and the impact of plant biochemistry on the measurement of the global carbon cycle. Prior to working at METER, he held a research faculty position at Washington State University in the Department of Crop and Soil Sciences, and so without further ado, I will hand it over to Jeff to get us started.

JEFF RITTER 1:16
Thanks, Brad. Thanks for the introduction, and I appreciate the opportunity to talk about something that I’ve been thinking about a lot lately. I recently had the opportunity to attend a meeting of the American Association of State climatologists. This was a gathering of their Mesonet community, and they had a whole day devoted to the operational aspects of maintaining their weather networks. And one of the big takeaways for me from that was that there’s a lot of universal issues that folks are facing when they’re maintaining a weather station, whether that’s maintaining a vast network of complex weather stations across an entire state, or whether that’s just maintaining one or two. Obviously there’s unique challenges there, but I thought it would be worthwhile to go through some of the challenges that you face when it comes specifically to understanding when you have quote, unquote bad data. And I wanted to dive into what some of those data parameters look like, so you can tell whether your data is actually suspect or whether you are seeing microclimate effects that you might be interested in. So you’ll, you’ll have to forgive a bit of frivolity in my my title there, that I’ve called to guess is human to measure divine, that that’s borrowing from Alexander Pope’s an essay on criticism. But really want to, want to get to is, when you see your data, are you just guessing that your data is good, or you just guessing that there’s a problem with it? Or do you have an actual way to to compare whether your your weather parameters are an error or giving you real physiological parameters? So we are going to attempt to decode some of those things here. Now this is far too big a topic to cover in just one webinar that would require a whole lecture series, so these are just some of some of the things at the top of my mind. I wanted to start by drawing your attention to this particular graph that’s looking at evapotranspiration. So if you’re not familiar, evapotranspiration is simply a measure or a calculation of how much water is being lost from a landscape, whether that’s evaporation from the surface and the combined effects of transpiration water being used by the plant. So the bigger the number, the more water is being lost by a landscape. In this case, we’re looking at reference evapotranspiration. So this is not a direct measurement of water flux. This is going to be a measure of different meter logical parameters, and then we can calculate the expected water demand over that landscape. So if you’re measuring things like solar radiation, wind speed, air temperature and relative humidity, you can calculate reference evapotranspiration. Now this graph is really interesting to me, because this is two weather stations that are located pretty, pretty close to where we are, here at meter group in in the southeastern corner of Washington State, in the United States, each of these stations is just about a mile apart from each other, so you’d think being that close, they would give you very similar values for the water demand over the landscape. But what you see over the past month this was being measured is a huge discrepancy between these two stations, sometimes up to 40% different between the water demand at site one versus site two, especially on on what some of the sites are being measured as hotter days. So by looking at this, can we tell, just by this graph, is this real, or one of or one or both of these sites in errors? Are both of these sites telling the truth? Are both of these sites lying? How can we tell. So that’s kind of what I’m what we’re going to be talking about today. Are some of the ways you can tell whether you’re measuring, what you’re measuring is real, or what you’re measuring is for some some way in error. This is building on a webinar that was given by Dr. Colin Campbell a year or two ago. He was focusing primarily, or more so, on how to choose the right station for your application. He went into great depth with sensor technologies and how one technology might work for you. The balance between spending $100,000

JEFF RITTER 5:46
sensor outputs, but I’m not going to talk too much about actually choosing your station. I mentioned that here just because whatever station you’re using will play a role in how you site and install your station, and we will be talking about that. So keep that in mind moving forward, that whether you’re using an all in one station or a tower, that’s going to impact your needs for for choosing your location, and that’s where I want to start. You can’t know whether your sensors are good and when you’re getting good data if you don’t carefully choose where you’re installing it. So we’re going to spend a bit of time first talking about siting and installation and the problems that can arise there and things that can get in your way. And so I want to start first by talking about your data collection.

JEFF RITTER 6:27
One of the things that’s fairly easy to overlook these days is, Will I have access to remote data? We’re very used to just having cell coverage almost everywhere that if you are installing in a very remote site or someplace that’s very rural, you might not have access to the telemetry options that you need, whether that’s cell coverage, Wi Fi, radio, etc. One thing that you can run into as well is you might have a station out already that has good coverage. You have great telemetry options there, but depending on the landscape, that might not be homogenous everywhere. As an example for this. A couple of years ago, a colleague and I were doing some sensor installs in the Rocky Mountains down in Colorado, and what we found at first is as we were climbing higher and higher, we might have really good cell coverage on one hill, and then we would cross over to different different part of the mountains, and we’d have no coverage at all. And it got to the point where we could have good coverage in one spot and move a couple of steps to one one side of the other, and we would have no coverage. So we had to very carefully find every every place that we were installing where we would have good cell coverage, to the point where my colleague, Chris Chambers here, that you see in that you see in that picture, was walking around with an antenna trying to find the exact location we had to put stuff to get access to remote data. So what does this look like? What? What does it look like if a station doesn’t have great connectivity?

JEFF RITTER 7:58
That’s sometimes a little bit tricky to display in a graph, so that’s what I’m attempting to do here, and I’ll walk you through what we’re looking at. This is a time series graph where we’re looking at both the time between uploads or how frequently a station is checking in with its server, how frequently it’s able to upload data. And that’s the solid black line that you’re seeing. Obviously, the lower the number there, the better. That means you’re getting almost real time data. Closer to zero means you’re constantly getting uploads. The light gray dots are the signal strength. So the higher the number there, the better, the the more cell coverage you’re getting. Now this station, for the first couple of months that it was out, was looking great. It had almost no time between uploads. Was getting near real time data and very strong signal. The reason I want to share this graph with you is you can see that there’s some some jagged peaks over there on the right. And I want to call out what happens here around the beginning of July, where all of a sudden it looks like the signal strength improves, and then it starts degrading over time, and we start getting a huge time between uploads. What happened here is this station was actually moved. So for various reasons, sometimes you have to move a station. This one was moved, and while the it still had cellular coverage, there were some issues with the cell server or the cell provider that the towers that it was able to access in the new location were from a different carrier, and all of a sudden the modem started struggling, and you started getting these big gaps where it would it would go a day between uploads, and then a couple of days not getting the real time day that you were expecting. So keep that in mind whenever you’re selecting a location that you need to know what your coverage is and the carriers that are providing and make sure that whatever logger system or communication system you’re using is going to work with that. So moving into some actual measurements that we make and what what can kind of get in your way, I want to talk about actual physical obstructions. So when we’re talking about choosing a location, looking at the surrounding area, is extremely important for knowing what you’re going to be measuring. We’re going to be referring a lot in this talk to the world Meterological organization guidelines. So if I refer to WMO, that’s what I’m talking about there. I’ll also mention the American Association of State Climatologist guidelines, aasc, as well as potentially the National Weather Service. So these organizations have guidelines both for exposure, so distance to an object, as well as the Installation height.

JEFF RITTER 10:36
So if we just start with wind, because that’s one that’s pretty easy to visualize what we mean when we’re talking about the importance of distance from an obstruction.

JEFF RITTER 10:44
WMO recommends at least 10 times the height of the obstruction for your weather station to be put away from that obstruction. So they call this out as the absolute minimum distance from that obstruction. Typically, they say if you are 10 to 20 times the height of the obstruction away, you don’t you might need to do some corrections. So it is very difficult to find a site that is perfectly within these these specifications. So we’re doing the best that we can when we’re looking for these locations. What you’re you need to do? I’ve got a little graphic there showing the attenuation of the wind behind a tree. You need to know the height of the obstruction and then be able to actually physically measure at least 10 times that height away to put your weather station. I also put on here the recommended height for WMO as well as AASC for where you are measuring that wind so WMO recommends it measuring at 10 meters as well as potentially at two meter simultaneously and AASC is measuring it at two meters.

JEFF RITTER 11:49
So we’ll get back to talking about wind and actually looking at some obstructed wind data here in a second. But I wanted to take a moment to discuss sensor installation heights, because these are all different depending on what sensor you’re looking at. When you were looking at what we would see on the right is more of a modular weather station where all of your different parameters are located within an individual sensor that you can put across an entire tower. This tower on the right would be something that is about 10 meters. On the left is what we would call an all in one weather station, where all the parameters are in one compact package. So the first thing that you’ll see is that these these sensors are spread out over the entire tower, where we have some at two meters, such as our air temperature and vapor pressure. Others are recommended. Like I said before, wind speed is up at 10 meters. And then we’ll also talk about solar radiation and precipitation, which is a separate rain gage there on the ground, the all in one station. While it has a lot of advantages as far as how convenient it is to use it, you can’t fully abide by these WMO recommendations. You can’t have a wind speed measured at 10 meters and an air temperature measured at two so typically, we recommend an all in one station is mounted at two meters. That gets you AAC recommendation for wind as well as air temperature and vapor pressure. So what does it look like if you don’t abide by these guidelines? What is What does a wind obstruction really look like? And can you tell that in your data? I want to show you a graph here. This is a time series of a weather station, again, one that’s close to our facility here. And if you just look at a time series of the wind direction, can you look at that graph and tell me whether or not this is properly sited, whether there’s a wind obstruction? Ah, not really like I look at that, I can’t tell you almost anything about what might be in the way. I mean, if I look at it for a long time, maybe I could tell you some general trends, but it’s very difficult when looking very difficult when looking at a graph in this way. And so generally, what we want to do when you’re trying to look at whether you’re actually seeing impacts from nearby obstructions is look at either wind rows or even a simple radar plot. What we’re looking at here is simply a frequency distribution in a circular radar plot of seeing where the wind is being measured at. So this starts to tell a picture. Now this tells you, yeah, all the wind is coming from one direction. But this, even by itself, is not enough. We need another form of data on top of this, because if we compare two different stations that are just a mile apart, you know? Yeah, maybe that soccer field the graph on the right is obstructed by something. But can you tell me that that one that was measured in the neighborhood is not obstructed just because it’s a different pattern? So we need another form of data on top of that, and typically, I like to go to satellite images or pictures of some sort. So. Where I overlay the wind direction over the satellite. So this is a satellite image of that second station, not looking at the soccer field yet, the one that we are interested in whether or knowing it’s obstructed, we want to see if this one actually represents regionally prevailing wind speed and direction. So that weather station is located about here in a field between these two houses. If you were to go and actually measure this, you’d find this is not fully WMO compliant for for exposure. This is closer than 10 times the height of that nearby house. And again, remember, WMO specifies that 10 times the obstruction height as the minimum distance you should be away the wind wake that you see from obstructions can easily be 12 or 15 times the height of the obstruction. So we might still be seeing some interference here. And when we overlay that wind pattern over here, we do see that it seems to be coming east, west winds potentially being funneled between, those those rows of houses. So while we can’t say that this is a, a we can’t be fully confident that this is regionally prevailing winds. We can at the very least say that it is not obstructed by the same things that this station is. This is the station that is at our facility here at METER Group. We have it out in a soccer field just behind our building. So you probably already see what’s going on when you see that where that weather station is located, you’ve got all these buildings located to the south of that weather station. So I’m going to overlay first the neighborhood wind graph over this so you can see if we were to assume these were the regionally prevailing wind directions? What that would look like on this graph and you can, you can probably tell that’s not what we saw here with those buildings in the way. When I put the soccer field graph up again, you can, I think, pretty clearly, see what’s going on. You can see the impact of the obstructions of all those buildings to the south preventing any winds from coming across there. So the impact of that obstruction on wind direction becomes almost immediately clear when you look at the data in this way. But keep in mind this can also then impact wind speed, obviously, but also air temperature and a lot of other weather parameters that are going to be impacted by these obstructions. So moving on to something else that can obstruct your data and impact things like calculations of evapotranspiration is if you have obstructions in your solar radiation, this also is oftentimes very apparent once you know how to look for it, I like to I stole this picture from Dr Campbell’s talk where he was also talking about solar obstructions. This is a good example of a solar radiation sensor actually mounted below the level of the data logger, which means that you are virtually guaranteed, at least certain points during the day to have a big shadow cast across your solar radiation sensor. But I’m it’s easy to say that that’s a problem, but the question is, what does that really look like in your data? So I wanted to find an example of what good and bad solar radiation data look like, too, so we can compare. So this is an example of solar radiation data over the past. Over about a week or so from this last month, you’ve got solar radiation on the y axis measured out in units of watts per meter squared, those first two days you might look at that and say there’s something funky going on there, but really, those are just cloudy days. It is difficult to tell if you are getting obstructions, if you are always measuring under under clouds, because cloudy days look very noisy from a solar radiation graph, you really need to be looking at clear sky days in order to tell if your solar radiation sensor is shaded. The other thing you might notice, if you look really closely at this unobstructed sensor is this step wise fashion that you see on the clear white on the on the clear sky days where it’s not a perfectly smooth graph like you would expect, and that’s what you’re going to see on any weather station that has a bird deterrent kit or a bird ring with spikes on it at the top. That is a very important accessory that I would recommend a lot of weather stations use if you have bird activity in your area, and we’ll talk about the impacts or the importance of that precipitation later, but that’s what you’re seeing there, that step wise fashion for solar radiation. So here you’re able to see a new, well calibrated sensor, and what that looks like in both cloudy and clear days. What though happens if we put a sensor up that is positioned too close to a tree? Well, if you were to compare these on cloudy days again, it’s very difficult to tell if there’s anything wrong. Because there’s a lot of noise in those measurements, and it’s hard to say whether that was actually shaded or whether that was due to cloud effects. On the clear sky days, so this one is very easy to determine that there’s an issue. The solar radiation sensor in red has got a unobstructed view of the sky up until around solar noon, and then it gets into deep shade for much of the afternoon. Normally, if you have your station near a post or an electrical pole, it’s not going to be this obvious or for this long, but you will see periodic dips around the same time every day, and that is very indicative of you getting a shadow that needs to be accounted for in your calculations of things like evapotranspiration, as that will play a big role in those calculations. So the there are some recommendations for how to site your solar radiation sensor. The biggest thing is just that it’s got an unobstructed view of the sky. Height doesn’t really matter for your solar radiation sensor. So for for installation, if you’ve if you’ve got a 10 meter, 10 meter station, it’s got to go at the top of that station. Otherwise, I recommend having it more like two or three meters off the ground, just so you can access it more easily. For cleaning, every once in a while, you do need to get out there and clean off the top of your solar radiation sensor so it doesn’t get covered in dust or bird waste. Okay, let’s also talk about precipitation. There are some recommendations for how close obstructions need to be to your rain gage, but I actually want to talk about first, the actual obstruction that your rain gage itself is on the landscape, so you’re putting your rain gage out, and when wind flows across your rain gage, it some of it will enter the funnel and actually create an updraft. So that updraft then takes precipitation, both both solid and liquid precipitation, and will deposit it. And it will deposit it basically everywhere except for where you want it to be. It will deposit it outside of your funnel. If you are interested in measuring snow, for example, and you have this in a very windy location, you can lose up to 50% of the actual snow being snowfall. It will miss your funnel based on these updrafts. So there’s a couple of ways to mitigate this. You’ll often see, when you have a standalone rain gage, that they’re put closer to the ground where there is less wind. You can use funnels of bigger diameters. The bigger the diameter of funnel, the less of an issue this is. The other thing you’ll commonly see are researchers using a shield around your rain gage so the one in this picture here is called a double altar shield. So it’s got two rings of this shielding material to prevent updrafts when the wind comes across, they hit the shield rather than causing an updraft at your at your rain gage. A lot of the time if you’re interested in measuring snow, you have to have them a decent bit off the ground. One of the requirements is, obviously, that needs to be above the level of snow accumulation. The other obstruction that we should call out here is the obstruction that can be caused by other things in nature. A lot of weather stations provide a very practical perch for birds to sit on. And when birds are perched on your weather station, they’re going to do bird things. And that can, that can foul your your rain gage up. It can, it can clog it. So it’s important to know what a clogged rain gage looks like, what your data are going to look like if water is unable to move through or only moving through very slowly. There’s sometimes we can get these things filled up with dead bugs or leaves. Anything that will be in your area can fall into your rain gage and clog it up. And we’ll talk about maintenance. Here. Here in a little bit about how frequently you should be taking a look at this. But I wanted to actually look at some data of what a clogged rain funnel looks like. Now it might be a little bit difficult to see on that top graph. If you look really closely, you can see a big slug of precipitation. So we’re looking at precipitation in millimeters. On the left hand side of that graph, you see a big, a big rainfall event. And then if you were to look really closely and zoom in, you could see individual little tick marks going out for a long time after the rain stopped. If we zoom in on that, that’s what we’re seeing on the bottom graph, you were seeing every piece of data that comes out after that is the exact same value. And what we’re seeing there is the minimum measurement resolution for this particular rain gage. So this, this rain gage that we’re looking at data for, can count every drop of rain. So what happens is that funnel gets filled up, but because it’s clogged, it doesn’t all run through very rapidly. So over time, it just lets a single drop of rain through slowly over hours or days, depending on clogged the funnel is. So what you can see are these very unnatural patterns of rainfall where it just looks like it’s raining once every 15 minutes or every hour, depending on the severity of the clog. So if you see unnatural patterns of rainfall like this, after a big rain event, it could be time to go out and clean out your rain funnel. So the recommendations for pre-cip are basically, you want your standalone rain gage closer to the ground than your other sensors. There is a minimum height recommended that is to prevent in splashing from water bouncing off the ground. So that’s 30 centimeters from WMO, or one meter from from AASC. And also try and keep this away from obstructions, although in this the recommendation is four times the obstructions height. So just to summarize that, you know, it needs to be open to the sky, obviously, and, um and above the level of snow accumulation. For diameter of the rain funnel, the recommendations that WMO puts out are largely saying that this 16 centimeter diameter is necessary for if you are very interested in solid precipitation, like snow, that 200 centimeter square, which is about a 16 centimeter diameter is necessary for that, and that, excuse me for the National Weather Service, they’ve got a standard eight inch gage, which is 20 centimeters. So a lot of you have probably got this far and are now thinking, you know, that’s great, but I my site that I’m interested in measuring at doesn’t conform to that. I can’t have a weather station that’s, you know, 10 times away from all the nearby obstructions that my site just isn’t isn’t even that big. And so that’s I kind of wanted to now turn to the fact that we are, a lot of the times, interested in measuring microclimates, and so understanding the difference between an obstruction and what actually is your area of interest is important. So we need to be able to measure where it matters to you. And to make that point, I wanted to look at, again, a comparison of a couple of weather stations we have here. Again, we’re going to use that First Station at at our soccer field, which is right outside our building, and another weather station that I have located nearby in a test bed. This test bed is just about 200 meters, if we walked in a straight line from one station to the other, but it’s down a very steep hill. This hill, you know, runs, runs north, south, along the landscape. And then there’s a river that that runs right next to that weather station. So the question is, if we were to compare a couple of parameters for these weather stations, what are we going to see? You know, what? What sort of if only 200 meters away, we would expect them to be very similar. But you probably already guessed that’s not what wouldn’t be very interesting for me to show you that. So I wanted to go through just a couple of parameters and show you how different these sites can be, even if, even if your weather stations are are installed properly. This is air temperature over a couple of weeks in in July and August from the soccer field weather station that’s near our building. And so you can see here on the Palouse, we get temperatures up to about 35 or 96 Fahrenheit over that time period, you know, down to about 10 degrees Celsius, 50 Fahrenheit at night at the building. What’s going on, though, down in the valley, just down that hill, 200 meters away, we’re seeing daytime highs, almost exactly the same, almost no difference there. But at night, we see this really interesting phenomenon where daytime lows are extremely depressed compared to where we are just just 200 meters up the hill, sometimes you can see temperature differences of five or almost 10 degrees Celsius. And this is, this is in win-, this is in summer, so as we move closer to the winter months, that station down there gets to freezing temperatures much earlier. This has a big impact as far as what sort of plants grow down there, what sort of disease pressures you have, etc.

JEFF RITTER 29:29
We see the same thing, obviously, with with relative humidity, relative humidity and air temperature are tied so we get a much higher relative humidity at night down in the valley than we do just up the hill. If you are interested in frost monitoring or spraying applications, this can play a big role in when you would make those make those decisions. And the last I wanted to show you was wind speed. Wind speed also, for the most part, is very depressed down in that valley. So you get, you get not, not much higher winds, but you do get higher winds up the hill, down the hill, you get lower winds, so you have lower temperature, higher relative humidity and and lower winds. So you might already be able to tell where we’re going, as far as evapotranspiration, if we’re talking about why two sites might be different, but we’re not, we’re not quite there yet. So just keep that in mind. We’re going to shift gears a little bit at this point in the talk we have discussed how you’re going to where you want to put your your station, how you’re going to install it and site it, and then how to determine what sort of micro climates are important to you. After your sensors are actually installed though it’s important to understand you need to plan going in upon installation of what you’re going to do for calibration and maintenance. If you take away nothing else from this talk, just know that sensors are going to drift. You need to plan on some of your sensors drifting. You should know which of your sensors are prone to drift and how you’re going to deal with that. You can’t have accurate or you can’t trust that your data is accurate if you don’t keep your sensors calibrated, and all weather stations are going to need service. There’s no such thing as a maintenance free weather station. So you need to have a plan in place of how frequently and when you’re going to go out and service your weather station. So let’s, let’s do a quick dive into sensor drift. I wanted to show you what a drifting sensor looks like, and we can use that same unobstructed sensor that we talked about before when we were talking about shading. So this, again, this is a new, well calibrated sensor whose results we can trust we’ve this has been recently calibrated against a NIST traceable standard. If we were to take an old sensor that’s been out in the field, say, about five years, one of the sensors that that we know is prone to drift. What can we expect that to look like? Again If we look at a cloudy day, it’s really difficult to tell if anything is wrong, but as soon as we look on a sunny day, you can see this big mid day sensor depression for that sensor in in red, that sensor has either been damaged where the lens is severely scratched or coated and in dust or bird droppings, or the sensor itself is drifted and needs to be recalibrated. You can see it’s more sensitive to the bird spikes that are on it, and it doesn’t ever reach the daily maximum that the new sensor does. And that’s that’s what you see in a lot of sensors that have drifted. You need to compare them to to a well calibrated sensor. The sensors that are are prone to drift at least of the parameters that are commonly measured in weather stations are your solar radiation sensor, relative humidity, and atmospheric pressure. And so I would strongly encourage you to follow the manufacturer’s recommendation for calibration. They almost are always going to tell you which sensors need to be recalibrated, and which ones don’t a lot of the time you don’t have to worry about recalibrating air temperature and wind speed, especially if it’s the son-, ultrasonic, anemometer. So if they are telling you these sensors need to be recalibrated, they likely are even telling you how much drift can be expected year to year. So you can you can know how much error you are potentially introducing into your data if you don’t calibrate them. Maintenance is another big one, especially with all in one weather stations. I see folks, they’re very convenient to install, and they tend to be very easy to maintain. But it doesn’t mean you don’t maintain them. And the other thing I would love to get across is that maintenance should not be just reactionary. If you are only going out to the field when there is a problem, you’re going to have more problems. You need to, in my opinion, create a maintenance schedule upon installation so you know how frequently you’re going to go out and schedule out when those times are going to be so that would include check in visits where I’m going to go out, even if everything looks fine, just to make sure that there’s nothing going on that I’m not able to see in the data, but also, obviously, as needed visits. If you see something going on in your data that you can tell is a problem, don’t feel like you need to just stick to your check in visits, as needed visits are oftentimes more time sensitive. I get the question a lot also about how frequently you should be going out to visit your weather station. And I put on here visit each station at least once a year. I would consider that the absolute minimum that you should be visiting your station a lot can happen at your site in one year. Typically, you want to plan on at least two check in visits a year, and then as needed. And that is kind of the minimum. Even if you have a really big Weather Network, you want to be able to get out to each site at least twice a year, so you can do, you know, regular maintenance, but then also get out back for for as needed. The other thing I really strongly recommend for maintenance is going out to each site with a checklist. I i prefer a pen and paper checklist, but if you also have one on, on an iPad or something, having an actual checklist that you can go and check things off and then sign your name and date is really great for metadata later, but it also ensures that everything that was supposed to happen at your site happened. Most of this is pretty routine, cleaning out your rain gage, like we talked about, maintaining the area around your station, so mowing and then doing any pre scheduled maintenance and your as needed repairs that you find. The biggest thing I wanted to point out on this checklist that I’ve recommended are these pictures taking both arrival and departure pictures. So I’m a big fan of using pictures as an important form of your metadata. When you go out to your site, if something’s happened, you want to document the state that it was in. So you can, you can flag your data and say, this is what this was our data collection until we got there to fix it. And you’ve got a picture to show that, and then you take a picture when you depart that shows that you fix the issue. But also you can look back later if there’s an ongoing issue, and then say we didn’t actually fix this. You know, I see in this picture here, we left our data logger, door open, the maintenance checklist for every organization, maybe even every site, is going to be slightly different. This is just an example template. And then moving on to the last section I wanted to talk about here is metadata. Make sure you you have a sense of what sort of metadata you’re going to record before you go out, but the more metadata you are able to record, the better. And the longer you are planning on having this weather station installed, the more important metadata becomes. So things like sensor replacements, who is visiting the field and when and what they’re doing out there, and then, like I talked about before, these site pictures, and I like using this picture as an example of why we take site pictures, because this is a case of where we didn’t actually take a departure picture at this at this site. So I noticed that all of a sudden we stopped getting data reporting from this site. So if you’re not familiar, that picture is of a data logger on the right, and the door is wide open, so it stopped reporting data one day. And so I trudged out there, and this is this data found in this was my arrival picture, and when we had visited it last we forgot to close the door of the data logger, and so it wasn’t actually enclosed. The circuitry was not protected. If we had taken a departure picture, we would have caught that, and at least the arrival picture now allows me to go in and flag the data as to why we might see any oddities there. So metadata document, what you can it becomes more and more important. So just to kind of round things out and revisit this graph that we were talking about in order to know whether these sites are giving us good data or bad data, whether we can trust the differences between these two sites, we would need to know what how these sensors were installed, and what we’re looking at right. We need to to realize that site one might be the correct evapotranspiration for a station that’s that’s positioned out in a field far away from from all obstructions.

JEFF RITTER 39:04
Site two just because it’s very different doesn’t mean that’s bad data. That’s simply the evapotranspiration down in the valley, in that in that area that I showed you, it simply has less wind and and overall lower temperatures, and so you see lower overall water demand. So being able to understand when these things are due to sensor issues, sensor drift, calibration or maintenance versus actual microclimate effects is critical. So overall, the best practices are, you know, after you choose your station, spend some time choosing your site. Those WMO or AASC guidelines are important depending on the question that you’re asking, and it’s important that you know what those guidelines are so you know when they don’t apply to your question. If you are trying to put a weather station down in that valley, because that’s what you’re interested in you’re interested in the plants that live down there, that’s where you need your weather station. And make sure that you you plan ahead, you make a plan for maintenance, you make a plan for calibration, and build that into your budget, because if you want to leave your weather station up for five years, some of those sensors are going to drift by then, and you’re going to need to have a budget for calibration. And lastly, spend some time familiarizing yourself with what bad data can look like. It’s better for you to know when your data is bad versus when there’s something interesting going on, so you don’t waste time and you don’t you don’t make the incorrect decision. So understanding microclimate effects versus sensor effects is very important, so that I really appreciate the opportunity again to talk about this weather data stuff, and so I’ll take any questions.

BRAD NEWBOLD 40:54
Thanks, Jeff. So we’re going to use the next few minutes to take some questions from the audience. Thank you to everybody who sent in questions already. There’s still plenty of time to submit your questions if you’d like, and we’ll try to get to as many as we can before we finish. If we do not get to your question here live, we do have them recorded, and Jeff, or somebody else from our METER environment team will be able to get back to you directly via email to answer your question. All right, so first question here, they’re asking, 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 41:35
Yeah, that’s a tough question. Mounting on a roof is not ideal, and it really depends on, on where your roof is and what question you’re trying to answer. So if you don’t have access to, you know, a nearby field to put your weather station in, and the roof is the best you can do, you have to understand that that’s going to be a very blunt tool for trying to understand, for example, the wind speed over a canopy, if you’re interested in measuring something like evapotranspiration, so especially if you are putting mounting this on a roof surrounded by other buildings, the turbulent effects of wind over an urban Landscape make that very difficult to deal with. There are some ways that you can model for wind height, but you need, you need some other, some other data. So, you know we can, we can discuss a little bit wind profiling and how you’re able to to model for wind height, but short term is, there’s no real good way to go from the top of a roof of a building in a city and determine what the wind speed is going to be out in the field.

BRAD NEWBOLD 42:52
This next one, you talked about birds and the issues that they that they cause. Can you go into a little bit more detail about bird deterrence, and is there any way to avoid birds? I don’t know if you can say altogether, but in many ways.

JEFF RITTER 43:09
So no, there’s. There’s no way to make something that is completely impervious to animals at all, not just birds, but other large animals. Sometimes you’ve got to have this out in a field where there’s there’s cattle, and so folks will build an exclosure around it, you know, and animals will still get through. Birds will still find ways, a lot of the time to land on weather stations, the bird deterrent kits that a lot of weather stations have are really good, and they will prevent most birds from landing and perching. I mean, the bird is looking for the easiest place to land, and so if it has to work at all to land on your weather station, it’s going to go somewhere else. I have seen some bird deterrent kits where the actual spikes are pulled out by either birds or squirrels, rendering it, you know, just a plastic ring. So they’re they’re very good, but there’s not ever going to be a 100% way to to keep birds off. Now, I should say these bird deterrent kits, again, they don’t hurt the birds. They just prevent them from from landing, and so some birds are still able to land and perch, depending on their size. I’ve heard that some owls don’t really care as much, especially some smaller owls. So there’s always going to be some issues whenever you are are dealing with animals.

BRAD NEWBOLD 44:31
I was going to say, Is it, is it possible to, uh, mount one of those fake statues of birds, you know that they have sometimes on roofs and put that somewhere nearby to scare other birds away?

JEFF RITTER 44:45
It is, but if you are, you know, putting this out in the middle of a field that’s 100 meters away from anything else that it might, might be hard to find a spot to hang a fake bird, but, yeah, I do see those a lot in like lumber yards and other places that could work, although I don’t have much experience with that.

BRAD NEWBOLD 45:04
All right, thanks. A couple questions about trying to get ET measurements out in fields, especially for agricultural purposes. First question here is, why, why do you feel that it’s best to install weather stations in either an idle or fallow field for ET measurements?

JEFF RITTER 45:26
So it’s important to kind of understand when we are measuring ET, are we making a direct measurement, or are we doing a modeled ET? So for these weather stations, we’re not actually talking about a direct measurement of ET we are measuring reference evapotranspiration. So that means we need to know the solar radiation, the relative humidity, the wind speed and the air temperature, and from that we can calculate a reference value that applies to a certain canopy type. For reference ET, it’s typically referring to a fully enclosed grass or alfalfa canopy, depending on we’re talking about, ETO or ETR. So that’s very different from measuring ET, where the site of the field is going to matter, we are much more interested in reference ET of the environmental parameters that are going to the energy balance for that land.

BRAD NEWBOLD 46:34
Along with that, there was question about exclosures with our, I guess, are chain link fences or chicken wire considered obstructions? And if you were trying to do ET measurements or modeling, should they have the weather stations within a smaller exclosure or outside of, I guess, of that fence?

JEFF RITTER 46:58
Yeah, it’s always a balance, and most of the time, you just need to make sure that the height of the fence is carefully weighed against the height of your installation. If you’ve got a 10 meter installation, then the fence probably is not creating an obstruction. But if you have a six foot chain link fence, and you’re mounting at about six feet all in one station, it likely is providing an obstruction there. I would say, if you are in an area that requires an exclosure, you probably want to be measuring et inside of that exclosure, because the reason for that exclosure is to to protect against against wildlife for the most part. So that’s you would want to make sure that your, your your gear, is protected there. So it’s it’s difficult to say without knowing the exact specifics of the installation, but likely, if you’re putting an exclosure out, you just need to make sure that you’re mounted higher than your than your fence.

BRAD NEWBOLD 47:54
I think we have time for two more quick questions. Any advice for measurements, if it’s done under tree canopies?

JEFF RITTER 48:06
Yeah, that that’s can be tricky for a couple of reasons. For one, you’re likely going to want to go with an ultrasonic anemometer anytime that you are interested in very low wind speeds. An ultrasonic is going to be be the way to go, and that tends to be what happens when you’re measuring under a canopy. The biggest problem that you run into under a canopy is precipitation, because though your rain funnel can get clogged very quickly from detritus from the nearby trees, and that’s actually when I have seen issues with bird deterrence being the spikes being stolen by squirrels is when it’s in and around a lot of squirrel activity. So the biggest thing with with that is knowing what you’re getting into and making sure that you’re going out there pretty frequently to clean out that, that rain funnel and that that being said, you know, that’s what the understanding that is precipitation even interesting to you in that application, under, under the canopy. In a lot of cases, yes, but sometimes, no. Sometimes you want another rain gage that’s measuring the actual rainfall at the top of the canopy, and you put that somewhere out in a clearing, and then the other other parameters you care more about, under the canopy.

BRAD NEWBOLD 49:25
Okay, this is gonna be our final question. Thank you again, for everybody who’s submitted questions, we have several that we will not be able to get to again. We do have them recorded, and Jeff, or somebody else from our METER team will be able to get back to you via email to answer your question. Final question here, going back to installation guidelines, so weather stations and weather instruments come with installation guidelines and manuals. What should individuals do if those particular guidelines and manuals differ from, say, the WMO guidelines? Is there, is there a preference or priority as to which they should they should follow?

JEFF RITTER 50:04
Yeah, that’s a really good question, and you have to keep in mind that those WMO guidelines are kind of the baseline for inter comparability between sites, for being able to to measure a what’s considered a regionally representative weather pattern. So if you are using, for example, an all in one weather station, you’re never going to be able to conform to all WMO guidelines. So you might have to mount it at two meters, and that might be the manufacturer’s recommendation. If you tried to put that up at at 10 meters, all of a sudden there’s another you know, you run into other other issues. So that comes back to that question of, when you choose your station, you have to know some of those things before you buy like, do I need to completely conform to all W guidelines? Or is there some flexibility based on, based on where I’m going here? So to your question, I would say, normally, the manufacturer is going to have already looked at that for most, what I would call research grade weather stations, and so they know the limitations of their weather station. But it’s still an important question for you to understand what the what the guidelines are if you’re if you’re not going to conform to them.

BRAD NEWBOLD 51:27
Thanks again, Jeff, that’s going to wrap it up for us today. Thank you everybody for joining us. We hope that you enjoy this discussion. Thank you again for all the great questions. And again, we’ll, we will be able to get back to you if we did not answer your questions here today, please consider answering the short survey that will appear after the webinar is finished, just to let us know what types of webinars you’d like to see in the future. And for more information on what you’ve seen today, please visit us at metergroup.com Finally, look for the recording of today’s presentation in your email and stay tuned for future METER webinars. Thanks again. Stay safe and have a great day.