NASA ARSET Introduction To The Aerosol Robotic Network (AERONET)

0:00 Speaker 1: What you see in last year, May 2023, the locations of all the stations all around the world. In the last 30 years, it has grown a lot. Currently, we have about 600 active stations. Over the time, we have about 1,800 station deployment, which means some of the sites have been decommissioned for various logistic and other reasons, and some of the new sites come along every year. Aeronet coverage is across 100 political gaps in many areas around the world, specifically over Africa and over northern hemisphere, over Russia and other places due to geopolitical issues and difficulty in maintaining stations on these locations.

0:55 Speaker 1: Here are some of the locations or the networks which are part of AeroNet program. The first picture you see here is from the NASA Goddard. We have a calibration centers and one of the major AeroNet lab here, which have capability to calibrate 100 instrument on a given day in time. And we have photons on the top, which is in little fronds, error span, We have NEON.

1:27 Speaker 2: NEON is very deep.

1:29 Speaker 1: These are also located in very difficult locations and typically these instrument, sun photometer is mounted on actually with many other instruments on these kind of towers which you see here in the picture.

1:44 Speaker 2: On the right corner.

1:46 Speaker 1: Then you have some sites in Australia. We have AROCAN Canadian sites. which are really in difficult locations under the snow and ice conditions. So they operate certain time of the years only. So this is just to give you a glimpse of what different sites looks like or where those different networks operates. Now, the growth didn't happen overnight, as we mentioned earlier, it took 30 years. And what you see on the bottom of the time series is the number of sites which came along added into the Aeronet networks over the years.

2:26 Speaker 1: So if you do a trend fit, it's about 25 sites which we include new sites in Aeronet networks combined from all the networks which are part of Aeronet. And you can see how the locations of those sites have been evolving over the time in the map on the top. And the colors are basically mean aerosol optical depth for that particular years which the station is displayed. The AOD is represented at 500 nanometer and we'll talk about that little bit more what it means. Here is some more details about some of the sites which came along in the last three to four years, starting 2020 to 23.

3:13 Speaker 1: As you can see, 2020 and 21, 22, those two and more than two years were heavily affected by the COVID pandemic.

3:26 Speaker 2: All over the world and that internet.

3:28 Speaker 1: Operation also got affected because it was very, very hard to maintain those sites. But still we were able to put a lot of sites during that period and after that period in many, many different parts of the world. You can see the deployment, the colors on the map doesn't have any relevance here, but you see the locations of the new sites which came along. We are already in discussion with many different countries, including U.S., Saudi Arabia, Iraq, India, and other countries in Africa where new sites in coming year will come up, which is about 20 more sites.

4:09 Speaker 1: Now, if you look at the data records, availability of data records for various climate and air quality research, then we have about 30 years data records from one site that was ordered. If you're looking for more sites, then we have about 20 years of data records from 71 sites.

4:31 Speaker 2: Spread across the region.

4:33 Speaker 1: So you can see very quickly in the first 10 years itself, we went from one site to 70 sites. The 10-year-plus data records is available from more than 350 sites, so huge change in last 10 years. all the way from 70 sites to 350 sites. And if you're looking for five years plus data records, then we have about 565 sites. As I mentioned, currently we have about 600 active sites all around the globe. Okay, now in additions to these permanent Aeronet sites. Aeronet also heavily involved in supporting field campaign, and these field campaigns can be NASA or non-NASA, depending on the purpose of the field campaign.

5:25 Speaker 1: And Aeronet participate in all kind of field campaigns since the beginning of the program until current time. What we call is Dragon, regional aerosol-graded observation network, distributed regional aerosol-graded observation network, or DRAGON. So the DRAGON, the concept of the DRAGON is that you put many, many aeronet sites in smaller regions to observe or to measure aerosol gradients in space and time over smaller regions. And one of the example, if you see here on the left side in this animation, is a similar dragon deployed in 2011 during a Discover AQ field campaign here in Washington DC and Baltimore area, where we had 40 plus signal instrument operated for many months to actually capture that spatial and temporal variability in this region.

6:28 Speaker 1: Similarly, we have deployed these Dragon networks in LA, in Houston, in Colorado, all supporting Discover AQ Field Campaign and internationally. We have these Dragon deployment in UAE, over India. Most recent one, which we just did earlier this year, was called Asia AQ Field Campaign, which was deployed in Thailand, Philippines, and South Korea, and Taiwan. So those four countries actually cover doing this. And these high density networks can be very, very useful, not only to assess the spatial gradient, but to validate the satellite data at high spatial resolution, which are becoming more and more available using the new sensors.

7:17 Speaker 1: So if you have these high resolution satellite data, which is about one kilometer to 10 kilometer resolution, this Dragon network can be significant and the value to the quality of the data in assessing the quality of satellite data.

7:34 Speaker 2: Okay, here I.

7:35 Speaker 1: Just want to quickly show some of the example of Aeronet deployment from different sites. And these are just some example. What I want to point out is that if you look these each deployment, Each of these are different in many things. Some of them are deployed on a tower, which is designed specifically for observation. Some of them are just on the roof of schools, or some of them are in open field on the ground, surrounded by mining activities or lake, and some of them are in the agriculture fields.

8:16 Speaker 1: All 600 sites provide very unique perspective in their own way about aerosol measurements, and they cover all kind of land use, land change, scenarios, different environmental condition, different climatic conditions, from extreme heat in Sahara to extreme cold in Alaska and Canada and Greenland.

8:44 Speaker 2: And South Pole.

8:45 Speaker 1: So we have aeronet measurements in different environmental conditions, different climate conditions to assess how the aerosols are actually evolving around the world. So before we move forward, I just want to pause for 10, 15 seconds here. If people have questions, they can put it on the chat box, the question answer box, and then I will move on. Okay, so I hope that allowed you to think about any question you may have or put in the question answer box. So let's move on. I talked about the bigger picture of Aeronet network.

9:33 Speaker 1: Before we get into more details, I want to make sure that everybody understands what Aeronet measures and what kind of things. And since this is an air quality application training, I want to make sure we make that link very early how this aeronet measurements are linked to air quality, how they are different and how they are similar. So let's look into some of these terminologies which you will hear throughout the first two sessions. Okay, so the first one is called aerosol optical depth. That's the fundamental measurement done by aeronet and it's an optical measurement of the atmosphere okay as some people it also called aerosol optical thickness commonly known as aod or aot it is always represented at certain wavelengths so whenever you're saying aod you must mention at what wavelength the measurements are done typically for most climate application the measurements are done at the green channel, either 500 nanometer or 550 nanometers, but Aeronet along with other instrument can make measurements in other wavelength as well.

10:52 Speaker 2: So what is aerosol optical depth?

10:55 Speaker 1: It is optical depth express the quantity of light removed from a beam by scattering or absorption during its path through a medium. So let's look at this first bullet a little bit more closely. So you have a sun, which is a source of energy, which is transmitting electromagnetic spectrum in all kinds of wavelengths, which reaches to the top of the atmospheres and enters to the atmosphere. When it enters the atmosphere, it interacts with the atmospheric component, which are cloud, water vapor, gases, and aerosols, which are small particles in the atmosphere.

11:39 Speaker 1: And then whatever is transmitted through these atmospheric component reaches to the surface, and that's where the aeronaut instrument is sitting and looking towards the sun directly within 1.2 degree field of view. So what Aeronet does is it makes measurements of this incoming irradiance in a specific wavelength by looking directly to the sun passing through the atmosphere, right? So now if let's say, if we know I0, which is how much top of the atmosphere irradiance in particular channel is, and if we measure I, which is measured by this instrument, then we can tell by looking the difference how much irradiance in that particular wavelength is attenuated or removed by these particles or clouds or water paper or anything else which is there in the atmosphere so that is the basic principle of making measurement of aerosol optical depth now in this case we have shown us and photometer we'll talk about that a little bit more but there is also another instrument which is shown here is called microtops and i'll talk to that about that as well Now the aerosol optical depth represent the loading of the particles in the entire column of the atmosphere.

13:02 Speaker 1: As you can see from the measurement picture, what we are measuring is how much radiation is reaching to the surface. So it represents the entire column. We are not able to separate in the surface or different layers of the atmosphere. It represents the entire column of the atmosphere from the surface to the top of the atmosphere. The value depends on many things. So we'll talk about how to separate into different component, but the value of this aerosol optical depth depends on particle concentration, how many particles are there, what kind of shapes, like dust particles have a different type shapes versus smoke particles versus Pollution particles in the urban environment may have a different shape.

13:45 Speaker 2: Size also matters.

13:47 Speaker 1: Their chemical composition, in other words, refractive index, will tell us whether they are a scattering type of aerosol or absorbing type of aerosols.

13:57 Speaker 2: And where they are in the atmosphere.

14:00 Speaker 1: Although we are making a columnar measurement, but it matters whether they are in lower atmosphere, upper atmosphere, or middle atmosphere, depending on their location, this aerosol optical depth value can have a different meaning. And then the wavelength at which you are making measurements. So like I said earlier, aerosol optical depth always specified by the wavelength and its value will change depending on which wavelength we are showing the measurements. And like I mentioned, it can measure from the ground as well as from the space.

14:36 Speaker 1: So looking a little bit more into mathematical form, how the AOD is calculated. So typically, this is represented by what we call Lambert-Weir law or Weir-Lambert law.

14:49 Speaker 2: And as.

14:51 Speaker 1: From the previous picture, I is the measurement which we are making at the surface using the sun photometer and the I0 is the top of the atmosphere irradiance in that particular wavelength. M theta is what we call is optical air mass. And theta is the solar zenith angle and one over cosine of solar zenith angle is what M is defined by. And this tau lambda is optical depth the atmosphere so it includes optical depth of the atmosphere this is not yet aerosol optical depth this is optical depth of the atmosphere which have air aerosols and gases so now let's assume this measurement of I is done under cloud free condition when there was no cloud we did this measurement so when there is no cloud what are the other component which are there in the atmosphere which will have impact on this I value which we measured.

15:55 Speaker 1: We will have air particles which is also called Rayleigh particles. So this is called optical depth by air molecules or Rayleigh optical depth which we can calculate very accurately using air pressures measured at the surface. So this is the quantity very accurately can be calculated and we can remove this from the total atmospheric optical depth. Then we have aerosol optical depth. This is our interest. This is what we want.

16:24 Speaker 2: Okay. And then we have gases.

16:27 Speaker 1: So for example, ozone, NO2, SO2, water vapor, all of the carbon dioxide, all the gases, depending on which wavelength you are making measurement these gases can have a different optical depth for on that particular wavelength and again based on our understanding and other type of measurement we can calculate this gas aerosol gas optical depth as well with high accuracies so once you have measurement of total optical depth using this equation we can actually remove the Rayleigh optical depth and gases optical depth to get aerosol optical depth.

17:13 Speaker 1: And that's how Aeronet actually get to the aerosol optical depth. Moving on, there's another parameter which is very important which comes from Aeronet called Engstrom parameter or typically alpha or some people also called it Engstrom coefficient.

17:35 Speaker 2: They're all same.

17:36 Speaker 1: And what angstrom parameter or alpha tell us about is the size distribution or what kind of particles dominated that particular atmospheric.

17:47 Speaker 2: Column where you are making measurement.

17:49 Speaker 1: And this is defined by this specific equation alpha, which is a measurement taken in two different wavelengths. So these are aerosol optical depth in two different wavelengths, wavelength one and wavelength two, and these are the actual wavelengths. So you take the logarithmic of the two and then ratio, that is what the alpha is defined. Typically, this alpha is always represented by a pair of wavelengths.

18:20 Speaker 2: So in ARONET, we.

18:22 Speaker 1: Define it for 450 to 870 nanometer. You can also use more than two wavelengths to calculate alpha by doing a least square fit. So there are many different ways. So again, whenever you are using external parameter, always ask the question, which two wavelengths are how many different wavelength it represents because the value will change depending on which two wavelength you are using or more wavelength you are using and their interpretation will also change. So let's assume we are using these two wavelengths 440 nanometer which is sensitive to absorbing aerosols in the low close to the UV part of the solar spectrum and 870 which is close to the near IR channels representing high.

19:09 Speaker 1: Now when you have the value of this alpha like greater than two, then basically it indicates that the particles in the atmosphere are dominated by fine-mode or smaller-sized particles. These are typically urban pollution particles like sulfate, nitrate, or fine smoke particles. When this value is near zero or less than 0.5, it typically indicates the presence of coarse-mode particles such as desert dust or soil, or sand particles in the atmosphere. Again, depending on the pair of the wavelength, this interpretation can change But the angstrom parameter is kind of a quantitative way in which we are using optical depth measured in different wavelength to assess the size.

20:06 Speaker 1: So the spectral dependency of aerosol or aerosol properties are quantified by this angstrom parameter and very, very useful in assessing or identifying type of aerosols in the atmosphere. Okay, now, how these things are linked or similar to aerosol optical depth or what, we call PM 2.5 in air quality community, how they are two quantities linked or what kind of differences they have or similarities they have. So let's look first the PM 2.5, right? PM 2.5 is typically measured near the ground or top of the building.

20:56 Speaker 1: And these are most likely point observations. The surface layer is the lowest layer of the atmosphere where we all breathe. Typically, its thickness of this layer can be few tens of meters, up to 100 meters, or few kilometers, depending on the mixing of boundary layer, how the boundary layer is mixed. So this surface layer can vary, okay? Now, satellite or even the ground-based aeronet aerosol optical depth, which we just discussed, represent the entire column of the atmosphere. So let's say this is a surface measurement of PM 2.5, but this is aerosol optical depth.

21:41 Speaker 1: whether it's come, if it is coming from a sun photometer like Aeronet, that it will be representing one point, point location, similar to PM 2.5. But if it is coming from the satellite, then it will have an horizontal resolution also. So that cylinder of the atmosphere, which extended from the surface to the top of the atmosphere, will also have a certain diameter or radius. And this resolution can vary from one kilometer to 10 kilometer depending on which satellite, what data product.

22:11 Speaker 2: You are using.

22:12 Speaker 1: So this is main difference between aerosol optical depth and PM 2.5 in terms of what they are measuring. Now there are other differences. PM 2.5 is a mass concentration and it is always representing micro-ohm per cubic meter. It's a dry mass, means it is measured under controlled conditions by heating the air sample at 40% relative humidity, 35 to 40%. Where aerosol optical depth is actually represent all the particles in this column, all size, all shape, all kind of chemical composition, everything which is in this column is represented by optical depth.

22:57 Speaker 1: It's a unitless quantity. Optical depth is a unitless quantity. And it is done in ambient conditions. So what I mean by ambient condition is we measure aerosol optical depth in the ambient condition, whereas PM2. 5 is in dry mass. Like I said, the sample is heated and then measured. So when you do the optical depth in ambient condition, quantities like water vapor and can have a direct or indirect effect on aerosol optical depth, right? So let's say you put the sulfate aerosols in urban environment with high relative humidity conditions, then the sulfate aerosols are hygroscopic in nature.

23:36 Speaker 1: They will grow in size and they'll absorb water vapor and grow in size and their optical properties will change and aerosol optical depth will increase. Whereas since we are only making dry mass of PM2. 5, your mass concentration may remain constant where aerosol optical depth can see the changes because of the high concentration of water vapor in the atmosphere. And there are other parameters which we already discussed. Although these two quantities have been heavily used in air quality research and application, there are significant differences.

24:13 Speaker 2: In both of them.

24:18 Speaker 1: But both of them, one thing important to note here is that both of them represent amount of pollution in the atmosphere. Yes, there are significant differences, but there are also similarities. Under certain conditions, under certain atmospheric conditions, with some assumptions, they can be highly correlated. And that is what we use, actually. If you have looked at any satellite-derived PM2. 5 product, those assumptions, those equations, that's what we use to actually derive surface PM2. 5 concentration using aerosol optical depth.

24:56 Speaker 1: But this is also, I just wanted to make sure that everybody understand the differences before we start using aerosol optical data for air quality application.

25:06 Speaker 2: Okay, so let's move forward. Let's get back to the Aeronet.

25:10 Speaker 1: And what we are going to talk now is what kind of instrument we use, calibration, which is a backbone of our Aeronet network, and some of the datasets which we produce using these instrument from all around the world.

25:25 Speaker 2: Okay, so the first one is the instrumentation.

25:31 Speaker 1: The fundamental instrument which Aeronet use is called sun photometer. A sun photometer is nothing but a light meter that measures the.

25:40 Speaker 2: Amount of light.

25:42 Speaker 1: Within a very narrow range of wavelengths or colors within a very small field of view. So as I mentioned earlier, this instrumentation head of the sun photometer what we call is collimator is looking or pointing towards the sun and it measures the incoming solar irradiance in particular wavelength at very small field of view 1.2 degree field of view once the light passes through this there are detectors different wavelength detectors which makes measurements and then we save the data the one thing to note down about this instrument is it's an automated system so it not only has this collimator and the sensor head where we record the data but it also has a robot and the robot is basically to track the sun So it tracks the sun from morning to evening using automated programming and then we have a control unit that's where we actually store the data and can program in different mode.

26:59 Speaker 1: Okay, so let's move on and learn a little bit more about this particular sun photometer which is used in Aeronet. As I mentioned in the beginning, Aeronet used the same instrument throughout the network. This is the latest version of the instrument of the Sun Photometer, what we call it Sun Sky Lunar Multispectral Photometer. It is manufactured by a company called CML Electronics.

27:29 Speaker 2: It is based in Paris, France.

27:32 Speaker 1: And the latest version of this instrument is called CE318T, more commonly known as T-Instrument or TCML. And this is, although we use the older version of CML, it's still in the network, but they are all calibrated with the new instrument. But since 2015, since the launch of the instrument by CML, Aeronet has adopted this new version and started implementing transitioning the entire network from the older version to the new version. About 60% to 70% of current network uses this new version of instrument, and the remaining 40% uses the older version.

28:19 Speaker 1: They are very similar to each other. They are only differences in their capabilities, but the data quality and the parameter which we get are the same. The new instrument is capable of not only making measurement of sun, but it can also scan the sky. And I'll talk about that a little bit more, how those sky measurements are useful. But the most important is that it can actually track the moon in the night also. So we are able to get not only daytime aerosol optical depth measurement by looking to the sun, but we are also able to get nighttime aerosol optical depth measurement by observing the moon.

29:06 Speaker 1: As I said, it's fully automated, very low power consumption. It can run on solar panel, although it has option to provide the power through direct wall power or we can use the solar power. The solar panels are being heavily used within the internet network, specifically in locations where we may not have continuous supply of electricity. We often use solar panels so that the operation can be continued. The daytime measurement, both sun and sky, and the nighttime measurement from the moon are all used to derive these aerosol optical depth, particle size distribution, number concentration, and water vapor.

29:56 Speaker 1: I'll talk a little bit more about the water vapor because that's one of the parameters which we get from Aeronet, which is also used to correct the aerosol optical depth, but we also provide that.

30:11 Speaker 2: As a product. Several models exist.

30:17 Speaker 1: As I said, this is a standard model, but there are For a specific application, there are other types of models available from Aeronet, including polarized. So people who want to use the data for other purposes to assess the polarization, there is a way to get polarized. There is a sea prism specifically designed to make measurement over ocean and lake. We'll talk about that more later. The instrument also come with flexible communication. It can communicate the data through satellite link, through internet by connecting to a computer, or it can also transmit data over cellular phone, cellular SIM using modem.

31:05 Speaker 1: The data are fully automated. We get the data in real-time process. If the location doesn't have any of these or problem with any of the communication, it has onboard data storage. It can store data for up to two years on an SD card. So for complicated or difficult location, remote location, data is stored on the SD card. And once in a while, site managers or PI upload the data to your internet server. Okay, these are different versions. As I was mentioning earlier, the one which we often use is called Teens2Main.

31:41 Speaker 1: It has these different channels, nine different lengths, starting from 340 nanometer to 1640 nanometer. All these channels have specific purpose in their own age.

31:57 Speaker 2: 937 channel is used.

31:58 Speaker 1: Specifically to derive the columnar water vapor and other channels, These UV channels are specifically put there to actually get absorbing property for aerosols, and the visible channel are most commonly used wavelength.

32:14 Speaker 2: There are other versions.

32:15 Speaker 1: Aeronet also use over some station polarized model. There are two other instrument which Aeronet use in their network called Sea Prism for ocean and for lake, and I'll talk about that a little bit later. If you want to know a little bit more technical detail, the link of user manual is here, which provides a lot more specific on the instrument. Okay, so the instrument operation, as you can see in this particular movie, it's very automated. It wakes up every 15 minutes, track the sun, makes the measurements in nine channels, and then go back on the, when it is resting, the collimator.

33:04 Speaker 2: Position is down, as you can see, and that is to prevent.

33:09 Speaker 1: Any water particles, any water vapors, or anything getting into the collimator or damaging the instrument.

33:15 Speaker 2: So this is always what we call is rest position.

33:21 Speaker 1: There are nine direct sun and four sky spectra channels. It takes about 10 seconds to make measurements. We also do a series of three measurements to actually screen the cloud. Whenever measurements are taken within one minute of interval, every 30 seconds, three sets of measurements are taken. Those three sets of measurements are actually take to what we call the triplet measurements. These triplet measurements, we take them to actually screen the cloud from the algorithm because instrument has no other way to actually detect the cloud or non-cloud.

34:00 Speaker 1: So these triplet measurements are often used in cloud screening algorithms to remove the cloud from the measurements. The frequency of measurement varies from 5 to 15 minutes depending on the sun angles. In the morning and evening, we have more frequent measurement. In the afternoon, we have less frequent measurement. And then similarly, we do the nighttime measurement, and we'll talk about that a little bit more. Okay, so in addition to those two direct sun and direct moon measurements,

34:39 Speaker 1: we also do these instruments also capable of making sky radiance in four spectral bands. uh those four spectral bands are basically 440 nanometers 670 nanometer 870 nanometer and 1020 nanometer currently we are operating in these four wavelengths but we are actually experimenting to expand it in uv channel as well allowing the so so these sky radians are along the solar principle plane which you can see on the first okay uh movie here and then here is uh what we call is Al McIntyre measurements.

35:23 Speaker 2: As you can see it actually.

35:25 Speaker 1: Tracks the sky in principal plane just looking so we can capture the different scattering angles and in the Al McIntyre you are actually looking scanning the sun along the sun angle, the entire sky from 0 to 180 degree back and forth. And these two measurements allow us to actually measure the scattered radiation, which is used along with aerosol optical depth measurement to derive the phase function, which is a radiative property or the optical properties of the aerosols and the size distribution and single scattering properties using a algorithm called developed by in a very early called Dubois and Nakazima inversion algorithm.

36:22 Speaker 1: So we use detection and sky measurements to actually get additional aerosol properties. And there we have several publication describing into much more details about this algorithm if you are interested. Okay, so let's move on to the calibration. Let's take a 20-second break here before we get to the next section. If you have a question, again, put in the question-answer box, and then we'll move on.

36:52 Speaker 2: To this next slide.

37:15 Speaker 1: Okay, so the next topic, which is most important for Eronet, its backbone of maintaining high quality data in Eronet network is calibration. Each Eronet instruments is calibrated using same method, same duration, as I mentioned earlier.

37:34 Speaker 2: So there are three major steps in calibration.

37:38 Speaker 1: What we call is pre-deployment calibration. Whenever either we buy instrument or anyone else who is deploying these stations, they can buy directly from CMLS. So one of the things I want to note down about Aeronet is anyone who is interested in deploying a station around their own institution or wherever their research interests lie, They can actually buy an Aeronet instrument directly from CML, and then NASA will calibrate for you free of charge. And your station will become part of NASA Aeronet network.

38:21 Speaker 1: The only policy is that you open your data to the public, and we process all the data in real time. So before, when you buy instrument from CML, before you deploy in the field or at any location, the instrument comes to either NASA or one of the other calibration centers. So if you can see in the map, there are a few other calibration centers all around the world. There is one which is started in Taiwan. We have one in the US. This is NASA Gardner. We have a couple of them in Europe. We also have some calibration sites, which may not be the calibration center, but they are calibration sites where the calibration data are gathered.

39:05 Speaker 1: One over is Hawaii, one over is Izania, another island, because we need very clean conditions to calibrate this instrument, and I will talk about that a little bit more. And in Australia, we have just another network. It's not a calibration site. So once the instrument, is purchased you do a pre-deployment calibration which is done either in one of those deep calibration centers then the instrument is sent to the field it is deployed it remains there for one years to one and a half year depending on the location and the purpose Typically, the deployment period for each instrument is about 18 months.

39:45 Speaker 1: After 18 months, the instrument come back to one of these calibration centers. We do the post-deployment calibration, and then the instrument is sent back again to the field for another 18 months or 12 months, depending on.

40:04 Speaker 2: Where you are.

40:06 Speaker 1: So this three-step process ensures that instrument we track the

40:15 Speaker 1: consistency of instrument over time, how the filters behaviors are changing, if there are dust or anything depositing in collimator or sensor head, if the instrument for whatever reason is declining in the performance, then those pre, post field calibration allow us to correct the data. This is very, very important for Aeronet to maintain high quality data. Okay, so in order to do the calibration, one of the things as you might remember from the equation of aerosol optical depth is the I0, which is extraterrestrial solar irradiance or top of the atmosphere solar irradiance in the particular wavelengths.

41:04 Speaker 1: And in order to calculate aerosol optical depth from the measured I value, we need this I0. And this, to derive this I0, we use a method called Langley method. And this Langley method required that the atmospheric turbidity or haziness or aerosol optical depth remains low and constant over time.

41:31 Speaker 2: Of the measurement.

41:32 Speaker 1: So to ensure that we actually make, set up these calibration sites in very clean condition with very, very minimal aerosol optical depth. Typically they are high altitude sites. So we have two major calibration sites, one in Mauna Loa Observatory in Hawaii, which is about 300, which is about 3.3 kilometer up in altitude and very clean conditions. Aerosol optical depth values are order of 0.0205, so very, very stable site. This has been used extensively since the beginning of Aeronet program for calibration all the So we have this something what we call it the master instrument.

42:22 Speaker 1: The master instruments are sent to the Mauna Loa on a regular basis. Once the instruments come back from the Mauna Loa to the Goddard or other calibration centers, they are inter-compared with the field measurements and then we transfer the calibration from the master instrument to the field instrument. And this process continues every six months. The master instruments are sent back to the Mauna Loa Similarly, we have another site, Izania in Canary Island in Spain, which is also an excellent site for the calibration.

42:59 Speaker 1: Okay, now let's, again, calibration has a lot more details. Again, our website, when we go to the part, in part two, next session, we'll discuss, I'll walk you through details, documentation on all of these specifics on the website which are available. Now let's talk about the datasets. So there are two types of datasets as we mentioned earlier. There are direct sun and moon measurement Like we mentioned earlier, we use that template to do the cloud masking, and then we use this VR Lambert law to get the aerosol optical depth.

43:41 Speaker 1: Once you have aerosol optical depth in different wavelengths, you can calculate angstrom exponent, or angstrom parameter, or angstrom coefficient, and then the water vapor is also calculated using 937 nanometer channel to correct the aerosol optical depth. The sky radians, which we saw earlier using Alkmaar-Kent and the principal plane measurements are used to input in an inversion algorithm by Dubois and Nagasima to actually retrieve or derive micro-physical and radiative properties. So micro-physical properties such as whether particles are sphere or not, and size distributions.

44:30 Speaker 1: Radiative properties include single scattering albedo, which tells about whether the particle is a scattering type or absorbing type, and then the imaginary and real part of refractive index, which is another parameter to define chemical compositions of the particles. Aeronet data, similar to satellite data, comes in different levels. Level 1, level 1.5, level 2.

44:57 Speaker 2: In terms of as people.

44:59 Speaker 1: Who are using satellite data, they are familiar that the time-to-time data are reprocessed using improved algorithms. So similarly, Aeronet also reprocessed the data and the current data are called version 3, which was released in 2018. Level 1 data are unscreened data, means it will have clouds and everything. Level 1.5 data are cloud screen and quality controlled. These are done using an automated algorithm. And level 2 data are quality assured. Level 2 data are only produced after the post field calibration.

45:34 Speaker 1: So level 2 data have a lead time of about two years after the measurements is taken and about a month after the calibration, post-fail calibration happens at one of the calibration facilities. In order to see the complete list of all the parameters which are actually derived or provided by Aeronet, you can see that link below, but these are major list of parameters which are used very often by the research community. Spectral aerosol optical depth means optical depth in different channels. Armstrong parameters using different combination of wavelengths.

46:17 Speaker 1: Colomer water vapor is also derived by aeronet. And then we have some of the inversion parameters like single scattering albedo, which tells us that the particles is absorbing or scattering size distribution. Refractive index tells you basically give indication of chemical composition by looking imaginary and real part of refractive index. We also provide absorption and extinction optical depth derived. Then you have phase function, fine mode fractions, coarse mode fraction, and so on and so on.

46:53 Speaker 1: There are like 50 plus different parameters which are provided by Aeronet. Okay, now let's talk about nighttime measurement. As I mentioned, this new generation of instrument, T instrument is capable of looking moon in the night and able to make nighttime aerosol optical depth measurement. Currently about 300 out of 600 stations, 300 stations have this capability to make nighttime measurement.

47:25 Speaker 2: The data are currently under.

47:28 Speaker 1: Research, I would say, and what we call is provisional data and only.

47:34 Speaker 2: Available at level 1.5.

47:36 Speaker 1: We don't process this data to level 2. Aeronet team continues to refine and improve the algorithms. And one of the challenges is the cloud screening during nighttime. So our team is continuously working to develop new cloud screening algorithm to check the consistency of this nighttime data with the daytime data. data and using some of the LIDAR measurement during nighttime data to validate. Once we are assured about the quality, the provisional status will go away and the data will become regular level 1.5 data and then will also mature to level 2 data after the full calibration.

48:20 Speaker 1: Again, this link provides actually documents which describe very specifically detailed description how the lunar algorithm works.

48:35 Speaker 2: Okay, so in the beginning.

48:37 Speaker 1: I mentioned a little bit, Aeronet not only have this permanent Aeronet stations on the land, but it also deploys sites during field campaigns and do the nighttime measurement, but there are more than that. We also have a couple of other components of Aeronet. One is called Aeronet OC. So the Aeronet OC is Ocean Color Monitoring Network. And this is part of Aeronet. It uses one of the single sun photometer models, which I discussed earlier, called Sea Prism. It is specifically designed to make measurement in ocean to actually get the ocean color there are so this is the current sites locations these sites are typically either inland which are on lake to get actually the lake water measurement, or they are on a platform in ocean or on the coastal sides.

49:48 Speaker 1: Okay, so how does this work? The Aeronet OC is not only capable of looking to the sun and the moon, but it also has capability to look down to the ocean, to the water surface, and make measurement of what we call water leaving radiance. As you can see in this picture, the water leaving radiance is very, very important when you are trying to validate the satellite data specifically designed to observe ocean color or to get properties of how the solar radiation.

50:29 Speaker 2: Interacts with the water surface.

50:33 Speaker 1: In very early, in like 2000, Aeronet OC started and became part of Aeronet Network. Since then, a lot of things have changed, and the network has grown significantly. These are some of the locations of the sites. We have about 20, 25 active sites. 20 sites are over ocean. Five are lake sites. and then some of the sites are decommissioned during different time of the measurement. One of the thing I want to note down, these aeronet sites are most difficult sites to maintain and do the measurement.

51:20 Speaker 1: As you can see, these platforms are often in the middle of the ocean, far away from the coast. They are in difficult oil and gas platforms or lighthouse or various kind of platform which are in the ocean and they are often experience extreme weather conditions continuous bombarding of sea salt particles winds and all kind of challenging environmental conditions so to maintain and make continuous measurement from these locations is very very hard so this although it is making actually measurement from about 20 such locations this data are very, very valuable, otherwise not available from anywhere else.

52:12 Speaker 1: One more thing I want to mention here about this particular network is that although NASA manage this network, but each site are actually managed by site managers or PI of that particular site and they belong to international institutes. Sometimes the instrument is also purchased by those institutions, and sometimes it is provided by the NASA. But overall, the whole network is actually benefiting the global community on the ocean color measurement, and that is maintained by the Aeronet. As you can see in these pictures, some of these sites, you cannot reach them directly through ship.

53:00 Speaker 1: Sometimes you have to use helicopters, sometimes you have to use the boat, and many, it's extremely difficult stations to actually make the measurements. We are really lucky to have these 2025 sites around the globe to get this valuable data sets for many satellite validation, including the PACE mission, which was just launched by NASA recently. Okay, now another very useful network or component of aeronet is called Maritime Aerosol Network or MAIN.

53:37 Speaker 2: The MAIN is.

53:42 Speaker 1: Done through the sun photometer which is commercially available called MICROTOPS2. And this MICROTOPS is a handheld sun photometer. works in four to five different channels, depending on the configuration. It provides ship-borne aerosol-optical depth measurements using the Sunvoltimeter, and depending on who is buying, it can become in two different combinations. One is with 340 nanometers, another one is one longer wavelength at 500 nanometer. It In addition to this microtops providing aerosol optical depth, it also has inbuilt temperature and pressure measurements so that we can get the other component of aerosol optical depth as we mentioned earlier.

54:31 Speaker 1: You can connect this microtop using a GPS, which enables the accurate location and time.

54:38 Speaker 2: Of the measurement.

54:40 Speaker 1: All these microtops, we have about 20, 25 microtops at Aeronet here at NASA, which are all distributed and provided to anyone who is interested in making ship bond measurement through a research vessel, field campaign, or through a commercial, or through a private ship cruise visit over ocean. So we provide to all kind of people all around the world. Once the cruise or the ship measurements are completed, the instruments come back to NASA Goddard. We compare and calibrate against the master CML instrument, sun photometer.

55:18 Speaker 1: So the data quality comes from these microtops are similar to what we get from the CML sun photometer. Although because of the limited channels and limited capabilities, it only provide aerosol optical depth and few other parameters, it doesn't provide all range of parameter which we get from the main aeronautic information. Since 2004, we have deployed these periodically on ships, research vessels all around the world. And this data provides an alternative source of information data sets for our ocean to validate satellite retrieval and model outputs.

56:00 Speaker 1: These are some example of different parameters which you get from main network from Microtops. You get angstrom parameter, AODs in fine and coarse mode, and then you can also get the precipitation water. Here is a bigger map showing Locations of this measurement you can see over last 20 plus years since the existence of main network. It has covered large part of global ocean and otherwise these measurements are not available to validate satellite data. So very, very important data sets comes from Aeronet, from main networks.

56:44 Speaker 1: to provide valuable information over oceans, which is otherwise not available from any other ground source.

56:52 Speaker 2: Here you can see some pictures.

56:55 Speaker 1: This man is actually a great example of collaboration internationally because indigenous people are spending their own time and effort to make this measurement and NASA is just providing the instrument and the data processing and calibration capabilities.

57:11 Speaker 2: But many people, students, researchers, individuals.

57:16 Speaker 1: Doing their family trips, makes all the time these measurements and help us build this huge data repository at NASA, which is benefiting global community. Okay, so let's talk quickly about some of the applications I mentioned many of them throughout the presentation, but specifically emphasize that aeronet data are heavily used to validate the satellite data, satellite retrievals. More recently, air quality application has started using aeronet derived product. For a long time, to address the implication or role of aerosols in climate change research, heavily relied on aeronet because of the high accuracy of the data and long terms of data records.

58:11 Speaker 1: More recently, solar industry also started aeronet data to actually account for aerosols particles in the atmosphere and how they actually can prevent solar energy production in different parts of the world, specifically in dust-prone regions.

58:32 Speaker 2: This can become very, very useful.

58:36 Speaker 1: Here are some examples of how we are actually supporting various satellite missions and their validation efforts. The first one is NASA's most recent air quality missions called TEMPO, Tropospheric Emissions Monitoring Pollutions. which was launched last year's and data became available earlier this year's. This is a part of a global constellation of three air quality satellites called GEM, Sentinel-4 and Tempo, which will be flying over U.S., Europe, and over Asia. In Tempo field of view, Aeronet has about 160 active sites covering actually out of 50, 37 states and corners.

59:20 Speaker 1: And then more recently, we have actually put 10 sites in Minority Surveillance Institute to bring minority students and faculties to the mainstream NASA research. These 10 sites are set up by NASA through a ROSES competition. PACE is another recent NASA mission which is focused on ocean colors and aerosols. Aeronautics is heavily involved in validating both ocean color and aerosols.

59:53 Speaker 2: From the base.

59:54 Speaker 1: And then we have another upcoming mission which is anticipated to launch in the next two years called MAIA, Multi-Angle Image for Aerosols. It is specifically designed to get PM2. 5 and speciation to assist the health impact of aerosols particles. and it will make measurement or specific location all around the world. And Aeronet is a critical part to validate these datasets. We have set up several sites around the world to actually specifically validate and collect data for Maya validation.

1:00:34 Speaker 1: As many of you know, NASA is building an inclusive open science community through various program. to actually foster greater collaboration in research and scientific investigations by removing barriers. So, Eronet stands on the foundation of open data policy. As I mentioned from the beginning, one of the Eronet data policy is to make the data open so that the global community can use. We have done an excellent job since the beginning of the program to make the data available through the web portal.

1:01:10 Speaker 1: But there's also opportunity to expand it to reach an impact to the larger community by providing data in user-friendly formats, by providing training, making more relevant data products for a quality community, for example. And partnering with DARSET is one of the examples or attempt for us to actually reach to the larger user community. Okay, so I think that was all about the Aeronet and different components. So I'm going to quickly summarize what we learned so far. As I mentioned many, many times, Aeronet has produced 30 years of atmospheric aerosol measurements and we are continuing.

1:01:57 Speaker 1: We have about 600 active sites all around the globe. Aeronet continue to provide high quality data for various application, which includes satellite validations, and including sensors starting from MODIS, OMI, TEMPO, Maya, which are more recent, VIRS, and then upcoming sensors like AOS and Maya. Now, Aaron, it is also heavily used to validate, evaluate, validate, and assimilate into models, including majors model like GEOS, and then UKMET, ECMWIF, NAVY model, and many others. Eronit also has been used to assist the reanalysis products, such as MERA, MERA2, or European ERA.

1:02:48 Speaker 1: And then, One of the focus of Aeronet is to do synergy measurements. So often, Aeronet alone does not provide complete pictures of atmosphere. So we are also co-locating our instrument with MPLNet, Pandora, Tolnet, which you will hear in following session of this webinar series, and then other networks like SPARTAN, Skynet, improve to assess their quality. The Dragon networks are very, very useful to get high spatial and temporal resolution data, which are often deployed to support various NASA and non-NASA field campaigns.

1:03:30 Speaker 1: And then Aeronet OC continue to support water-living radiance, which serves as a primary tool for validation, including the PACE mission. We also have quarterly newsletter, which you can access through this QR code. And that's all I have. We are looking for part two. In part two, we will go over some of the application example. We will get into the website. browse through different parts of the website. We look into how to download the data, how to visualize the data. We also have some Python Jupyter Notebook, which we will play with the Google Collapse.

1:04:19 Speaker 1: If you want to actually work along with us, please make sure you have a Google account. And that particular Python Notebook will also include an exercise to how we are using aeronet data to validate the satellite retrieval of aerosol optical depth. So that is for all the part two. And one of the most important thing as I have been mentioning from the beginning is aeronet is a Effort of international collaboration. Although NASA takes the leadership in managing the networks and providing a lot of infrastructure, but it is a true collaborative works from hundreds and hundreds of people from all around the world, coming from many, many institutions, universities, agencies, government, non-government, private and public sectors.

1:05:16 Speaker 1: This picture just shows our network team here at NASA Goddard. We have Aeronet site PIs and site managers who actually manage and ensure that Aeronet instrument is working fine and operating and sending the data. Our funding comes from what we call US validation program under NASA from NASA headquarters. So I would like to acknowledge and thanks everyone actually who directly or indirectly contribute to the whole Aeronet program. because without support of collaborators and global team, this would have not been possible.

1:05:57 Speaker 1: So I would like to thank everyone for that. And I think that's all.

1:06:01 Speaker 2: I will.

1:06:04 Speaker 1: Move over to Carl for homeworks and other logistics for this webinar series.

1:06:13 Speaker 2: Thank you.

1:06:18 Speaker 3: Thank you very much, Dr. Gupta, for that introduction to the Aeronet Network. Just a reminder about the homework and the training certificates. The homework will be issued at the end of the last session of this training. That's on August 22nd. And the certificate of completion will be handed out about two months after the end of the course for those who attend all five of the live webinar sessions and complete the homework assignment by the deadline. This slide lists contact information for myself, Dr. Gupta, as well as the RSET program and our sister programs.

1:06:51 Speaker 3: We also encourage you to join the Aeronet mailing list by emailing the email address listed here, aeronet-join at lists.nasa.gov, if you're interested in receiving news and updates about the Aeronet network. Finally, here's a link to the Aeronet website. And thank you very much for your attendance and attention. And we'll be moving into the question and answer portion now.

1:07:23 Speaker 2: All right. Yes. Sorry, go ahead. Okay. Yeah.

1:07:27 Speaker 3: Well, thank you. We'll get through as many questions as we can. Any questions we don't answer live will be collected in this document, which we'll post to the website later. I will note there were several questions about sharing the slides. Those will be posted on the website along with the recording of this training, so you can go back and look at things you might have missed. And I will note that we had several questions related to climate change. If you're interested in that, that's not the focus of this presentation, but we have several other RSET trainings dealing with climate and resilience.

1:07:57 Speaker 3: So we put links to that as well as the U.S. Global Change Research Program website for those asking questions about the U.S. All right, so we can... Take it away with the other questions. I don't know if, Pawan, you had specific questions you want to answer. Otherwise, we'll just go through. Pawan Kharatekaraman Sure.

1:08:15 Speaker 2: Thanks, Carl. So what we're going to do is I'll.

1:08:20 Speaker 1: Pick some questions and answer them because I noticed a lot of questions earlier were actually already answered in later slides, and many of them are actually also going to be answered in part two of the presentation. but we will answer some of the important ones. And just want to note down that I have my colleague Thomas Eck from Aeronet team also in the panel. So I will direct some of the question to him who has a lot more experience with Aeronet. He is associated with Aeronet from the beginning of the program and he can provide you more insight.

1:08:59 Speaker 2: Perspective.

1:09:04 Speaker 1: Okay, so let's see. Okay, let's go to the question three.

1:09:13 Speaker 1: So Carl, are you going to read them?

1:09:15 Speaker 2: Yeah, sure.

1:09:16 Speaker 3: I'll read them. So the question is, what's the distance between the instruments in the Dragon campaign? And this is related to evaluating the MIAC data product, which has a one kilometer resolution.

1:09:27 Speaker 2: Yeah, so that's a great question.

1:09:28 Speaker 1: So the Dragon deployment distances are not fixed.

1:09:35 Speaker 2: It depends.

1:09:36 Speaker 1: It varies for every single deployment. And it depends on what is the scientific objective and measurement requirement for a particular field campaign. It can be a few kilometers to tens of kilometers.

1:09:50 Speaker 2: It really varies.

1:09:52 Speaker 1: The one of the densest Dragon deployment we have done was in Baltimore DC area in 2011 as part of Discovery Equifield campaign where we have 40 plus sites actually ranging from DC in the DC Baltimore metro area. So that's one example of the dense network.

1:10:15 Speaker 3: Okay, thank you. I think we can scroll past. Some of these were answered in the questions. There's another question about Dragon. In terms of the difference between active and passive remote sensing, basically passive remote sensors rely on an external source of energy, usually the sun. Active remote sensors provide their own energy, and we'll be learning later in this training in other sessions about some active instruments as well.

1:10:40 Speaker 2: We can go to the question eight. Mm-hmm.

1:10:44 Speaker 3: So what is the region of the world that needs the sun photometers, for example, African deserts or the Mediterranean region?

1:10:51 Speaker 2: Yeah, so that is correct.

1:10:53 Speaker 1: We have Africa, Mediterranean, and Asia where we have less dense network. And again, we'll go through all this a lot more in detail.

1:11:02 Speaker 2: In part two.

1:11:03 Speaker 1: And if you look at the map, you will see the U.S. and Europe are most densely networked currently.

1:11:10 Speaker 2: All right.

1:11:12 Speaker 3: Maybe there's a question about, again, about the maps. Maybe the question 11 about what are the constraints or requirements for establishing Aeronet sites?

1:11:22 Speaker 2: Yes.

1:11:23 Speaker 1: So there are a lot of requirements, which one is it required a lot of resources to set up and maintain an Aeronet station. There is initial investment from financial perspective, but there is also commitment from the institutions who are hosting the instruments because instrument required regular checkups, maintenance.

1:11:45 Speaker 2: The instrument has.

1:11:46 Speaker 1: To come to one of the calibration centers every 12 to 18 months.

1:11:51 Speaker 2: So we need to facilitate that part also. So there are several requirements.

1:11:56 Speaker 1: Again, when we go to the part two, I'll talk about those a little bit more. And anybody who is interested in learning more, please email me and I should be able to provide you more specifics.

1:12:09 Speaker 3: Okay, thank you. So there's a question about urban versus high altitude sites. They're different. There's no necessarily a preference between them.

1:12:21 Speaker 4: Yeah.

1:12:21 Speaker 2: Yeah, both are valuable depending on the purpose and what you're looking for. We don't have any specific preference.

1:12:30 Speaker 3: Okay. And we're not aware of any data sets or campaigns that look at AOD at the meter scale resolution. As Pawan mentioned, the Dragon networks are pretty much, for Aeronet at least, the highest resolution spatial scales that were explored. And then all the methods that Pawan presented on are detailed in a lot of different publications, and those are listed on the Aeronet website for people to look at. Another question, question 15, about the number of sites in India specifically. Again, next week we'll talk about how to go to the Aeronet website and get a list of active sites there.

1:13:06 Speaker 3: And you'll be able to use that to determine where the active sites are. How about question 16 on the level of accuracy? So Aeronet doesn't measure trace gases. We'll learn about the Pandora instruments, which you do in part three. But what about the accuracy for aerosols?

1:13:24 Speaker 1: So the aerosol optical depth, which is fundamental parameter which Aeronet provides, we provide the accuracy of order 0.01 in the visible and then around 0.02 in the visible. in the UV part of the solar spectrum. And I think, let me ask Tom if he wants to add more on that.

1:13:48 Speaker 2: Tom, do you have more? Yes.

1:13:52 Speaker 4: There's also parameters such as absorption of the aerosol. And we measure that to an accuracy of single scattering albedo is the parameter for absorption. And we measure that to 0.03. It's an uncertainty of the retrieval. And that's only for optical depths that are moderate to high, so greater than 0.4 at 440. And then we measure size distributions quite accurately for fine mode, probably better than about 0.01 micron radius uncertainty for fine mode particle radius. So that's all.

1:14:36 Speaker 2: OK. Thanks.

1:14:37 Speaker 3: Thank you. And then a question about ensuring the accuracy that's done through the calibration, which was discussed in the presentation. In terms of the operational timeframe for the instruments, I don't know if you want to answer that.

1:14:51 Speaker 1: Yeah, so the instruments, depending on the location, it operates typically throughout the year. If it is in polar region or high latitude, depending on how much sun is available, If there is snow and ice on the surface and the instruments are covered, then also it can stop the operation. And during monsoon season in some parts of the world, if there's continuous rain, then we don't have measurements. Sometimes there can be gaps when the instrument is getting calibrated, but otherwise it's year-round measurement.

1:15:32 Speaker 1: And for calibration, we work closely with the site managers and site PIs from individual institutions to get the instrument back to one of the calibration center and then we send them back. Sometimes we send one swap out of the instrument in advance so that we don't have any measurement gaps. So there are a lot of combination which works and every site is unique in that perspective. Okay.

1:16:03 Speaker 3: Maybe question 20 is about what the wavelength that these AOD measurements are taking. I know there were several questions related to, you know, why do we measure at 500 and 550 nanometers specifically?

1:16:15 Speaker 4: Okay.

1:16:16 Speaker 2: So let me say this, and then I'll ask Tom to add more.

1:16:20 Speaker 1: But I think one of the reasons is 500 and 550 nanometer is kind of a wavelength where solar energy peaks. So if you look at the solar spectrum, that's where the the green bank, that's where the solar spectrum peaks. I think probably that is the reason where most models, calculations, solar energy budgets are using to actually assist the impact of aerosols on different climate and air quality using that particular web.

1:16:52 Speaker 2: Tom, do you want to add anything on that?

1:16:54 Speaker 4: And then most satellite retrievals are made at 550 nanometers.

1:17:03 Speaker 4: The ARINET measures that 500, and we can interpolate spectrally between the 500 and 675 nanometer data from ARINET to compare it to satellite data for validation of satellite data.

1:17:21 Speaker 3: Okay, thank you. Question 21, again, deals with where the technical information can be found. That's on the website. Question 22 is, again, about quality control. That was discussed in the presentation. Question 23, I think we kind of answered that already. Why do we measure certain spectral channels? Question 24, and I think there was another question later, is about the top of the atmosphere solar radiance. I know that during the slides it was mentioned that the Langley method is used, if you maybe want to elaborate on that a little bit.

1:17:53 Speaker 2: Tom, do you want to take it?

1:17:56 Speaker 4: Okay. We calibrate our master instruments for ARENET at Manalo Observatory in Hawaii. And so it's a high-altitude station in the mid-Pacific, and the optical depths are only about 0.01, and the stability of the atmosphere is exceptional in the morning. So we get highly accurate calibration from the Langley method, and that accuracy is on the order of a quarter of 1%. So this is why we calibrate there. It's unitless. It's top of the atmosphere signal for that instrument. And so because AOD is the unitless quantity also that we measure.

1:18:42 Speaker 2: Okay.

1:18:44 Speaker 3: Maybe question 25 is a question about capturing local sources with Aeronet. I know this is is not necessarily what they're designed to do. I don't know if you want to discuss a little bit about the limitations there.

1:18:58 Speaker 2: No, I think that's a great question.

1:19:00 Speaker 1: So, it all depends on what is in the path of Sun and Aeronet instrument, right? So, and it's point measurement, like satellite data are typically averaged over certain horizontal space. So, they cover certain footprint on the Earth's surface where Aeronet is much more like a point measurement. But data Aerosols are often homogeneous in a small spatial scale. So if there is a constant source of local emissions and if it is pumping aerosols in the atmospheres, then it will be measured by aeronaut.

1:19:38 Speaker 1: It really depends on the strength of that source, whether we will be able to capture it or not, I think. And then like mentioned here, wind directions and other factors can play a role there.

1:19:53 Speaker 3: Okay. Questions 26 and 27 are also about the specific wavelengths used. I think we addressed that. It looks like also question 28. Question 29, it's about the shape of the Earth. I don't know. It doesn't really affect things so much, for Aeronet at least.

1:20:17 Speaker 2: That is correct. Yeah.

1:20:19 Speaker 3: So question 30 maybe. So we know that there are, or there's evidence that PM1 is very closely related to health effects. Is there any relevance in the data collected by Aeronet for inferring PM1 rather than PM2. 5? Do you know about that?

1:20:38 Speaker 2: I think, let me refer to Tom.

1:20:40 Speaker 1: Tom, do we have more sensitivity towards very fine particles, ultra-fine particles?

1:20:45 Speaker 2: Yes. Tom, can you hear?

1:20:57 Speaker 3: I'm sorry.

1:20:58 Speaker 4: I was muted. We have more sensitivity to the size of fine-mode particles in our size distribution retrievals. In terms of AOD, we have equal sensitivity to both fine and coarse-mode particle sizes. So it depends on what product's being used. If it's a retrieval of the size distribution or whether or not it's the AOD that's being used. But we do have a tremendous sensitivity to fine mode particle, both optical depth and size.

1:21:38 Speaker 3: Okay, thank you. And I guess just a reminder, you know, PM1 is a mass measurement. Aeronet's collecting optical depth, which is not a mass measurement. But as Tom said, Aeronet is sensitive to those fine mode particles that contribute to the PM1 mass. How about question 31? What about polar regions where there's a very shallow angle to the sun? Does that affect, how does that affect the measurements?

1:22:04 Speaker 2: I think that is again Tom's question.

1:22:08 Speaker 4: We have actually greater sense, greater accuracy in the polar regions because of the large path length of the atmosphere. So the uncertainty in aerosol optical depth measurement is proportional to one over the path length or one over the air mass. So we actually have more accurate data in the polar regions. Even though the optical depths are typically low there, there's the ability to measure with higher accuracy in polar regions.

1:22:49 Speaker 3: Okay, thank you. In terms of calculating PM 2.5 from AOD, in the introductory training, we recommended as a prerequisite for this. We do talk about that a little bit. I don't know if you want to add anything to that now or just leave it at that.

1:23:06 Speaker 2: Yeah, I think so.

1:23:07 Speaker 1: I don't know if that, we do have some old ARSA training we did a number of years ago focused specifically on calculating PM2. 5 from AOD, so I would like folks who are interested can explore some of the older trading, I think, before 2000.

1:23:29 Speaker 3: Okay. So we'll put a link to that in response to this question. And again, the document's going to be posted to the website after the training. We're a little bit past the scheduled dead time. We'll go, I think, until 45 minutes past the hour, so another 10 minutes about And again, any questions we don't answer live, we'll write up the answers and post the document to the website. Here, question 33 is another question about Dragon. I think we basically addressed this before, but just, you know, what's the advantage of this high spatiotemporal network and what's the comparability to satellite footprints?

1:24:06 Speaker 3: I don't know if you want to add anything to that or just move on.

1:24:09 Speaker 1: I think, yeah, the bottom line is if you want to capture the small scale spatial variability, specifically in urban locations, which can be high depending on different road system, different traffic pattern, different industry locations.

1:24:31 Speaker 2: And some of these can be very useful, actually.

1:24:33 Speaker 1: So I think that is one of the main purpose to deploy this. And they can also be very useful because when you're comparing with the satellite data, often satellites one pixel or multiple pixel of satellite data compared with one particular aeronet locations. But in this case, this gives you an opportunity to average several aeronet locations within the footprint of satellites. So you have a little bit better measurement or representation of satellite footprint from multiple aeronet sites rather than one aeronet site.

1:25:08 Speaker 2: So I think that helps.

1:25:12 Speaker 3: Okay, great. Question 34, in the images of the Aironet instruments, there were two collimators. It looked like there were two collimators. So the question is, what's the purpose of the second collimator?

1:25:24 Speaker 2: Tom, do you want to take it?

1:25:26 Speaker 4: Sure. There's two collimators because there's essentially two separate instruments. In one, there is a silicon detector for most of the channels from the UV and up to 1020 nanometers, and silicon cannot detect higher than 1020 nanometers or 1050 nanometers. So then the second collimator is for the longest wavelength, which is the 1640 nanometer wavelength, which we have an in-gas detector, Indian gallium arsenide detector. So it's a separate detector and a separate filter that's measured and the second collimator, and that cannot be measured by the silicon side.

1:26:15 Speaker 4: So essentially you have two instruments in one sensor head, one for the bulk of the wavelengths and the other for the 1640 nanometers.

1:26:26 Speaker 3: Okay, great. So question 35 we answered already. Question 36 about applying for more sites. I think you can reach out to Paulan or the other Aeronet leads with requests for setting up new sites. Maybe question 37 about the standard working schedule of Aeronet. How do you decide on the pointing angles and the schedule for the different measurement modes?

1:26:53 Speaker 2: I think let's let Tom answer this.

1:26:58 Speaker 4: Okay. The schedule is dependent upon solar zenith angle, and the instrument calculates the solar zenith angle based on the latitude-longitude that's entered into the instrument and based on the accurate time it gets from the GPS. So solar zenith angle is used to measure through the Langley period in the morning, which we start measurements at air mass seven, seven path lengths through the atmosphere. And then we go through the entire day based on a predetermined schedule. The AOD measurements are made every five minutes.

1:27:41 Speaker 4: and the sky scans are made every hour. So we have Balmacanthar scans and hybrid sky scans that are made hourly, but direct sun measurements of AOD are made every five minutes.

1:27:55 Speaker 3: Okay, thank you. Question 38 about the Angstrom parameter. Can it have negative values?

1:28:03 Speaker 2: I think yes.