 Hello everyone and welcome for today webinar with Veronica. For myself, I will be moderator for the webinar and the presenter will be Veronica Escobar-Ruiz. First of all, I will give you a brief introduction about EGU-GI Early Career Scientist Committee in the European Jewish Science Union. Well, actually what we are doing here is to work with the young scientists and make their work and of course more visible in the science and in the Jewish science and instrumentation part of the GI division in the European Jewish Science Union. We have a committee for GI division, but here we have Early Career Scientist. We are just 2% for now. Myself, I'm Veronica, excuse me. So what I'm doing, I'm a postdoctoral research fellow in Gustave Eiffel University in Paris. I'm working in infrastructure assessment, including non-destructive tests and data analysis. Veronica, I will give her to talk about herself and her background. Also, later we will show you some interesting sessions and special issues after Veronica presentation. So, Veronica, the stage is for you now and thank you for your presentation for today and availability, actually. Yeah, thank you, machine. Thank you for organizing this and giving me the opportunity to present this webinar. A little bit about myself. I am working as a research support scientist at the University of Reading in the meteorology department. My background is a multidisciplinary background. I did a PhD in physical geography, specifically in hydrology. But I have a master and a bachelor's degree in electronic instrumentation engineering. Then this has allowed me to incorporate both knowledge in the areas of physical geography. And at the moment I'm working with projects related to runoff, solar ocean, radar characterization of soil moisture, forestry and canopy, and a little bit of the atmospheric electricity. And then that's it. That's about myself and I will now present the webinar. That's okay. Well, the idea of this webinar is to show you how I developed two instruments and I wanted to do this for the early career scientists that are here in this webinar. Because it's important to see the tools I use and the hard words I use to in order for you to implement it in your research for for your projects. About the outline, I will, you know, cover a little bit of the theory, radar, backscattering and polarization, synthetic upper two radar, and the technique or the technology we use at the University of Reading, which is a demographic profiling. And for each project, sorry for each instrument, I will explain the aim of the instrument, the instrument development and the application of this instrument. And a little bit of, I won't call it preliminary results, I will call it just like a real mapping for you to see what we have done with the instruments. Verinika, before, before starting, I will just comment on the question and a section. Sorry for interrupting. No worry. So for for the attendance, if you have any question during the presentation. If you have any doubt, you can just put your question in the section below. It's a question and a Q&A. So, final in the in the last part of the presentation you will answer. Verinika will answer your questions one by one. So please use this Q&A. And thank you. Thank you very much. Go ahead. Okay, well, what's radar radar in comes from radio detection and ranging and basically uses electromagnetic waves to determine the distance velocity or angle of the scanning objects. The magnetic waves are consists of a couple of forces, one electric and one magnetic, which are 90 degrees of one each other, and they propagate transversal to the, sorry, they go transversal to the direction of the propagation. In this example, or in this figure we see the electromagnetic field perpendicular to the incident plane. And when we, when you have that type of propagation you call it, it's horizontal polarized. If now we think that this field is rotated 90 degrees and now it points in which the magnetic field it is, you call it is vertical polarized. And this is important for a certain analysis would do with the radar. But also the, the, the frequencies or in order to call it a radar and electromagnetic waves they need to run certain frequencies between 30 hertz to 300 gigahertz. And the bandwidth we are focusing. It's the, the band C, which is, you know, between four gigahertz to eight gigahertz. And basically to basic radar system consists of transmitter receiver. What's back scattering and polarization. Back scattering is the future collection of the radio signal and allow us to see, for example, proofness of the soil volume and certain materials that will give us an idea of a man main structure. The polarization signal I was talking about, you know, like the vertical polarization and the horizontal polarization allows to identify features or areas that the polarize the signal. In this, you know, table, you can see that to see the roughness of the surface, you can have transmitter in the vertical polarization and receiver in the vertical polarization to then that will give you an idea of the roughness of the surface. In terms of the double bouncing, which allows you to see the material or, you know, certain surface like buildings, you use a transmitter in the horizontal polarization and the receiver in the horizontal polarization. To see volumetric and specific specifically for vegetation, you need to use both type of sources, horizontal polarization, transmitter and a vertical polarization receiver or the vertical horizontal. Then as you can see that's the idea of using different polarizations. This video shows you how synthetic upper to radar works and I really like it. It's a very good example and imagine you have an aircraft flying above the ground at a certain distance. And then you send these electromagnetic waves to the ground and by this Doppler effect you will receive the signal back. If you were only going to analyze one of the scanning surface you will see a blue point. It's until you send all that train or don't you analyze all those pulses or those kinds of pulses that you can reconstruct the signal. As you can see, synthetic upper to radar is a technology to reconstruct objects. One of the main characteristics like a site looking antenna. If you were having an antenna looking at the nadir position, you will talk about zero degrees. But if you had it as a certain degrees, you talk or you say it's a site looking antenna. And maybe you have here about this site looking antennas in the satellites at 40 degrees 30 degrees. But that's mainly the characteristic of the of the SAR. But also, in order to observe the high or of the objects, you need to apply a technical interferometry, but the disadvantage with the SAR is that you need at least two observations of the same target. The technology that we use here at the university already is calling tomography profiling and basically gives you vertical backscattering profiles. The main characteristic is like it has an antenna along the scan direction. In this figure, if you imagine that the scan direction is going to the X axis, then the antenna will be pointing or at the nadir position or at angle, but in the along the scan and with a single pass that it will give you a idea or information of the height of the objects you are scanning. If you want to know more about this technology, I will recommend you to read more reason and Bennett 2014 it explains a imaging processes and the resolution and compares it with the SAR technology. Well, I will talk now about the drone based SAR system. This is a project between the University of Reading and the National Physical Laboratory. The aim of the project is to estimate forestry carbon stocks using another radar drone system, and it has three main objectives. One of them is map the radar backscattering in a high detail through a forest canopy at the individual tree level. The second one is to quantify uncertainties associated with current and near future satellite techniques. Finally is the characterization of biomass from translated object radar signatures to forest in conjunction with high resolution 3D LiDAR maps. And as you can see what we want to do is scan individual tree levels, quantify uncertainties compared with the current satellite, and also compared with or this high resolution technique with a 3D LiDAR maps. The study area, we are aiming to fly in these within woods. Within woods are located in the south east of UK in Oxford. They are dividing three areas. One of them is an Asian semi-natural woodland that is 300, 400 meters by 300 meters. And basically the most important thing to show here is that some of those trees date back to the last ice age. Then that gives a very good estimation of the forestry canopy stock sort of. Okay, but how did I develop this system? Basically consists in six devices. We have, of course, a drone, a differential GPS, a pair of antennas, actually three antennas, a radar, something called a low range frequency, which allows me to do some remote communication and a Raspberry Pi. The drone we are using is from the DJI company. The name is Matrix 600 and has a maximum speed of 60 kilometers per hour, but we are actually flying in a very slow speed. It has a payload of six kilograms. Our system actually weighs two kilograms and then allow us to have a very good health hour of flying. And we have at least three sets of batteries, then we can have one and a half hour of flying. The drone comes with an integrated GPS with an accuracy of 50 centimeters in the vertical position and 150 in the horizontal position. And as you can see for us to scan individual trees, it's not very accurate. However, we purchased a differential GPS. This has a centimeter accuracy of two centimeters in the vertical position and one centimeter in the horizontal position. And in order to fly or to create flight routes, we use the DJI pilot software. And here is an example of how we are aiming to fly the drone. It basically takes over here and follows like a grief path. And every time we want to scan an area, we will do it remotely. The radar we are using is an anchor tech radar has different waveforms and options. The one we are using is the south tooth waveform has a frequency between 5.2 gigahertz to six gigahertz and a bandwidth of 800 megahertz. The bandwidth is the difference between six gigahertz and 0.2 gigahertz. It has different sweep times between 0.25 milliseconds up to eight milliseconds, different scan time scan 10 and 30 seconds. We can use different receiver channels, one or two, and also has the number of samples of sweep changes. Of course, the bigger it is, the more data you're getting, but the more time you will take the radar to obtain this data. The advantage of this radar is that we have a USB communication and this allows us to connect it to a Raspberry Pi. In terms of the antenna array, this antenna is a patch antenna has a range of the same as the radar, 5.2 gigahertz to 6 gigahertz, but it has a bandwidth of 400 megahertz. This means that I only can run the frequency for 5.2 gigahertz to 0.6 or 0.4 to 6 gigahertz because I am limited in the bandwidth. But the mounting device that we put in the drone has a transmitter at the middle and two receivers. The idea of the receivers with these mechanical things that you can change antenna to the vertical position or to the horizontal position. And remember when I talk about these polarizations, having an antenna transmitter in the vertical position and a receiver in the vertical and in the horizontal allow us to characterize these trees. But also we have in the drone mounting device that allow us to change the angle of the antennas. Remember when I was talking about this tomographic profiling I was explaining to you that the antenna or the angle can change along the scanning path. And that's the idea. The idea is that we will do two imaging modes, one called the nadir position and the other one at the changing the angle of these antennas. Remember this one for example is unable to allow us to compare it with the satellite data. And now I will talk about the low range low power transmitter. This is a device that allow you to communicate between another device up to a five kilometer distance. And I basically I'm using these devices to turn on and off the radar remotely. And I use a Raspberry Pi Raspberry Pi is like a very small computer has exactly same characteristics as a computer. It runs in a in a Linux operative operative system. And what I have done in this Raspberry Pi is I use the SSH server that allows me to connect with the Raspberry Pi remotely. This is because I am not using any keyword or mouse or screen in the Raspberry Pi. Then in order to see what is happening there. I use this as a SSH server, which is just connecting my computer with Ethernet cable to the Raspberry Pi. And I run a Python program there. This Python program allows me to run the radar and scan these six different or do six different scanning options. And at the moment I'm working in this last part which is basically what I want to do is read the differential GPS data every time I'm scanning and save it from the drone to the Raspberry Pi. But now I will explain the complete system because I just explained each of the devices and I have what I will call the ground based system and the air system in the ground system I have an application running in Matlab. And the first thing I do is I connect this application with the Raspberry Pi but by this SSH server, which is just an Ethernet cable that allows me to send a Linux command to say run the P program, sorry the Python program. Once I have done that, I disconnect the cable and then the drone can be in the air and then I select depending on the scanning options, I have six different scanning options that I will explain later. I also select the name of the project and then I go to the connect. When I connect, what I do is connect to this Lora A device, which is a serial communication. And once I have done that, this application is able or allows me to now do the scan. If I was sending or switching this switch, then I send a command saying turn on the radar and scan the option one. And that's what it does remotely, sends that command to the Raspberry Pi, selects the scanning option and then runs the radar. All the data that I am scanning is saved in the Raspberry Pi. I am not sending it again just because we could have some, you know, lost of the data then it's better to save it in the Raspberry Pi, which actually I had it in an external USB. And here I just want to show that I could turn on this switch as much as I can and what it will do is scan several times and, you know, the project will be exactly the same and it will have the number one, two, three, which will allow me to know which scan number I have done. And these are the scanning options. As I explained before, the radar allows you to change certain parameters. The main difference between these options is the time scan. All of them use the south tooth option, sorry, waveform, starts and ends in the same frequency and has the same bandwidth, but they change between 10 seconds of scanning or 10 seconds, 30 seconds of scanning or 10 seconds. And also the sweep time could be one millisecond or 500 nanoseconds. And also I can change the day receivers could be one receiver or two. At the moment, if I am using only one receiver, I can have a very good sampling points, but if I am using two, it's divided in those two and I am currently working in improving that, it's a little bit more complex. But yeah, that's basically the six different scanning options. And this is just an example of the scanning option 5. As you can see is the south tooth waveform, 5.2 gigahertz to 5.6 gigahertz to channels. Again, I told you that one of the things we have seen is that using 200, sorry, 128 samples, it's not enough. And then with this image that you're seeing here at the right, we flew the drone at an altitude of 40 meters, and the drone was flying in 10 meters per second. What you can see here, which is just the real big mapping, which is the Fourier transform of the complex data is the soil, which is here in yellow. Probably if you use your imagination, you can see the trees. But as I mentioned before, we have two main problems. First of all, we don't have enough number of samples for a better image for a better resolution. And second, it was flying too fast, then we we see some noises, then at the moment we are working with this. Just to let you know these image corresponds to a fly with it in the in a farm here near to ready because, well, because of COVID situations, we haven't been able to to go to this with them woods yet. And that's it about the drone. I would like to acknowledge Professor Keith Morrison, which is the PI of this project. Also Andrew Lomas, which he is the technical manager who created the drone structure for the antennas and our certified pilot Ian read. And I will just drink some water. Now I would like to talk about the radar rig. And the radar rig it's an instrument founded by the land wise project. Basically the aim of this rig is a characterized field scale heterogeneity of so moisture for different national food and measurement practices. The idea is to have a transportable radar that allow the description is a software controlled instrument that runs a radar on a real from and starting to an ending point scanning every given distance. The radar rig consisting three units. The first one is the motor unit which has three parts. And by the way, this is the version one, we have an improved version, but it consists of the motor, the controller and the driver. Then we have the radar unit, which is a vector vector network analyzer and a rate of antennas a transmitter and a receiver. And all of these runs in an application again in Matlab. The motor unit, as I say before, consisting the driver and controller and the motor. The driver I created a printer circuit circuit board using this open software target 3001. And in that printer or PCB, I put a microcontroller, a microcontroller is like a mini computer that allows you to manipulate signals like turn on an LED, or in this case, a run a motor. In this PCB, I also have a switch switch that I call home because will allow me to know when the motor is at home, and a switch that I call in that will allow me to know that the motor is at the end. And then I have these connectors which have three main signals that are going to the to the controller. Well, these are just power signals but the signal. It's a naval signal, which will turn the motor on and off. There were direction signal which will tell you fastward or backward and the pools, pools. I send a series of pulses that will be translated or converted to velocity. And this driver requires a 24 voltage to be connected to the to the sorry that to the controller. And these bolts are already this is a battery that goes to the motor, but the driver will come the controller will convert these a voltage to current and will send it to the coils of the motor. The motor we are using is a stepper motor. The advantage of a stepper motors is that you can know the location of the motor. For example, in this case, a pulse will allow you to move the shaft 1.8 degrees. Then for a complete rotation, that means 360 degrees, you will need to send 200 pulses. If you send 200 pulses in a second, then you will have a one revolution per second. And that's how basically works the pulses are converted or transformed to these current signals that will allow you to move the shaft of the motor. And all of these so the way I am sending or telling to turn on the motor the direction and the pulses are manipulated by a serial communication that material I also know where is the position of the motor. And remember I say in the PCB I have a microcontroller and this microcontroller I use an Arduino IDE to program the microcontroller. I'm not sure you have here about the Arduino Arduino is a company that develops softwares and hard words. One of the softwares is Arduino IDE. It's an open source software. I use the one in Linux but sorry Windows but you can have it for Linux or Mac. And basically it's a C++ language. And I also use the Arduino Uno as programmer. And this is important to mention because I have seen a lot of people buying Arduino Uno every time that they need to do a new project, but actually you can use your Arduino Uno only as a programmer for for microcontrollers actually for different microcontrollers. And my program is the ATMEGA328. And if you want to know more about how to do it just go to the website of Arduino, the Arduino to breadboard, and you will see the instructions are really easy to follow. And then you can just do it. And here I will just show the flow diagram of the program I put in the Arduino. I habilitated the serial port. Remember to enable or disable the motor and the directions. I use the serial. I use some call interruptions. This interruptions will be connected to the switches. And then the motor pins remember enable direction and pulse. And I also put some constants. One of the important constants is like this shaft of the of the motor. And this 1.8 degrees at how many distance or movement it's equal to and it's 0.75 millimeters. And basically what the program does is like it waits for a serial command, the serial command consistent six numbers divided by comma. The first number tells me the option. I have option one which is a scan option to which is moved to home. Option three, which is a calibration option for which is moved to the absolute position and option five, which is a stop the motor. The second common, sorry, the second number on the common. It tells about the direction zero is felt forward. One is backward. This third number is related to the position to the start position, what I will call a location or lock. The next number is about the end of the scanning. In this example, 2000 millimeters will be up to two meters know like of a scanning and the other common or sorry number. It's related to the increments of of these scanning. And finally, the last comma, the last number is related to the velocity of the motor. In this case 15 is equal to or is in millimeters per second. Then if I send this common, what what the problem will do is is because the number is one will go to the scan and will compare his location which at the beginning is zero equal to the end, which is 2000. If it's not, then sense a pulse and increase the location for this step. Remember, I told you is the step of it's equal to point 75 millimeters. Then comparison location now is equal to the increment location will be now point 75 millimeters is not equal to 20 millimeters. Then it will return to this instruction and it will be here until the location it's equal to 20 millimeters. Then it will be half second that half second allows me to send an instruction to the DNA to scan, and then we return here. And as you can see it goes here until the motor have run up to two meters. If I am sending a number two here and a number one here what it will do is go to the instruction home and basically the motor runs backward. It will run until it press the switch button home. And remember I told you about these interruptions well what what will happen is that this program is break or stop jump to the to the sub routine. And what it does is create initial conditions again, and it won't return to this party will jump to the again to waiting for the serial command. I can do the same for the calibration I sent a number three, and what the calibration will do is, it's only run to the motor to the middle of the reef. Or I could tell the move to absolute position will be to 10 centimeters 20 centimeters, or in even in millimeters, or I can tell them what the program to stop. And that's basically the flow diagram of the of the program I put in the microcontroller. Now in terms of the radar unit and the radar unit, as I say before consistent antennas and the DNA in the antennas we have a transmitter and a receiver, but the mechanical part of the of the rig allows us to have the different polarization configurations, that means vertical vertical vertical horizontal horizontal horizontal vertical. And again, we will with the mountain device we will able to rotate or to change the angle of in which these antennas are pointing between zero to actually 45 degrees. And then, yeah, that's that's a deal with a mountain device and antennas. And in terms of the vector network analyzer, we have, we said that I can default configuration that means s to one means that transmit for port to receives for port one. Remember runs to four gigahertz to a gigahertz because we are centered in the band C and has a resolution of 401. So this means that how in how many steps you can divide this range of the frequency. And then we have a bandwidth, when we are scanning, we are having a bandwidth of three kilohertz, and when we are calibrating we are have a band with 300 hertz. The bandwidth of I F bandwidth is related to the filter, when the signal is returning to eliminate some noise that as you can see from scanning we we had it higher and with a higher filter we have a faster data than in a with a lower filter. The power the power we are using is between zero and two, we kind of use a bigger or smaller because we are limited to the to the frequency, because of the frequency. And the data we are receiving real and imaginary data. And in this device for this device we have an Ethernet communication TCP IP protocol. And the application well what I did I developed as application using the design app of MATLAB, and this it's just very simple I create a two bottoms wants to browse the project and another one to give a name to the project. Then I have these options to configure the BNA. You know the, as I told you is a, as an Ethernet protocol then I have, I need an IP address. The continuous mode, it's off that means I every time I send a signal, I will scan. They remember the BNAS parameters is transmits for the port two and receive, receives from port one antenna polarization this is this actually it's manual you need to click, depending on how your antennas are. And then I have a sweep, an average of one because if I increase it, the scan will take longer and I want a very fast scanning. I remember the band C frequency, the number of frequencies is 401 power of zero dvms. The bandwidth of 3k. And then if, if I'm happy with the configuration I press configure BNA and you can see remotely how the BNA have been configured. I didn't create this app, this came with the BNA. But as you can see I can, I can see the parameters are well configured. And then this other section of the application allows you to control or to put the parameters on the motor. And here I will just say it like you have two options auto or manual. If I'm clicking the switch, what I can do in manual is a run the motor to home or run it to an absolute position. If I put it in auto, it tells me, well, what's the start position you want to scan the end position you want to hold the limit of the increments and the velocity. And I, when I have done that I press scan and the motor will just run and scan, run and scan. Now remember I talk about calibration. Well, if I want to calibrate this device or this instrument what I do is click here and it will open a new window. I, again, put a name in the, in the, in the file. The antennas, what is the configuration of the antennas and again, frequency start and number of points. And this is just for the, for the calibration. Now if I want to, or I will talk about the calibration. It is important to calibrate this, this device because we want to know what are we seeing and the way of doing it is to use reflective surface that we know the shape of them. In this case, we are using a diahedral and to try a hydrals one of five millimeters and one of eight millimeters. We did a calibration in the laboratory, in which we place absorbent material, a pole, and then we put these shapes. And if you see the image of these shapes you will basically see like floating in the, in the, in the space. This will allow us to remove certain noise can be created by the, by the coaxial cables or, you know, when you connect and disconnect these cables. And then because this device is transportable what we do is place it again in the field and conduct a very simple or a small calibration at the beginning of the, of the, of the day, and at the end of the day. And I will just show you here how it looks. And this is just what we call big diahedral at the start of the day and at the end of the day we cooperate with the laboratory data and these are just some results that we did from a test we did outside of the university. At the start of the day with a R square, we have 0.7 and actually improve at the end of the day with a big diahedral and the same with a small triahedral. And this is just a video that I wanted to show you how the, the rig works basically remember it starts in a, in a zero position this case and will end to the this end, which is more or less two meters and scans and stop well. Every two centimeters stops and scan and then you will receive the data in this example the antennas are a vertical vertical and are, I think about 10 degrees, and we are trying to see this path of so bear soil. This is by the way the version one of the rig. We have an improved version. And this is the real been mapping. Here an example of antennas at the BB polarization at zero degrees. You can see here is the the soil return, more or less. And then in the bear so we see a bigger effect of on the signal. There is no vegetation and more compaction perhaps. But again, this is not this is not related to soil moisture. This is just a real been mapping of the of the signal. And here I just wanted to show you the effects of the of the polarization when we use the vertical horizontal you can see there is a change in the image. And this is because certain areas or parts material on the soul are depolarizing the signal. And this is the second version of the rig. The photo has a lot of things at the back, but you can see now there runs in two reals, the legs are a little bit stronger because with the other version there was a lot of instability when the when this the background wasn't quite even. And now it's really really stable. Now for this project I also want to acknowledge Professor Keith Morrison, Dr Wilma Flanka, which is the postdoctoral researcher in the language project, and Andrew Lomas, which he developed the second version of their break. And before I read the questions just wanted to tell you to please look at our suggested session. In the EU 22. It's a geo infrastructure infrastructure monitoring complex data analysis and instrument applications. And that's it. That's it from about the, the instruments I hope you'll like it. Thank you for the great presentation actually and great work have been done from you and your team at University of reading. In terms of the inside and also rather Rick. A great project and to wish you a very very successful project with this actually. Before we start questions I don't know if anyone has questions. I will just share with you one two slides and then we can come back to the presentation for if anyone have any questions. And so, yeah. Here as Veronica already mentioned that for our session in the EU conference assembly 2022. It will help in Vienna, Australia. We will be back to 8th of April 2022. So you have the geo infrastructure monitoring complex data analyze and instrument application with the team from Gustav E. Phil and the University of reading and also the University College of Dublin. I would be honored to receive your abstracts for this session. And also to other special issue in mdpi remote sensing. When it's related to rather techniques for structures currency characterization and monitoring. Here is the link. You can use mdpi.com si and this number you can reach the special issue. It's 31st of December. And the second one we have remote sensing for the infrastructure assessment using entities and intelligent that analyze new trends and challenges. So again, you have this link using this code you can reach the website of the special issue and we are pleased to receive your papers and plan papers even if it's a paper that you're planning to publish you can contact us with emails and we will give you instructions how how can be submitted in before the deadline. But yeah, I have several questions actually very unique to to talk about the inside instrument and also the rather read read. So both of them they are more than in the range of the frequency is four to eight. I guess according to the European standard and harmonization of the frequency. It's allowed no lately until I'm not sure until 24 or how many gigahertz is allowed so it depends. For example, one of the problems we have right now in the within woods is like there is an American base. We can't fly to the we can use a frequency of point five point. There is there is a range we can use. And that's the reason why we haven't been there. We need to have a special permission to fly. Exactly. And then we had the permission before but because of Kobe it, you know, it was just like not longer available and now we need to renew it and it had been like crazy. Because I remember we were in the conference your GPR. So this was one of the topics that your GPR was trying to attend this meetings about the harmonization of these GPRs and permissions and the range is how how it's possible because new new manufacturers of rather I guess they are facing this problems lately. With the Rick, we don't have that problem because it's it's more like a ground system. Yes, with the drone. Yeah. I was. I saw that there is a range between the Laura and the SSH server. So I how long this can be connected. Well, the drone. It uses also something gigahertz. I don't remember right now like, I think it's five, two gigahertz, five gigahertz of communication. And Laura uses megahertz then they never interfere. And that was one of our issues at the beginning because we were trying to use a system to connect it and it was interfering with the with the with the drone. We have a lot of interference the drone normally tends for security reasons just go lands. Great. And in terms of the vertical and horizontal solution, I don't know if you did any analyze to see how how they how it works for the vertical and horizontal resolution of your rather rich, especially which is what I am working on mostly not inside. And insert also you did any analyze any checking and assessment for the device has a calibration that you can see and tell what type, what range of the horizontal and vertical resolution it can be valid. Yeah, well to be honest, I'm not involved in the data analysis that's more a professor Kate Morrison, then I can't give you too much information about that. It's a little bit, you know, my limit in the in the in the project. We have actually two interesting questions if you will like to read those. Yes, actually, one of the questions is from Daniel. The first one actually actually from Akim Akim is asking if this rather great can be useful for the environmental engineering or geology application, or which areas that you are supposing to use this rather great. Actually, the radar rig has a lot of applications, we can, you know, analyze soil moisture, but actually tomorrow I am flying to Sweden to characterize some vegetations we are thinking of using it in certain areas but for vegetation and it has, you know, as I think it's as much as imagination we want to use it for ice I know it's no there is vegetation and soil moisture. Then I think those are the quite interesting applications. So for ice and vegetation and soil moisture these areas know so it's environmental engineering and could be also used for geological applications. Okay, and there is another. Thank you for your question Akim. I hope you got your answer. Tania is asking the special resolution. It was again the same question about. Yeah, that's actually quite an interesting question I can't tell you right now. We flew the drone in 10 seconds at 10 meters per hour, and that can give you a little bit of an idea of the resolution, but we just realized we have a lot of noise. And, but yeah, that's a very good question I don't know really I need to do some, I think calculations to get that you know more about that. And for rather we did some, well, each rather is different as you know the boat that will be depending on the frequency. And in what application you're using but we did some calibrations for for horizontal and vertical resolution so it's very interesting area to see how how this rather is working and according to your application limits and the range. And so you can see the vibration and this type of thing so you can see the wavelengths and how it works. Could be another presentation. Yeah, I know. One of the problems we had with the radar. The radar is connected to the Raspberry Pi using a library for Linux library gives me a limit of the amount of data that I can have on the boat in the in the in the transfer. And I use two channels I tend to, to loss a lot of data, I said is 120 samples. And right now I'm working to try to solve that in terms of a C program in a Linux and quite crazy. I am sure that this will be a great project ahead. Actually, this is, well, all all the manufacturers usually they mentioned in the guidelines and instruction of the rudder so what's the special resolution and how is it approximately well it depends on the application but usually they mentioned so it's it could be something that a task for for the project that you want to develop more or extend in some other parts. I hope Tonya you got answer. There is another question in QA. Russell is asking, can the drone based are used in the mineral explorations for mining and this type of applications. That's a good question I think we have a deep penetration. The thing is, if you want to increase. Well, for the presentation, do you want to like a deeper presentation you will need lower frequencies. And if with lower frequencies is not that expensive to create or to construct a SAR a drone SAR system. It becomes more problematic when you want to do a higher frequencies because components to be more expensive, but with a with a lower frequency it will be possible, but not with the system we have because we have a limit a frequency between 5.2 to 5.6 gigahertz. Well it's high resolution, which the depths of penetration with will be automatically less. And I guess for mineral exploration and mining it could be. I'm not sure it's not my field but how how are the material in what depths and the minerals sorry and in what depths. So I'm not sure if it's very useful for this for this application. But yeah, maybe Russell can be contact with you and you explore more about maybe a ground a ground radar will be better for mineral exploration but I don't know if there is any implication. You're again using a higher resolution so I don't think it will be useful again for for for this application for mineral because usually you will need approximately several meters. I will, I will assume. Yeah, I suppose. Here, here with this with the drone we are penetrated, maybe five is five meters. Yeah, I think so with this one, but it could be could be something. Yeah, it looks like with the image I saw before. Okay, so maybe it could be useful for some type of minerals which I don't know how in what depths and how I don't have any other questions from attendance but I have some questions actually very unique. So you're using very interesting feature in your rudder Rick, the development of this rudder Rick. Well, I use rather but I couldn't yet see this feature, which is using different angles, 01030. No, you're using different angles I was using polarization like diaper orientation of the antenna. Manually changing to to use in two different polarization, which I receive different information from the seconds. Sometimes, usually now I use both polarization dipole antennas vertical perpendicular and parallel to my, for example, if it's a crackline. I get some information in from both, which is helping me supporting me for the interpretation part of the data. But with angle, the question here I don't know what type of information, you can get. I'm not sure what type of information you can get from changing the angle. This is similar to SAR, you know when you get the angles you get like, like another visualization of the object. You have any examples or not yet. No, not yet actually. Okay, all the things I show is with the, with the antennas looking at the ground, I guess. Yeah. This is just the real been mapping is not it's not data analysis. But yeah, I can give you other type of information when you when you change the angle of antenna. I see that we can extend this presentation to another one. Interesting. The vegetation, we are using that 10 degrees angle and it's okay to get that information. You had made you had took any scans that in different angles for example 10 and 30 and you can see the information, how they are like for example making some comparisons to see if if inter of course there will be different implications and details of the vegetation or any application that you're using for, but I'm, I'm, I just wonder, I am curious to see how, how they are looking like. And yeah, for any questions from attendance. From my part. Thank you again, Veronica. It was a great, really presentation, a lot of information, a lot of technical things, technical parts of the of both the instruments. And one of the ideas with this webinar was to show people that it is not that complex to create an instrument because now with Arduino Raspberry Pi, and even Matlab Matlab the application I use is very friendly. And I wanted to show people like, you know, it's not that complex and I mean you will require a lot of like, you know, really, I wonder to ask Veronica I guess this type of devices when you are developing because I know some colleagues from Croatia and Serbia they were developing some parts of the radar. So, I guess it's very low cost. It's, it's not. I mean, comparing to the drone. Yes. And that was expensive and with the DNA the DNA is expensive too. But yeah, it is compared with, you know, other systems, you know, compared to buy a new one. Exactly. So it's possible for students and master PhD students and other students could really develop these instruments at lab. I'm sure there is a lot of skills and with the attendance that they can really work on that. We have another. No, actually, there is a lot of thank yous for you, Veronica. We thank you a lot. Just one note for attendance. This webinar will be available in the YouTube channel for EGU channel. I am not sure what I will show you maybe the EGU channel. So you can just search in the YouTube EGU channel in a few days you will have this presentation. Exactly. This is, this is how I will share with you. Just one minute. Here you can see the European Geoscience Union. So there is a lot of webinars not only our there is different webinars related to ocean plastics ocean conservations air pollution. How to get the most out of the online conference and different webinars hours will be available in this YouTube channel in in few days. We will ask the or ask the team to send you the link of the webinar. No more questions. Someone is asking will there will be certificate given to attendance. Please just send us an email. Here I will share. Okay, when you registered you got an email from the system know please contact the same email with Simone. I guess he will provide you a certificate with the logo of EGU and he will contact us to prepare this template for you. We will, we will consider that with with Simone. Thank you. Here we are in the very very last few minutes to say goodbye and once again Veronica thank you very much for the presentation and your availability very valuable information and details. I hope all the attendance got some information some very very I'm sure they got a lot of new informations from this presentation by myself I got a lot of new new updates. Thank you very Nika. And we are looking forward to meet you all again at attendance and very Nika for new presentations. Okay, well, thank you. See you. Thank you for all.