An Overview Of SAR Data Sources And Tools

Speaker 1: Hello, everyone, and welcome to the third and final session of this RSAT webinar series titled Introduction to Synthetic Aperture Radar and Its Applications. I'm Dr. Erika Potus, a scientist at NASA's Jet Propulsion Laboratory, and I'm also an instructor with the RSAT program. Today's session will be an overview of SAR data sources and tools, which will be delivered by invited experts Dr. Franz Meyer and from the Alaska Satellite Facility and the University of Alaska Fairbanks, as well as Heidi Christensen from the Alaska Satellite Facility.

Speaker 1: And as a reminder, this webinar series has consisted of three sessions. The first one was an introduction to SAR. The second one, an introduction to InSAR. And today's session, an overview of SAR data sources and tools on the Alaska Satellite Facility. There is a homework associated with this webinar series, and you can access it as of today on the webpage for this training. The certificate of completion will be awarded to those participants who attend all live sessions and complete the homework assignment by the due date, which is December 4th.

Speaker 1: So let's get started with today's session. Let's start with a review of prior knowledge. SAR is an active microwave sensor that images the Earth's surface. It uses different sensor configurations, including wavelength and polarization, to provide different information about surface characteristics. The surface parameters that influence the radar signal are related to moisture and to structure. SAR datasets include both amplitude and phase components, so the amplitude is used to generate imagery that displays the intensity of the returned signal, and the phase indicates the distance between the sensor and the surface, and is the basis of interferometric SAR, which can be used to identify and quantify surface deformation.

Speaker 1: SAR imagery has distortions that are caused by that side-looking geometry or nature of of SAR sensors. And radiometric terrain correction adjusts for both geometric and radiometric distortions, allowing SAR imagery to align with other geospatial data sets. How to ask questions. Please write your questions in the questions box, which is at the bottom right. There'll be three points. Click on that, and there'll be a pop-up menu. Select questions. And write your questions in that box. And We will address your questions at the end of this session.

Speaker 1: Feel free to enter your questions as we go along, and we'll try to get to all of your questions during the Q & A session. The remaining questions that we cannot answer, we will answer them in the Q & A document, which we will post to the training website in about a week. So the objectives of today's session are the following. To understand the role of the Alaska Satellite Facility within NASA's Earth Data Organization. To be aware of the wide range of SAR data available at the Alaska Satellite Facility, especially analysis-ready data.

Speaker 1: to know how to search for data, SAR data available on the Alaska Satellite Facility, and how to access analysis-ready SAR datasets, including archived data and customized data generated on demand. Today's guest trainers are Heidi Christensen, who's a GIS specialist at the Alaska Satellite Facility, and Dr. Franz Meyer, who's a professor in remote sensing at the University of Alaska, Fairbanks, and chief scientist at the Alaska Satellite Facility, as well as member of the NYSAR science team. I'm extremely grateful that they'll be sharing their expertise, so welcome Heidi Christensen and Dr. Franz Meyer.

Speaker 2: Thank you, Erica. My name is Franz Meyer. I'm the chief scientist of the Alaska Satellite Facility, and I'm also a professor for remote sensing at the University of Alaska Fairbanks. And in the next few slides, I want to just briefly introduce the Alaska Satellite Facility to you and tell you a little bit who we are and what we are doing to support the science community and why this is a really good time for you to get started with looking into SAR as a resource for your work. So the Alaska Satellite Facility is one of so-called 12 NASA DACs.

Speaker 2: And DAC, in this case, stands for Distributed Active Archive Centers. So what NASA has decided to do is to spread out their archives to a variety of different locations and a variety of different DACs that are then supporting different parts of the science community. ASF specifically is shown here on the slide on the top left. Let me turn on my laser pointer real quick for you. So this is ASF up here. We are located in Fairbanks, Alaska, and we are the DAC for SAR, for Synthetic Aperture Radar Remote Sensing Datasets.

Speaker 2: So the DACs overall, they are custodians of NASA mission data or NASA-linked datasets, and they ensure that these datasets are kept safely and are being easily made accessible to the science community. And so again, ASF is the data center for SAR data in that space. We provide reliable and robust services to the science and applications community to discover, access, and work with the datasets that we are the custodians of. ASF does a lot of user engagement also with the community through myself as the chief scientist, through an organization that's the so-called user working group.

Speaker 2: This is close to an advisory board that works with ASF to make sure that we are providing good services to the community. And, of course, ASF also has a user support group that is working with you with answering your questions, does trainings, et cetera, to the community.

Speaker 2: So ASF specifically is the Alaska Satellite Facility DAC. ASF sits in Fairbanks, Alaska, as mentioned before. We've been around since 1991, initially acting as a downlink station for NASA. So we operate these dishes that you see on the bottom of the slide to provide space communication services to NASA-linked satellite systems. But we also are the data center itself And currently ASF is housing about 23 petabytes of synthetic aperture radar data sets in its archive. Most of these data sets are sitting in the cloud through a partnership of NASA with the Amazon Web Services Cloud.

Speaker 2: So all of our data sets are available for immediate download and immediate use. ASF has probably been known to you for a couple of reasons. One of them is we are a host of a copy of the Global Sentinel-1 Archive, so we're working with the European Space Agency to provide access to this valuable resource, Sentinel-1, which is a C-band star system that was launched by the European Space Agency and is maintained by ESA for the benefit of the community. Also important for ASF is that we will be the host of an upcoming NASA mission, the so-called NISAR mission, and we will be hosting all level 0 to level 2 LBN SAR data coming out of NISAR.

Speaker 2: So I'll talk about this in a few slides here in a few minutes. ASF has a total of 41 years' worth of SAR data for your consumption, and all of these datasets are available through this URL. at search.asf.alaska.edu. And when you access these slides later, this is a live link and gets you directly to our search interface. So the SAR community or the space of synthetic aperture radar has been really exciting for the last few years. And it's because we have a series of satellite missions that are providing globally observed free and open SAR datasets to you.

Speaker 2: One of the notable ones I mentioned earlier is Sentinel-1. This is a mission that has been around since 2015 and 2016 and has been providing this global access to free and open SAR data sets that you can see here on the right-hand side. Soon to come in spring or early 2025 is NISAR, which I'll talk about in a few slides. This is a NASA-supported mission in collaboration with the Indian space research organization ISRO launched this in 2025 and will provide similar quality data, meaning that the data sets will be global, will cover all land masses once every 12 days, and will provide all these data sets for free.

Speaker 2: And similar missions are also in the works. The European Space Agency is working on an LVAN mission called ROSEL, which will launch also in the next few years and also will provide free and open datasets. So this space of free and open data is becoming more popular. And ASF provides access to many of these modern SAR sensors that provide these regularly sampled free and open characteristics on a global scale. So one of these missions that we support is Sentinel-1. As mentioned, Sentinel-1 has been available since 2014.

Speaker 2: Heidi Christensen will talk a little bit more about the specifics of this mission. For me, the main milestones here is, as mentioned before, the datasets are free and open. This mission was particularly designed for interferometric SAR. which some of you may be using that allows users to monitor surface displacements on a global scale. And it images all land masses, coastal zones, and shipping routes every six to 12 days. So it's a very exciting mission. Sentinel-1 is also what's called a constellation of satellites.

Speaker 2: In order to provide its global capability, they are using specifically a constellation of two satellites, initially launched as Sentinel-1 A and B, With the two constellation system, it can provide six-day sampling over Europe, over the Arctic Ocean, and for some global hazard locations, and 12-day sampling for the globe. Unfortunately, one of those two satellites, Sentinel-1B, had a failure in December 2021, but luckily for all of you, there is a replacement satellite in the works And so Sentinel-1C is slated for launch late this year.

Speaker 2: So hopefully very soon we have a constellation of satellites available for you again.

Speaker 2: Along a similar vein, we will be hosting NISAR. NISAR stands for the NASA ISRO Star Mission. So this is a collaboration between NASA and the Indian Space Research Organization it will provide, it will launch in 2025, and it's going to be the first Spaceborne two-frequency L and S-band SAR mission. Out of the L and S-band data, ASF will be hosting the global L-band data set. So all of the L-band data will be available through ASF. Also there, NYSAR will provide full global 12-day coverage in both ascending and descending, and ASF will be A data center for the LBNSAR data, we also are a ground station, so we are receiving some of these data sets directly to Fairbanks through our ground station antennas.

Speaker 2: NISA, a revolutionary mission in many ways. One of them is simply by the sheer amount of data that NISA will provide. NISA will bring in about 50 petabytes of data per year, or 150 petabytes of Earth observation data in three years. And all these data sets will also be free and open. It's a very exciting mission that many of you, that will accelerate the research that many of you are doing.

Speaker 2: So with this, or sort of suffice to say that with all of these missions, the archives of free and open datasets are exploding rapidly. So for ASF alone in 2014, our total archive was about one petabyte. Currently we have an archive of about 23 petabytes available. And in just a few years, our archive will grow another factor of 10 to about 180 petabytes in our archives. So there's a tremendous amount of free and open SAR data becoming available that will keep the science and applications community busy for a very long time.

Speaker 2: And because of the volume of data, and to make this volume of data available to you in an effective way, we have migrated all of our archives into the cloud And this opens up some of the tools and services and capabilities that Heidi Christensen will talk about here shortly. So ASF has focused very strongly in the last few years to develop efficient mechanisms that allow you to access and work with this large volume of SAR datasets effectively. And so We focused on a sort of a cloud-based archive, so all of the data sets that come in to ASF, they are now stored in the cloud, in the Amazon Web Services cloud, and then based on this cloud-based archive, we have developed efficient mechanisms to discover these data, to access them, and then also build some services on top of these data sets that allow you to create value-added products and work with data sets more effectively whether you work in the cloud or whether you prefer working on your local machine.

Speaker 2: And so for many of these cool ways for how you can find and work with datasets through ASF, I will hand it off now to Heidi Christensen, who will tell you all about how to find datasets, access them, and do some custom processing with these very large volume datasets. And so, Heidi, I'm handing it off to you.

Speaker 3: Thanks, Franz.

Speaker 4: ASF has quite a range of data that's available in our archive. We serve synthetic aperture radar data that was collected by NASA itself, but also collected by partner agencies, including the European Space Agency, the Japan Aerospace Exploration Agency, and the Canadian Space Agency. Most of the data in our archive is collected by sensors that are on a satellite. They're orbiting the Earth and are on a continuous path in collecting data. We do, however, have airborne sensors as well. These missions have the SAR sensor that is mounted on an airplane, and the footprint is much smaller, and generally these are much more targeted missions.

Speaker 4: We even have data from a mission that was mounted on the space shuttles.

Speaker 4: There are a range of different sensor configurations that can be used for SAR, and when you're looking at a data set, it's important to know what these configurations are. Most of the data in our archive is collected in either C or L band. C band has a wavelength of about five to six centimeters, whereas L band is much longer at about 25 centimeters. We do have a few data sets that are in other bands, but they're very few and far between. You're probably going to be encountering mostly C and L band data.

Speaker 4: SAR can collect data using different polarizations, and the different polarizations tell you different things about the characteristics of the Earth's surface. Similarly, there are different beam and acquisition modes that can be used that also tell you different things. So if the incidence angle is different, that tells you different information about the targets that the signal is interacting with on the surface. So if you're doing a time series analysis in particular, it's very important to make sure that you're using a consistent data set, that it's the same, preferably the same mission, and using the same mode and polarization.

Speaker 4: The repeat intervals vary for the different missions. Some of them have fairly frequent repeat intervals.

Speaker 2: So some of the.

Speaker 4: More modern satellite missions have fairly dense coverage. Whereas some of the older missions, you may only get an acquisition over your area of interest every month or so. So it's important to be aware that not every mission provides the same amount of coverage. All of the data that ASF serves is free and open and can be directly downloaded. Most of our data sets are actually hosted in cloud storage, and some of the data sets can be accessed directly where they're sitting in cloud storage. So if you're into cloud computing and you want to explore that option a little bit more, check out the link to learn more about direct S3 access.

Speaker 4: I'll be going into a lot more detail on the access options a little bit later, and we'll be doing a demo with one of the web-based interfaces, but there are a lot of ways to access the data and it's all easily available. ASF supports quite a range of missions, including those that have continuous coverage such as the very first Spaceborne SAR mission, which was CSAT in 1978, all the way through to currently active missions, including Sentinel-1 and ALOS-2. For Sentinel-1, we actually have the entire archive of data that's been acquired from that mission from 2014 to the present.

Speaker 4: For the ALOS-2 PALSAR-2 sensor, we're actually, even though that mission has been around for about as long as Sentinel-1, We are just now getting the data from JAXA, so we're in the process of ingesting that into our archive. So hopefully by the end of this year or early next year, you'll be able to search for the ALOS-2 PALSAR-2 data, which we're pretty excited about. We also support quite a range of historic missions starting from the 1990s onwards. There's a lot of data in our archive, again, with different wavelengths and different sensor configurations, so a lot to choose from.

Speaker 4: We also support a number of discrete missions, including the shuttle-borne missions and the airborne missions. For the airborne missions, because they're so much closer to the Earth, the footprint tends to be a lot smaller, and so these missions are usually for fairly targeted projects, specific events, or particular areas of interest. You can learn more about all of the SAR missions that AFF supports on our website, including the NISAR mission, which is coming up soon, and.

Speaker 5: I'll be talking about that a little bit more later.

Speaker 4: The Sentinel-1 mission is the most commonly.

Speaker 5: Accessed data set that we serve.

Speaker 4: This is an effort of the European Space Agency and European Commission and is part of the Copernicus Initiative of satellites.

Speaker 1: Many of you may be.

Speaker 4: Familiar with some of the other Sentinels. Sentinel-2 is the optical mission. and Sentinel-1 is the SAR mission. It provides global coverage with C-band SAR, and it was designed as a two-satellite constellation. The first satellite, Sentinel-1A, was launched in 2014, followed by the second in 2016, and we had five glorious years of having those two satellites collecting data. Each one of the satellites has a 12-day repeat cycle, and they have the same orbit pattern, but it's offset.

Speaker 1: By 180 degrees.

Speaker 4: So this allows the satellites to collect data every six days, potentially. For most of the globe, the acquisition plan was for a 12-day interval, but over the European Union and some other geologically active areas, there was six-day coverage during that time.

Speaker 4: Because these are polar orbiting satellites, we also see a lot more coverage closer to the poles. So if you're interested in amplitude data in particular, There are a lot of repeat passes, even if they're not quite from the same path, you can get quite a dense time series. Now, unfortunately, Sentinel-1B is no longer collecting data. Its mission ended at the end of 2021. And Sentinel-1A is also reaching the end of its life and will need to be deorbited in the not too distant future. So Sentinel-1C is waiting in the wings and is due to be launched at the end of this year.

Speaker 4: And 1D is also ready to be launched once there's a launch vehicle ready. So we're hoping that at least one of those satellites will be in orbit by the time we lose Sentinel-1A, and that would allow for a nice continuous data set. But at this point, it's hoping for the best. For this mission, new data must be available to download within three days, and In general, we actually see the data available within 24 hours of acquisition. The data is free and it's easy to download. We provide both the raw and the level one products for download.

Speaker 4: So ESA has done a lot of the heavy lifting in terms of processing to provide level one products from the raw data. So the two level one products that we have in our archive are GRDs, which are ground range detected products, and SLCs, which are single look complex products. There are two components to a SAR dataset, the amplitude and the phase. The amplitude is how much of the signal is sent down that actually is returned to the sensor to be measured, and the phase is where that signal is in its wave cycle when it returns to be measured.

Speaker 4: If you're interested in the imagery part of SAR and looking at an image, you're essentially looking at the amplitude part of that dataset. Whereas the phase is used for SAR interferometry, where you're comparing the phase measurements and looking to see if the ground has moved relative to the sensor, and there are a number of other workflows that also consider the phase measurements. If you're only interested in the amplitude part, you may be best off using the GRD. The GRD is a single cohesive image.

Speaker 4: There's one geotiff for each polarization, and it is fully geo-referenced. So if you bring this into your GIS environment, it will show up where you would expect it to be. There are square pixels. It can be projected into whatever projection you need. But it's important to keep in mind that there are distortions that are inherent to the SAR dataset due to the side-looking acquisition approach. And so if you want to use the GRD data, the amplitude data, In comparison to other data sets like optical imagery or institute data, you will have to correct for those distortions.

Speaker 4: A common approach for doing that is radiometric terrain correction, which uses a digital elevation model to not only adjust for the terrain distortions, but also to normalize the amount of backscatter based on the actual surface area that is contributing that backscatter back to the sensor. So the GRDs are a fine place to start if you want to do a radiometric terrain correction. The other nice thing, in the SLCs, you can see that there is this grid, and this delineates the individual radar bursts.

Speaker 4: The SLCs and GRDs are actually made up of a number of different individual radar bursts that have been packaged together into one file to create a much larger frame. Most SLCs contain about 27 individual bursts. And in the SLC, all of that individual data is retained. So even though the bursts actually overlap, if you look above and below the guideline, that data is repeated because there is overlapping data. So in the GRD, any of those overlapping portions are averaged together, so you just get a single output image.

Speaker 4: Same thing with the sub-swas. If they overlap, that's all just combined so that there's one value per pixel. For the SLC, the important thing to know is that it retains the phase information. The GRD loses that phase information. It's only the amplitude, while the SLC has both the amplitude and the phase. So if you want to do interferometry or anything else that requires the phase measurements, you'll need to use the SLC. It's a much larger file size than the GRD, so if you're only interested in amplitude, again, it may be easier to start with that GRD, but for phase measurements, you'll need the SLC.

Speaker 4: Historically, we've not had a lot of analysis-ready products available for SAR datasets, but that's really been changing. And in the past five years in particular, we're seeing a lot more analysis-ready products that are available to the public. We have a couple of different main pathways of providing analysis-ready data. The first is a pre-processed data set that is archived and ready for download, but we also have on-demand processing available. So for any Sentinel-1 acquisition that's in our archive, you can submit that for on-demand processing to either RTC, the Radiometric Terrain Corrected products, or to an INSAR product.

Speaker 4: In addition, this is fairly new. As I mentioned, all of the SLCs are actually made up of individual bursts, and those can all be processed individually as well. So we now offer a way to search for the Sentinel-1 individual bursts, and you can submit those for processing to INSAR as well. Currently, it's just INSAR, but we do hope to have RTC processing available for bursts in the not-too-distant future. The OPERA project at JPL actually generates RTC products based on individual Sentinel-1 bursts.

Speaker 4: This is a new project. They just started last October, so we have just over a year of data now where every Sentinel-1 acquisition, except for Antarctica, has an RTC product that is processed and in our archive. So you can search for it, download it, and it's ready to use. The ALOS PALSAR mission ran from 2006 to 2011, and originally we just had a subset of the data, and all of the products that were in that subset, we processed to RTC and had those products available in our archive. So those are still there and available to search.

Speaker 4: Since that time, we've now acquired the entire archive of ALOS PALSAR imagery, and only the original ones have the RTC. So you may not find a terrain-corrected product for every product that you search for, but for most of North America and the northern regions especially, there's fairly good coverage with the terrain-corrected products. The ARIA project is another project that is at JPL that provides Sentinel-1 GUN-Ws, which are geocoded unwrapped interferograms. These are over specific areas of interest, but they tend to be geologically interesting places.

Speaker 4: So if you're looking at volcanoes or earthquakes, you can check out the ARIA S1-GUNW archives. There are more than a million products in there, and they cover these areas that are geologically interesting. So there may already be an interferogram that's been generated for your area of interest. Lastly, the UAVSAR mission, which is one of the airborne missions has a lot of analysis-ready data that's part of many of the missions. So again, these are much more targeted, but if they happen to be in your area and time period of interest, it's a very rich data source to use.

Speaker 4: I do want to make a quick shout out to NYSAR, which is due to launch in 2025. The earliest possible launch window is in February, and this is a joint venture between NASA and the Indian Space Research Organization, or ISRO.

Speaker 2: This will be an.

Speaker 4: L-band mission, which has– L-band has much greater penetration through canopy and soil than the C-band missions do. And what's interesting about this mission is that it also has an S-band sensor installed. it's not going to be collecting data globally. There'll be global L-band data, but for the S-band, it's mostly over India and a few other particular areas of interest.

Speaker 3: And we're really excited to see how this.

Speaker 4: S-band data can be used for research applications. It's a fairly unusual sensor to have on a space-borne mission. And what's really cool is that there will be acquisitions for both L- and S-band in those areas. And so, the ability to compare and contrast those data sets will be really interesting. It's set to have a 12-day average repeat cycle, which is similar to the Sentinel-1 acquisition plan, so it will be a nice complement to the C-band coverage that's already provided with Sentinel-1. What we're really excited about is the fact that analysis-ready products will be produced as part of the mission.

Speaker 4: Right away as the data is acquired, it will go through a processing step, and quite a range of analysis-ready products will be available in our archive just from day one. There is a suite of sample data products that's available at JPL's website, and we've also put together some guidance for working with these sample data. So, if you want a sneak peek on At how these data sets will look, these sample products are generated using ALOS2 data. So it's L-band data, so similar to NISAR, but it's in the NISAR format.

Speaker 4: So you can get accustomed to how you will be working with these data sets if you want to take a look at that.

Speaker 4: At this point, I'd like to move into the nitty-gritty of how we actually can search for, discover, and download the datasets that we're interested in. ASF Data Search provides access to all of the datasets that ASF archives. It's really specialized for looking for SAR. The Vertex interface is a map-based graphical user interface that's on the web that you can use to find all of our datasets. In the background, there's actually the ASF Search Python package, which is doing the work of finding the data.

Speaker 4: And you can also make use of that package if you have scripted workflows that you'd like to include, download, and search functions within. We also have the ASF Search API, if that's your preference. There is another graphical user interface that's available out there that allows you to find these datasets as well. And this is the Earth Data Search Platform. Now the difference here is that this gives you access to all of the data from all of the DACs. So you can find all of our SAR holdings, but you can also find all of the other data sets that the DACs host.

Speaker 4: Because of that, if you're familiar with Earth Data Search and you're comfortable with its environment, it makes it very easy to also search for SAR there. And if you're working on projects that have resources that you want to pull from a number of different DACs, this is very convenient. It can, however, be a little bit overwhelming, especially if you're new to SAR and not sure exactly what search terms you need to use. So you might want to start with the Vertex portal, and then once you get more comfortable, and if you'd prefer to work in Earth Data Search, then you've got that knowledge of how to find what it is that you're actually looking for.

Speaker 4: So the convenience here is that it provides access to all of the holdings from all of the DACs all in one place. Similar to the Vertex data portal, in the back, there is the Earth Access Python package that's doing a lot of the work. So you can make use of the Earth Access Python package yourself in your scripted workflows. At this point, I'd like to transition to a demo, and we will be working in the Vertex environment in a web browser. In Vertex, there's a little help button that links to all kinds of information, including video tutorials.

Speaker 4: So if I go through this too quickly or you kind of want to explore a little bit more in detail, I encourage you to check out those tutorials. And we'll also be searching for the Opera RTC S1 products. And there is a story map tutorial that goes step-by-step through looking for the Opera RTC products and also provides a lot of other information for accessing and making use of these data sets. So that's another resource to check out. So here I've opened a tab with ASF data search, and I can find that at search.asf.alaska.edu.

Speaker 4: By default, Vertex shows a geographic search for Sentinel-1 data. Now you'll note that there are almost 20 million files available. So using search filters are going to be really important The one thing that you'll need to do in order to download data is log in using an Earthdata login profile. If I'm already logged in, but if I wasn't, you would see a prompt to sign in. And if you don't already have Earthdata login credentials, you can register and it's quick and easy, it's free. And then that gives you access to all of the data that's served by all of the different DACs.

Speaker 4: So not just ASF, any DAC that you go to, you can use this one single unified credential. The first time you log on with your EarthData login credentials in Vertex, you will need to accept the end user license agreement for ASF, but you only have to do that once and then it's saved in your profile and then you're good to go. You can download anything from our archive.

Speaker 4: A great way to restrict your search results is using a geographic location. So I can zoom in by default. If I draw on the map, there's a box. I can also go to this area of interest menu and select the option to place a point, a line, a polygon, a circle, or I can also upload a geographic file, a geospatial file. So if I already have a shape file or GeoJSON or KMZ for my study area, I can just upload that, use that as my criteria for the geospatial extent, and I'm good to go. I can also just search for a location too, so if you're interested in a particular city or country or county, you can search the map for that location.

Speaker 4: There are also additional filters available, and these filters are different based on the mission that you're searching for, but one thing that pretty much all of them have in common is the ability to use a date filter. So if I'm interested in only recent data, I say the last year, I can select October 1st of 2023. And if I leave the end date blank, that shows me everything up to present. If I want to specify an end date, I can do that as well. At this point, already the number of expected files are down to just over 2000.

Speaker 4: For the Sentinel-1, another way to restrict the data is by selecting the file type. So if you're not interested in looking at the raw data, you don't have to have that in your search results. I can restrict it to just the single look complex data sets or also the GRDs. Now you'll notice that there are a couple different options or four different options for the GRDs. Most of the GRDs that are collected over land will be these GRD-HD, which is high resolution dual pole. But there are some other modes as well.

Speaker 4: So a mid-resolution dual-pole and then single-pole imagery, both in high resolution and mid-resolution. And usually that depends where on the earth the data set is being collected. So for most cases, if you're looking over land, the GRD HD will give you all the returns that you're looking for. But if you want to make sure that you have them all, you can check all of the boxes. At this point, I'm now at 656 files, and I'll go ahead and click the search button. And my search results come up in the pane at the bottom of the screen, and the results are listed on the left.

Speaker 4: And you can see that they're kind of combined, so if I click on one of these entries, it links to both the SLC and the GRD. Now, it's important to note that these browse images Browse images that you see are actually from the GRD. So if you only search for SLCs, you won't actually see any browse imagery. So if that's important to you or if it helps you make sure that it's covering your area of interest well, you can choose to also search for the GRDs as well as the SLCs, even if you're only interested in the SLCs.

Speaker 4: So if I click on an item in the list, it shows me the footprint on the map and also a browse image if it's available. I can also just click on a footprint in the map, and that will show select an item in the results list that has that footprint. One feature in Vertex that I really like is the More Like This button. So if I find a footprint that covers my area of interest well and I want to build a time series just using that footprint, I can just click the More Like This button. And it will restrict my search to only those images that have that same footprint.

Speaker 4: And basically what it's doing is just adding an additional filter that shows the path in the frame that you selected. So now I've got a list of 30 different scenes which provide 60 different files, 30 SLCs and 30 GRDs. If I'm ready to download, there are a few options I can Just click the download button here and download individual files one at a time. Or I can add things to my download cart. If I click on this icon, both the SLC and the GRD or any, if I haven't restricted it, all of the items that are bundled in that item will be added to my download queue.

Speaker 4: If I only want the GRDs, I would have to restrict my search that way. But I can also, if I want everything that's in my list, I can click the Q button at the top and add 60 files to my download. If I visit my download queue, I have the option to remove anything. So if there's something that I decide that I don't want, I can just click the X button. I can go through and download each file individually. Or I can make use of some of the bulk download features that are available at the bottom of the screen.

Speaker 4: If you're using Chrome, there is the download all option that is very handy. It makes use of the multi-threading capabilities that Chrome has in its browser. And you can just click on this button. It'll go through and download several items at a time, whatever your browser can support, until you've downloaded everything in your list. So for most people, that is a very effective way to do your downloading. There is also the option to download a Python script. And this is very handy if you're doing a search on a device that you don't actually want to use to download the data.

Speaker 4: You can just download the script, email it to yourself or to a colleague. And if they have Python on their machine and it's on their path, they can just run that script. It'll prompt you for your EarthData login credentials and then go through one by one and download all of the files in your list. There's also an option to copy URLs because these data sets are actually hosted in cloud storage. You can just copy the URLs, and if you already have a script that uses URLs to download online content, you can just add that into the script, paste that right in.

Speaker 4: All right, I'm going to clear out my download queue, and now we'll do a search for the Opera RTC products. So I'll just open a fresh browser tab, and instead of Sentinel-1, I'll select Opera S1. Now, there are two data sets currently in our archive that are generated by the Opera project.

Speaker 1: There are CSLCs and RTCs. I'm going to go ahead and draw an area of interest. And I've got 10,502 files that come up. Now, some of those are going to be CFLCs. And the CFLCs are not analysis-ready data. It's a co-registered FLC that does streamline the generation of INSAR products. But we're interested today in the RTCs. So a quick way to restrict your search right away is just to select the radiometric terrain corrected or RTC products instead of everything that's available. And this brings my total files down to 1,088 in my search area.

Speaker 1: Go ahead and search for that. Keep in mind that the RTC products are available starting October 4th, 2023. Currently, we've got just over a year of data that's available. I'll click this button.

Speaker 4: To zoom to my area of interest. And I'll point out one other thing here. So there is actually an option to set a limit of how many search results you return. I tend to have this set to 1,000 because I tend to be searching for a lot of things or larger areas. But if you have limited bandwidth and you can have a nicely refined search, you may be better off setting it to the lower option, the 250 files, because it will complete a lot more quickly. If you're searching for a lot of data and you want a lot of results, you can set it up to as high as 5,000, but be aware that that does take a while longer to complete that search.

Speaker 4: So if you're limited in bandwidth or if you have a slower connection, you may want to stick to.

Speaker 5: The lower limits.

Speaker 4: All right, so you can see right away that these footprints are very different than the footprint size of the full Sentinel-1 image. So there are a lot more results. And so again, filtering is going to be really important. Each one of the Opera products has quite a number of different files associated. And for most people, the files of interest are actually the GeoTIFFs, the polarization, and the GeoTIFF file. So for each polarization, there's one file. And what's cool about the Opera datasets is if I want to select everything from my queue and add it, everything from my list and add it to my queue, I can actually add by product type, which is super handy.

Speaker 4: But before I get to that, I want to point out that we no longer have the More Like This button. Instead, there's a Source Data button. So if you've selected a product, you can click the source data button and that shows you which of the SLCs that data was extracted from. So the burst that was used to generate the RTC, you can then place it within a specific Sentinel-1 SLC product. So the More Like This button isn't available there, but you can still do essentially the same thing by using this Opera Burst ID field.

Speaker 4: And there's a little icon right next to that that I can click and add the Opera Burst ID to the search. So now if I go into my filters, you can see just like similar to path and frame, it adds that filter and I've restricted my search now to only 32 files. So I have a nice little time series.

Speaker 5: Actually, I didn't press the search button.

Speaker 4: I have a nice little time series now of 32 scenes over this one area. And the nice thing about the burst IDs is that they always have the exact same footprint. So SLC footprints can kind of migrate around. They may include a few more bursts on the top or the bottom. So it's not always exactly the same footprint, whereas the bursts are always located in the exact same place every single pass. They might be included in a different source SLC, but the individual bursts always overlap with each other.

Speaker 4: So it's really handy for time series analysis because you have a very co-registered stack. So now if I want to add datasets to the queue, again, I can add all of the associated files by clicking this button here, or I can add individual files or directly download individual files. Or I can add everything to my queue, either all of the files that are in the product.

Speaker 3: Packages or, again, I can add by product type. So this is a really handy feature. I can just go in and add all my VV images, all my VH images.

Speaker 5: That's all I want.

Speaker 3: And now I've got a download cart that's ready to go that just has the products that I'm interested in.

Speaker 4: Again, the download options are exactly the same. I still have those. the bulk download options or the individual browser-based download options. At this point, I'm going to transition back to my slide deck. So now we're going to shift our attention to more of the on-demand products and the ways that you can order data if there aren't already analysis-ready products that meet your needs. ASF offers a number of different on-demand products for Sentinel-1. They include radiometric terrain correction, INSAR, which uses the full Sentinel-1 Level 1 SLC product as an input, and then also burst-based INSAR, where you can select just the particular burst that you want to process.

Speaker 4: We're also working on multiburst processing, and that should be available soon. And this allows you to select a suite of individual bursts that are all contiguous within the same path and merge them together before doing InSAR. So it will really allow you to target just the area that you want. If you are interested in more than just a burst-based size, but less than a full Sentinel-1 SLC level one scene, this will provide an excellent option. So the platform that's working in the background when you order an on-demand product is AFF Hype, which is the hybrid pluggable processing pipeline.

Speaker 4: Anytime you see this icon when you're searching in Vertex, that indicates that the product is suitable for an for creating an on-demand product. We have a lot of documentation available. This QR code takes you to our Hypedocs, which has all kinds of information on the platform, the products that you can generate, and includes links to all kinds of tutorials. We're going to move back into Vertex for another demo. But before we do that, I do want to point out two tutorials. the RTC on Demand and the INSAR on Demand tutorials.

Speaker 4: Each one goes through step-by-step how to search for, submit, and download on-demand products, either for RTC or INSAR. It's very step-by-step, so if you missed anything during this presentation, you can go back and work through it at your own pace. So here we are back in Vertex. Because I'm logged in with my EarthData login account, it not only gives me access to download the datasets, but it also allows me to do things like save searches, which can be pretty handy.

Speaker 1: So if you do a.

Speaker 4: Search and you want to save it for later or share it with a friend, you can save your search and come back later and select a saved search. So I'll launch this Ridgecrest search.

Speaker 5: And zoom to my area of interest.

Speaker 4: And you can see that my saved search has kept my area of interest, but also all of the filters that I applied. So it's got my date range, my file types, and all of that information is saved. I'll point out that if you do want to share a search, you can always just copy the URL entirely and send that to your colleague. If they put it into their browser, they'll come up with the exact same search results. So again, we're searching for Sentinel-1 data, and here I see my on-demand icon. If I click on that, the only option that it presents is the RTC option, and I can submit either the SLC or the GRD for processing.

Speaker 4: Now, the reason why I don't see INSAR as an option is because INSAR requires pairs of images, and this is just a single image. So there are a couple of different ways that that Vertex helps you find pairs of images to submit for INSAR. But for now, I'm just going to go ahead and submit my GRD RQC job, and you can see that now one item has been added to my on-demand queue. I'm going to use the More Like This button to restrict my search to just this footprint. And in this area, there was an earthquake in 2019.

Speaker 4: And I know that this acquisition was taken just before the earthquake occurred. So if you're doing an interferogram for a specific event like an earthquake, often you want to start by creating an interferogram that sandwiches the event as closely as possible before and after, so that the effects that you're capturing in terms of deformation are mostly due to just that one event. So I know that I want to use this particular granule as an input for one of the items in my pair. So in that case, a handy tool is the baseline tool.

Speaker 4: So I'll select that option, click the baseline tool, and then this will show me all of the other acquisitions within my date range that I selected that would be appropriate for creating an InSAR pair. So the selected item is shown with this big black dot, and then all of these other smaller gray dots are potential pairings. So this one was acquired 12 days after the initial image, and I'm going to go ahead and click on the on-demand button, which now shows me the option to submit for INSAR. So I can add this pair to my on-demand queue.

Speaker 5: And so I've got.

Speaker 3: Two things in there now.

Speaker 4: And I'm also going to add the 24-day pair as well.

Speaker 3: If I open my on-demand queue, I now have two tabs that are displayed and I can scroll down and see.

Speaker 5: What items I have in my list.

Speaker 4: So for RTC, I'd added that single item and for INSAR, I'd added those two pairs. Each one of these on-demand types has different processing options that are available. If you're interested in learning more, I definitely encourage you to take a look at that RTC on Demand tutorial. It goes through in much more detail than what I can cover today, what all of these options are and how that might impact your output. So you can choose between gamma-naught or sigma-naught radiometry, the three common scales for SAR, power, amplitude, and decibel, and then also the pixel spacing.

Speaker 4: You'll see here that there is a credit listed afterwards, and this is because to make on-demand processing available to the widest possible range of users, there is a limit of how much you can process in a month. So every user has 10,000 credits that they can use to process a product. And this gives you the indication of how many credits it would cost to actually submit this product to process to an analysis-ready output. So The 30 meter is the least expensive option, and for many analysis workflows, this will be sufficient.

Speaker 4: If that doesn't give you the resolution that you need, I recommend trying the 20 meter first before trying the 10 meters. 20 meter is actually getting quite close to the native resolution of the SAR sensor, and often there's not that much of a win in terms of resolution if you choose the 10 meter option. But the 10-meter file is huge. It takes a long time to download. It takes much longer to process, and it's just more unwieldy to work with. So if 30 meters isn't sufficient for your needs, definitely check out the 20-meter.

Speaker 5: Before going directly to the 10.

Speaker 4: There are a couple of different options that you can apply during processing, and you can also choose to include a number of ancillary data sets. The one that I really like is the RGB decomposition, which there's always a browse image that's included that gives you that nice color image that you've been seeing in the browse imagery. But this option gives you the full resolution, so the same pixel spacing as the RTC output. And this combines both the VV and the VH polarization into a color decomposition, which can make it a little bit more intuitive when you're looking at the product.

Speaker 4: So I will now go to my INSAR tab, and there are a few options available here as well. Again, the looks, that translates into the pixel spacing as well. So a 20 by 4 looks would give you 80-meter pixel spacing. 10 by 2 would give you a 40-meter pixel spacing. And again, there is a difference in the number of credits that each one of those options cost. If you're not familiar with the phase filter, I recommend that you just leave it as the default. And in fact, same thing with the RTC. If you're not sure and you're just trying to check things out, start with the defaults and.

Speaker 3: Then go from there.

Speaker 4: There is the option to apply a water mask. And if you're working in areas that have a lot of water, like if half of your scene is ocean or if it's bordering a great lake, especially these larger water bodies can really impact the quality of your phase unwrapping results. We do have a tutorial on water masking and how it can impact your results. So if you're interested in learning more about that, check out that tutorial. But in the meantime, if you are working in an area that has a lot of surface water, you may want to apply that water mask option.

Speaker 4: There are a lot of, again, ancillary data sets that you can choose to include if they're of interest to you. If you'll be using MintPi to do a time series analysis, there is this button that if you click it, it selects the options that you absolutely need if you want to use MintPi to run an analysis with these products. So you can also add other things in addition, but that's the bare minimum of what you need in order to do a MintPi analysis. So at this point, I'll submit my jobs. And it prompts me for a project name.

Speaker 4: And I can either select from a list and I can just start typing and it will show me the things that match. Or I can just type in an entirely new project name. As you get more and more items in your collection, it can be helpful to sort through to find what you're looking for if you've associated a project name. I do have the option if I don't want to submit all of the processing all at once, I can deselect one.

Speaker 2: Or the other.

Speaker 4: It stays in my queue, but it just doesn't submit it yet. So if I want to have different project names for the RTC versus INSAR, I can just uncheck that box and it will only send the INSAR for processing. I'm going to check them both and click the submit button. And you can see that I have a notification that they've been submitted. and I can click on this notification, and it will take me to my list of.

Speaker 1: Submitted jobs.

Speaker 4: I can also access that same interface from the On Demand button by selecting Submitted Products, or also by the Search Type dropdown if I select On Demand Products. That takes me to my list of submitted products. So here you can see the jobs that I've just queued up

Speaker 4: They were listed as loading and now pending. Once they're actively processing, that status will change again to running. And once they're done, you'll see some download options, the same download options that you're familiar with from before. I can add, click on an item and download it directly, add it to my cart. I can use this interface to add items to my download queue.

Speaker 3: Or again, I can just add all of the things that I have available for download to my cart and use the same download approach as for any of the.

Speaker 4: Other datasets that we've gone over. If I scroll down my list, you can see I've got some expired jobs. Once the processing is complete, you have 14 days to download your on-demand products and after those 14 days, the actual product goes away, but the record of your processing doesn't. So you always retain that in your history. And so if I forgot to download this product, I can go back, click on the expired button, and resubmit the job. This will send it with all the same parameters as before. I can just rerun it and make sure to download it this time before it expires.

Speaker 4: If I had processed a whole bunch of jobs and used the 30 meter pixel spacing, and then I come back a few weeks later and I'm like, oh, you know, actually, I think that it would be better to use 20. I'd like to try that. I can go into this expire button and add the whole project to the queue, and that will select all of the items that were in that project, list them in my queue, And now I have the option to change those parameters. So I can just select 20 instead of 30 meter pixel spacing. I can also just, if I'd like to reprocess with whatever parameters, any one of these jobs that I've processed before, I can also simply use that same interface for the on-demand processing.

Speaker 4: So the baseline search, gives me very specific pairs. And it's handy if I'm just doing very specific InSAR pairings. But for some use cases, you'll want to look at more of a time series approach. So instead of selecting an image and using the baseline tool, I can select that image and choose SBOS instead. This stands for small baseline. And in this approach, It helps you build a whole time series of individual InSAR products. So when I start, it shows me a dot for each one of the acquisitions through the entire Sentinel one time for this footprint.

Speaker 4: And it also uses the date range that I had set originally. So if I want to look at a longer date range, I can go in and change that. For a time series analysis, like if I'm looking at surface deformation due to groundwater extraction, for example, or something that happens slowly, it may be better to have a series of interferograms that you look at over time to get that longer signal. So I'm going to start back in 2018, and I'll actually start in May of 2018.

Speaker 4: So May 1st, and then I'll go through till this year,

Speaker 4: September 1st. Okay, so now I have all of these dots, which are all the acquisitions. And then in the date range that I've selected, there are these individual lines that indicate the potential INSAR pairs. So each one of these lines represents an INSAR product. And again, if I click on a line, it shows me the details of that pairing in my in the left-hand pane. By default, this is set so that only pairings that are 12 days or fewer apart are included. But if I want a denser time series, I can also extend that so that I have slightly longer interferograms as well.

Speaker 4: And that makes my time series much more dense. And if you look here, you can see in this time period, both Sentinel-1 A and B were up and running. Since the end of 2021, you'll see that there are fewer options, and that is an artifact of not having quite as much coverage in this area since the loss of 1B. So here I have 800 pairs that I can submit. If this is my time series, if this has everything that I want, I can just submit all of those pairs for on-demand processing. But I can also make a few more adjustments to hone in on just the data sets.

Speaker 5: That I really want.

Speaker 4: So there's also the perpendicular baseline. This is how far apart the sensor was in space from one acquisition to the next.

Speaker 1: And in general.

Speaker 4: You want those acquisitions to be quite close, that distance to be very small. We only present options that are within the critical baseline. So beyond this, you wouldn't be able to do a valid IN-SAR pair. So we only include things that are valid for IN-SAR. But you may actually want to tighten that criteria even more depending on your particular application. So this time, this slider here on the left side allows you to change the tolerance for the perpendicular baseline. So now it's down, close to 150 meters.

Speaker 4: And so any of these acquisitions that are quite a ways apart are excluded. Now, when you're doing an SBOS analysis, you do not want gaps. And so there's this warning here that came up because there's a gap now detected because there's not an acquisition less than 24 days that is within that perpendicular baseline setting. So I can actually add a custom pair here to bridge that gap. So I click on this button here, select the two acquisitions that I want to use as a pairing, and that automatically adds to my list a new pairing.

Speaker 4: And I can do that again to fill that more completely and fill the rest of my gaps as well. I'll mention one other approach. For areas that have strong seasonality, you may not want to be comparing interferograms throughout all of the seasons because there are just too many differences between winter and summer. And so there is this seasonal search, and this is available in all of the date searches as well, even if you're just searching for Sentinel-1 data or other data products. I can set this so that I only...

Speaker 4: include particular date ranges. So, this will acknowledge that I'm still looking for data between 2018 and 2024, but it will only show me the returns that are between May 1st and September 1st for each one of those years. So if I submit that, now you can see I've got all these seasonal chunks of INSAR pairs with the big gaps in between. So I can use that same custom pair approach to create annual interferograms. So that's a common way to bridge gaps in SBOS analysis where you don't want to have the same, where you want to have the same conditions on the ground, essentially.

Speaker 4: So May to May, June to June, that sort of thing, so that you're comparing apples to apples instead of snowy, wintry conditions to summery conditions. So to do that, I would just select my custom pair, the plus sign, and find an acquisition. So this is May 5th, 2020, and I'll find another May acquisition. So here's May 6th, 2021, and I can create that annual interferogram. And I could do the same thing here with July. So July 10th to July 11th, and I can build up my SBOS list in that way. Now, when I'm finished, I can either just download a CSV that shows all the pairs if I want to do some additional massaging and create a list to submit programmatically, or I can just add all of these 230 pairs to my list, and they're ready for download.

Speaker 4: I will point out also, you can add RTC jobs as well. So, if you wanted to make sure that you had nice amplitude images for all of the acquisitions that are represented in your InSAR time series. You can also just add all of the unique SLCs in this list for processing to RTC. And then you're ready to go and submit for processing. Before we wrap up, I'd like to go over one other option for finding on-demand products. And this is the event search. So under search type, I can click on event. And this platform monitors the earthquakes and volcanoes that are occurring throughout the world and submits automatically on-demand products for INSAR and RTC.

Speaker 4: So if you are interested in the impacts of a recent earthquake or volcano, this is actually a great place to start because there may already be on-demand products that have been generated for.

Speaker 5: That area of interest.

Speaker 4: So I can either just look at what's recent or I can click on an item in the map and that brings up all of the associated products that are available to explore and download. So that's another great way to access on-demand products without actually having to order them and wait for them.

Speaker 2: To be processed.

Speaker 5: Okay, so let's sum up.

Speaker 4: The Alaska Satellite Facility is one of 12 DACs that make up NASA's Earth Observing System Data Information System, or EOSTIS. Each one of the DACs has a specialty, and ASS specializes in synthetic aperture radar data. All of ASS holdings are free and easy to access. You can search and download them using interactive map-based web interfaces, including Vertex or Earth Data Search, but you can also search and download the data programmatically. There are a range of analysis-ready SAR products available from ASF, and there are a lot of options in particular for Sentinel-1.

Speaker 4: The OPERA RTC products and the ARIA-GUNW INSAR products are both pre-processed and available for download, but you can also submit any Sentinel-1 scene for on-demand processing to either RTC or INSAR. In addition, there are some other missions that include analysis-ready datasets including the UAVSAR and ALOS PALSAR products. And so that's also worth checking out. They are more limited areas, but that data is available as a resource as well. We do have a list of resources. So throughout the slide deck, there are lots of links and QR codes and things that you can use to track down more information.

Speaker 4: And then we also have this summary that includes some of those key references along with some other training options. So if you're looking to get a little bit more in the weeds, or again, just have another review of this similar content, there are some links to other options for that.

Speaker 1: And with that, I'll.

Speaker 4: Pass it back to Erica to go over the course summary.

Speaker 1: So in summary, you learn that the azimuth or the along-track resolution is different from the across-track resolution, that radar has three main parameters that are important, wavelength, polarization, and incidence angle. And the longer the wavelength, the greater the penetration depth. The length of the wave will determine the interaction with the objects in the surface. So objects or the wave will interact with objects that are approximately the same size as or the same length of the wave. Polarization provides information related to the structural characteristics of the objects on the surface.

Speaker 1: Incidence angle will influence the signal penetration into the target. The two surface parameters that influence the radar signal are structure and moisture. And the main backscatter mechanisms are specular scattering, rough scattering, rough surface scattering, volume scattering, and double-bounce. Radar images have geometric distortions in areas of complex topography, which you need to correct for. Speckle is that graininess effect inherent in star images, and it can be reduced, not fully removed, but reduced to multi-looking or with a spatial or temporal filter.

Speaker 1: Multi-looking and spatial filtering will reduce the spatial resolution of your image. And radar can be used for different ecosystem studies, such as, for example, mapping land cover, crops, wetland inundation, and soil moisture.

Speaker 1: SAR interferometry measures the distance from the satellite to the ground with high precision by using the phase of the reflected radar signals. The coherence of InSAR phase is sensitive to the surface or surface cover stability at the radar wavelength scale. Phase cycles in a repeat pass interferogram show change in distance to ground by half the radar wavelength. So 2.8 centimeters for Centome 1 and 12 centimeters for L-band, which NYSAR will be an L-band sensor. New pre-processed INSAR products enable user analysis of interferograms with few additional steps.

Speaker 1: And then finally, in-star measurements of surface motion are useful for a variety of geological processes, some hydrological processes, dynamics of glaciers, and other effects that displace surface or large structures.

Speaker 1: And as a reminder, there is a homework associated with this training. The homework is on the webpage for this training as of today. The due date is December 4th, and.

Speaker 6: A certificate of completion will be awarded to participants that attended all three of the live sessions, completed the homework assignments by.

Speaker 1: The deadline, and they will receive a certificate via email approximately two months after the completion of this webinar. And of course, for those of you that have any questions about the material that was presented today, please feel free to contact Dr. Franz Meyer or Heidi Christensen through their emails listed here. So with that, we've reached the end of this session and the end of this webinar series.

Speaker 6: I'd like to thank all of the participants for the questions.

Speaker 1: They've been sending, and we've been collecting them into a Google Doc that we will be sharing with you shortly as we begin our question and answer session. I'd like to thank Dr. Franz Meyer and Heidi Christensen for the excellent presentation and demos today. So let's begin our Q & A session now. Wonderful. So thank you for all the questions we've been receiving. Dr. Meyer and Heidi have been responding these questions. And we also have someone that's been helping respond some of these questions.

Speaker 1: He is from the Alaska Satellite Facility. Forrest Williams is also here with us. So thank you to Forrest for his support.

Speaker 7: All right.

Speaker 1: So they're all online. Let's start with question number one. I'm looking for free SAR data within the 1998 to 2006 time period. I'm preferring LBAN SAR data. Can you help me figure this out? Please go ahead. Any of our guest instructors, please go ahead and respond.

Speaker 2: Can you hear me?

Speaker 1: Yes.

Speaker 2: Go ahead. Yeah, so this is Franz Meyer. So I've been trying to answer this question. It is indeed a time period, the 1998 to 2006 time period. There was a mission that ended mid-1998. That mission is JERS-1. This was a mission launched by the Japanese Aerospace Exploration Agency. So until about mid of 1998, airband data from JERS would be available. Then there is, unfortunately, a gap in the event data coverage, at least from the perspective of openly accessible data sets. There's another Japanese sensor, ELOS Pulsar 1, that was launched in 2006.

Speaker 2: So then at some point in 2006, the time series picks up again. There is indeed a gap from 1998 until about 2006. from the perspective of globally available and accessible resources. There are some airborne sensors that are filling in this gap in some locations, but on a global scale, there is unfortunately not a lot of accessible data available for this timeframe.

Speaker 1: Great, thank you. That's quite unfortunate. But fortunately, since 2006, there seems to be quite a bit of continuous L-band data with some short gaps in between. Okay, question number two, is Amazon Web Service freely available?

Speaker 2: Yeah, I can take this one as well. So NASA has is storing all of their remote sensing data sets in the Amazon Web Services Cloud these days, and it makes these data sets available to all of its users for free. So you will be able to go in and discover, access, and download all of NASA's data holdings from the Amazon Web Services Cloud for free. So that's number one. If you want to do certain processing in the cloud, like Heidi talked about, the on-demand processing workflows that ASF maintains, NASA has actually invested some money into making these workflows available to approved users or to the user community.

Speaker 2: And as was mentioned, I think in the presentation, there is a certain amount of credits available, up to 10,000 credits per month. So that's going to give you a good amount of processing that you can do to create interferograms or RTC images. Also, that service will be available to the user for free. If you need more, so if the service that we provide for free is not sufficient for you, then, unfortunately, you have to look for other options. We can, as one option, build you a deployment of our services that you can run in your own cloud accounts.

Speaker 2: We would even maintain these services for you, but we would send you a monthly bill for the processing that you incur. These costs are not very large, but of course there's still costs that somebody will need to sever. The other option is to download the data sets to your local machine and do the processing there locally. There are a number of tools available that can give you similar type of products. Options include, for instance, the Snap platform, which is an open tool that ESA is releasing.

Speaker 2: There's also NASA-based tools, such as ICE-3, that can help you generate RTC data sets into the pharograms from Sentinel-1 and ISAR data in a more local way. processing approach. And then lastly, some of the community has been using a Google Earth Engine. I know there's a question about that later, so maybe I'll pause on that until we get to that question.

Speaker 1: Okay, thank you. Question number three, is it possible to apply SAR or InSAR data to change detection between two dates? And if it is possible, at which resolution? And is it possible to do it in Google Earth Engine?

Speaker 2: Boris, do you want to take this one? If not, I can answer the question. Yes, so there is certainly pairwise capabilities to do change detection. And so you can start with just a pre and post event pair. Actually, one of the really nice capabilities of SAR is that it's a very repeatable data set. So if you have repeat acquisitions and you see differences between these acquisitions, they are almost entirely related to physical changes on the ground. So it's actually a quite good change detection engine in that way if you want.

Speaker 2: Using pairwise change is good if your signal, your change signal is fairly large. If you are interested in more subtle changes, it is often recommended to not use one pre-event image, but take a whole time series of pre-event data sets and average them. This is because SAR has a lot of noise in the data set. We call that speckle noise. So if you look at a SAR image, there's a lot of pixel to pixel salt and pepper type noise contained. And the more you can suppress that noise pattern through averaging, the more sensitive of change you can expect.

Speaker 2: So that's a recommendation for these change detection techniques that often averaging before and after with improving your overall performance. You can also use INSAR data sets. What you would do here is you would look at the so-called interferometric coherence. If you say had a landslide between two INSAR positions, you will see a strong degradation of coherence in that area, and you could use the loss of coherence as a metric to extract pixels or areas that have changed.

Speaker 1: Okay, so the last part of that question, which is also question number four, can you access ASF data by using the Google Earth Engine platform?

Speaker 2: Yes, and so... So, you know, considering ASF data broadly as sort of, you know, international star datasets, there is a number of datasets that are available on Google Earth Engine. Google Earth is maintaining a complete archive of Sentinel-1. They are partly retrieved from ASF and partly from ESA. So those are available. I think there's a little bit of a delay with, you know, so that the latest data set may come in a few days after it was actually acquired, but they have the full archives available.

Speaker 2: As far as I know, there's also now ALOS 2, Pulsar 2, ScanStar data available on Google Earth Engine, and I assume, but of course in the end it's Google Earth Engine that has to make that decision, I assume they will be interested in also adding nicer data sets to their archive once they become available. So yes, Google Earth Engine has been pretty good at making these star resources available to the broader community. Can you guys all hear me now?

Speaker 1: Yes, Forrest, we can hear you loud and clear.

Speaker 2: Great, awesome.

Speaker 7: Yeah, I just want to mention one thing to add on to what Franz was saying, which is I put a link to a good paper that uses Google Earth Engine and the amplitude based on the one data that they have there to do some landslide change detection. That's a good one to check out if you're looking to do star-based change detection inside of Google Retention.

Speaker 1: Wonderful, thank you. Okay, question number five. How can we get a map of displacement from the interferometric rainbow fringes? And is there an easy tutorial or Python script that you can recommend. I'll take this one to start.

Speaker 3: We do have a tutorial that I've linked in the question and actually I might be able to post it to the chat right now.

Speaker 1: So it goes.

Speaker 3: Through our insert on demand products and talks about what the fringes mean and how that becomes then can be used for displacement measurement. France has also added some links to Jupiter notebooks that we have available. And what we're pretty excited about, and maybe I'll pass this over to Forrest for more detail, but we will be hosting the Opera displacement products, which, Forrest, do you want to talk a little bit more about that?

Speaker 7: Yeah, so the opera displacement products are essentially the end result of what you would get from an INSAR analysis, but it's using the best and newest methodology in the space to get as good of displacement estimates.

Speaker 2: As you can. That data should be going live at the end of the year.

Speaker 7: It'll be available for most of North America for the entire length of the Sentinel-1 archive, and we're very excited about And we're looking at providing some new visualization options for that data set so that you can explore the data interactively through Vertex.

Speaker 2: So, yeah, stay tuned in 2025.

Speaker 1: Super. Very exciting times in 2025 with so much SAR data now becoming not only available but analysis ready. Okay, question number six. Where can we check for updates on the release of ALOS 2 PALSAR 2, and is it currently available through JAXA? I'm hoping to incorporate that data set into a project I'm starting in mid-January.

Speaker 2: So I'm sure the data set is available through JAXA, through their distribution mechanisms. Regarding ASF, We are– we will be releasing shortly in December this year the global ALOS 2 Pulsar 2 ScanStar archive, which ASF has been working with JAXA on to host a mirror of that archive at ASF. So we are currently just finishing up skinjet pipelines that they show up. in NASA's centralized metadata framework and are then discoverable through the ASF interfaces. So if you are attending AGU this year, we'll actually have a presentation on this when you see some links to find these datasets.

Speaker 2: But the platform that Heidi presented earlier today will be hosting these ADOS to Pulsar to ScanSAR datasets. So these are the somewhat lower resolution products that they roll out with like a 50-meter-ish resolution. The higher resolution SOC, strip map SOC-type datasets, those are not publicly available on a broad scale. So those are available to approved science PIs.

Speaker 2: ASF is hosting some of those restricted datasets, and there's a mechanism that you can email us about to try to get access to those through NASA, but also JAXA is routinely offering PI possibilities, so calls for proposals that then, once approved, would give you limited access over limited areas, but free access to then these high resolution strip map type ALOS PALSAR 2 datasets.

Speaker 1: Okay, question number seven. Are there multispectral and SAR image pairs available at ASF? I'm working on SAR colorization.

Speaker 3: So most of ASF's holdings are SAR. We have one exception, which is the Avenir data. So the ALOS 1 mission had both the PALSAR sensor, but also the Avenir sensor. And the Avenir is a multispectral sensor. So that's kind of the one exception that we have. We do have both of those data sets. So it's not from very recent. That mission was flown from 2006 to 2011, but that is kind of our one exception. But if you are looking for both multispectral and SAR data, you can use the Earth Data Search Platform.

Speaker 3: which has data sets from.

Speaker 1: All of NASA's DACs.

Speaker 3: And so you can search all of ASS holdings there as well.

Speaker 1: As all the other DACs.

Speaker 3: So that would be a great place to search for some of the multispectral stuff as well as the SAR data all in one place.

Speaker 1: Great, thank you, Heidi. The next question, number eight, is it true that we use GRD data only for Pulsar and SLC data only for INSAR? Or can we use these data products in other ways?

Speaker 2: I can take a first stab at this. So SOC data is your sort of most versatile level one data product for SAR. The SOC data are a prerequisite for doing inter-processing, as you had correctly mentioned in the questions, but also for some of the pyrometric workflows. Also, those often use the phase information in the data sets, and so for some of the more sophisticated involved workflows, SOC data sets are also necessary. The ASF workflows to create geocoded image products, they actually work on both the SOC and the GRD product.

Speaker 2: I would also there recommend to start from the SOC mostly because the SOC frame actually has a little bit more coverage than the GRD frame does, and you ensure that you don't have tiny little gaps between the image products that you create. If you want to download data locally, and if your application is mostly mapping, the GRD product can be helpful, mostly because it is smaller size, there is less download volumes, and workflows that then fully geocode these products, they work both on the SOC and the GRD.

Speaker 2: So if you want to work locally and sort of data volume is a concern, then I would switch to the GRD product instead of the SOC. And I don't know, Forrest, if you want to add other stuff to this.

Speaker 7: Yeah, I would just say that when working with GRD data, I think my personal preference is like, when you have the ability to acquire it. RTC data is generally a better version of GRD, so I would just try and use RTC whenever possible.

Speaker 1: Okay, thank you very much. The next question, number nine.

Speaker 1: I would like to ask Dr. Meyer about the possibility of using SAR imagery to estimate water saturation in slopes.

Speaker 2: Right, so water saturation in the, so SAR is sensitive to variations in soil moisture. So what drives SAR scattering is the roughness of the interface and sort of the dielectric properties which are driven by the moisture content of the object that you're looking at. So soil moisture plays a role, and so typically when soil moisture increases, the rate of brightness also increases. To do soil moisture analysis on slopes, I would proceed accordingly. I would access SOC type data sets and at probably a longer wavelength.

Speaker 2: It depends a bit on the slope you're interested in, but in order to mostly see the soil properties and be less dependent on the vegetation above. A longer wave length gives you a better ratio between sort of the soil signature and the vegetation signature. So an L-band dataset might be useful. I would then perform RTC processing on these datasets to remove basically the slope impact on the radar brightness. And so once you have RTC processed the dataset, you should be able to use a time series of these products to analyze variations in water content in the soil over time.

Speaker 2: I hope that answered your question, and I hope I understood the question correctly.

Speaker 1: Thank you. So the next question, looks like we have a space enthusiast. Is NYSAR going to be launched from Cape Canaveral? Would love to watch it go up.

Speaker 2: Yeah, so unfortunately, no. So NISAR is being put together in a collaboration between NASA and the Indian Space Research Organization, short ISRO. And so the two organizations contributed to the mission almost 50-50. And one of the things that are a responsibility of ISRO is the launch. So ISRO will provide the launch vehicle, which is a GSLZ Mark II rocket, one of India's you know, heavier lifting launch vehicles. And the launch will also happen from India and not from the United States. So unfortunately, if you're close to Cape Canaveral, in this case, you'll miss out on the launch.

Speaker 1: Question number 11. For Sentinel-1 SLC data, apart from interferometry, you mentioned you can use other workflows that use phase. Can you please provide some examples?

Speaker 2: Yeah. Yeah. So there are other.

Speaker 7: Applications that use the phase component. In particular, Pulsar or SAR polar imagery. This can be really useful for doing things like trying to estimate biomass and trying to understand more about a forest canopy structure by using SAR. And then in addition, Other than interferometry, one thing to mention is that people use interferometric data in different ways. Some people are particularly looking at the displacement information that you can acquire from interferometry, and then other people are using the coherence information from the interferometric analysis to actually do change detection.

Speaker 7: So those are just a couple different additional ways that the coherence information, or sorry, the phase information can be used.

Speaker 1: Great, thank you. Question number 12, is there a data limit to acquire on-demand products?

Speaker 2: Yeah, I can take this one as well.

Speaker 7: Yeah, we do have to limit the amount of on-demand data each user gets just because, you know, we have a budget and we have to stay within it and we want to make sure everybody gets a fair shot at requesting the data they desire. For the full details, I've linked our cost page in this question, which goes through the details, but The general numbers are, if you're using our default settings, which have somewhat higher resolution, you can process 1,000 INSAR pairs and 2,000 RTC jobs per month. And then if you want slightly higher resolution products, that number will drop.

Speaker 7: But we hope that this is enough for most analyses. And if you need just like a little bit more than this to finish off your grad school project or something.

Speaker 2: Like that, please get in contact with us. we do have ways.

Speaker 7: Of providing a little bit more capacity to users.

Speaker 2: Great.

Speaker 1: Question number 13. Does an infographic exist that summarizes the observation platform, observations, data sets, including process levels, the tools where the data can be obtained? And these presentations are great, but I was also thinking about something that can be easily and quickly digestible by the general public as something very useful.

Speaker 2: Yeah, I can take this one. Actually updating a little bit the answer I gave. So there are actually some infographics that some communities have developed. There was an effort that wrapped up a few years ago that was called the SAR Handbook. This one was led by the Marshall Space Flight Center and resulted in a sort of free and openly distributed handbook that summarized the application of SAR to the mapping of the biosphere. And so as part of this, there's some infographics that are more starting not so much from the platform level, but from the application level.

Speaker 2: So basically say, if this is your application, what sensor would we recommend and what processing flow would we recommend? That was presented a few years ago at an AGU conference and I don't have the link handy right now, but I can try to find where that link is located and put that into the answer later. But so yes, this is a good idea. And so some people in the community have been starting to think about putting such infographics together. It's a bit tricky because if you ask probably 10 scientists, they may not fully agree what these infographics should look like, but attempts have certainly been made.

Speaker 1: Okay, question number 14. Is all the ASF data preprocessed?

Speaker 2: I can take this one again.

Speaker 7: So when we think about Sentinel-1 data, and NYSERDA in particular, we offer a varying number of levels of processing for this data. You know, we offer everything from the level zero completely unfocused data to SLC data, and then higher level products through a variety of options. For Sentinel-1, the offer products are a great source of higher level processed products. And then we also have our on-demand options for Sentinel-1 as well. When NYSAR launches, they're going to be producing a much larger suite of products, so you'll be able to get some NSAR and RTC data right out of the gate without any preprocessing.

Speaker 2: But if there are more.

Speaker 7: Details on particular flavors of Sentinel-1 or NYSAR data that you're looking for, don't hesitate to get in contact with us.

Speaker 2: Okay, thank you.

Speaker 1: Question 15, can we apply only data taken from the Copernicus browser without using the ASF, and what would that lead to?

Speaker 2: I can start answering this? Absolutely. I mean, we are very much appreciative of the work that ESA, the European Space Agency, and the Copernicus program has been doing to make these datasets available. to build a sensor that is freely available and monitors the planet so regularly. So their program is fantastic. They recently rebuilt their discovery platform to be now also cloud-based, so they have better accessibility and throughput these days. I'd say that what ASF does, has available that you wouldn't necessarily get currently through the Copiaticus platform is this ability to submit products for on-demand processing.

Speaker 2: So this built-in capability to create your interferograms and create your RTC images is at least currently not as built-in available in their discovery interface. But that being said, if they offer a very good service, We definitely encourage you all to check it out. We have a very, very productive and good relationship with ESA, and we can say anything but good things about their fantastic program.

Speaker 1: Super. Question 16, can I use only Sentinel-1 datasets for comprehensive BAM deformation monitoring?

Speaker 2: I can start this and then Forrest can chime in also. So for dam deformation, you would use an INSAR workflow. In concept, Sentinel-1 would support this kind of application. In practice, it will depend a little bit on your location and the characteristics of the environment around your dam. One of them is, what is the vegetation cover? If it's an earthen dam, how vegetation covered is it? Sentinel-1 is a C-band satellite. It doesn't penetrate as effectively through vegetation, so you may get some de-correlation issues at the C-band wavelength, depending on your vegetation cover.

Speaker 2: And a move to a longer wavelength, such as L-band, might be more useful. The other thing is, what is the orientation of your dam relative to the orbit geometry? these inside techniques are very good at seeing changes of distance between the satellite and a point on ground. And depending on the orientation of your dam and the orientation of the surface motion, there can be situations where the movement of the surface doesn't actually change the distance to the satellite much, particularly if the motion is sort of orthogonal to the look direction of the satellite.

Speaker 2: So these are also things to consider. You may find that either ascending or descending are more suitable, or it may also show that neither one of them is particularly good, and it might be better to use a ground-based INSAR instrument to observe that particular dam. So, yes, INSAR. can serve that use case and has been serving it. One good example, if you do a Google Scholar search, is there's a lot of work going on in the Netherlands where INSAR is used at broad scale to ensure the security of their dams.

Speaker 2: And I'll see if Forrest has additional thoughts.

Speaker 1: Yeah, unfortunately, Forrest had another meeting to jump to. So perhaps he can write his thoughts here when responding, elaborating on the answer. So, okay, let's go on to the next question. Number one. What data sets can be used alternatively? Which satellite sensors can be used in place of another? Are there any fused data sets currently?

Speaker 2: Yeah, I interpret that this is a follow-up question to the previous one. And so, yes, I would encourage to either use several sensors in parallel. So one approach is, if you're not sure about the performance of the different sensors, is to use all of them. So through ASF, you will be having availability for both Sentinel-1 and NISAR data stacks, and you could run displacement analysis for both of those wavelengths. If both of them are successful, then this actually gives you a very nice cross-validation of your results.

Speaker 2: If CBAN fails, then the LBAN approach may give you a different performance. And if both of them succeed, you can merge their time series in post-processing to gain a more detailed understanding of what the displacement behavior is like. So if you have the capability and through some of the ASF resources you would have, it would be recommended to actually try various wavelengths.

Speaker 1: Okay, question number 18. What are the advantages of looking for SAR data from Sentinel-1 versus Copernicus website? We've responded to that one already. How about let's move on to question number 19. What are or will be some use case examples for NYSAR?

Speaker 2: Yeah, so we are developing, will be developing workflows in the form of Jupyter Notebooks and other data recipes as the NISA dataset show up in our archive. In the meantime, in the answer, I put a link to the NISA website. The NISA website actually very nicely summarizes some of the science applications, but also some of the more sort of hands-on day-to-day applications that NYSER can support. They are highlighted in the form of sort of two-page white papers, and I encourage you to peruse them and see if you find the application you're interested in in that list.

Speaker 2: The NYSER science team has also developed a set of Jupyter notebooks that would allow you to derive higher-level data products and science products from Lysa data. These notebooks are available now on GitLab. I'm going to put the link to these notebooks in chat. And ASF is planning to work on integrating these notebooks into their Open Science Lab platform so that users could run these notebooks in the cloud for their data processing of Lysa data.

Speaker 1: Okay, question number 20. So you've answered this one already. What are the differences between GRD and SOC? But in addition to that, what is the difference between GRD and the Opera RTC products? Is there a better one, better than the other one for analysis?

Speaker 3: So I'll just weigh in on this one as well. So in general, if you are doing a time series analysis or looking at, you want to track change through time Often the RTC is a better option than a GRD. In terms of the Opera RTC products, that is a pre-processed data product. They are burst-based, so the footprints are very small. So that makes them very quick and easy to download. So that's an advantage for using those over the on-demand RTC that we offer at ASF. But if you do want to have the. RTC product for the entire Sentinel-1 theme.

Speaker 3: You can submit those for on-demand processing at ASF. The advantage there is that you can submit any product that we have in our archive. So the Opera products, they do generate RTC products for every Sentinel-1 acquisition, but they only started doing that in October of 2023. So there's just over a year's worth of data.

Speaker 1: In the archive.

Speaker 3: So if you're looking for data during that time, that's a great option But if you're doing a time series that expands beyond what is available through the opera RTC products, then you may find that our on-demand RTC products are a better fit for you.

Speaker 1: Thank you, Heidi. Question 21. ASF searches for all images which touch the area of interest boundary, and it returns then many images that are only on the edge and not useful. Can we tell ASF to search only for images which contain just the area of interest entirely? So that's a really cool request actually.

Speaker 3: We don't currently offer that as just like a one-click solution. When you do run the search, often I'll do this, I'll create an area of interest and launch the search and it has these frames that just barely touch and it's not that useful. So what I'll do is just go in and with the edit capabilities in the vertex interface, you can actually edit your area of interest. So you can pull in the individual vertices so that you can avoid those frames that are just barely touching your area of interest.

Speaker 3: And so I find that a very useful visual way to exclude those fringe scenes that are along the edges. But you can also use path and frame filters, which once you know which area is covered well by a certain path or if there's a very particular footprint, you can restrict your search to just those footprints or a single path or a series of frames. And so that can help restrict your search a little bit more. But I'll definitely pass along that idea to the Vertex team.

Speaker 2: Great.

Speaker 1: Question 22, can you provide application examples of how INSAR and burst INSAR are used? How do you know when it is suitable to use burst INSAR instead of INSAR?

Speaker 2: Yeah, I would say it's mostly a question of the size of your area of interest. The burst-based interferometry is particularly useful if your area of interest is small, much smaller than the original Sentinel-1 frame. So if you would have to rely on the full frame processing, you would have to create lots of data that you actually don't need. So these are use cases that are most suited for the burst-based processing. If you're looking at very large-scale applications, sort of this subsetting benefit that the burst processing brings you gets a little bit lost, and it might be equally useful to do the frame-based processing.

Speaker 2: The other thing where it's useful for is if you want to start trying something out. If you move to a new area and you want to explore the capabilities of Sentinel-1 for that area. You know, starting with a birthplace process just means that as a user you are spending less of your processing credits and at ASF we are not spending lots of money in the cloud to provide you data sets that in the end might not be so useful. So I personally do this if I move to a new area. I start processing just a first or two to explore a little bit the coherence conditions and what kind of displacements I might expect.

Speaker 2: And then potentially go to a full frame processing once I know that the data set is actually useful for the application that I'm looking for. Okay.

Speaker 1: We're coming to the end of our session. How about just one more question? Question number 23. For the Python script bulk download option, I believe you need the MetaLink file as well. Has this changed or is the MetaLink file not required?

Speaker 3: So the MetaLink file is not required if you're just using the Python download script. So that option where you can download the script that you can then launch to cycle through and download those products, those do not require the download. It's completely self-contained. So the meta link is for kind of external download managers. So if you're building something that's going to take advantage of that functionality, but for most individual users, the meta link is not required. You can just download the Python script and you're ready to go.

Speaker 1: Okay, how about one more? Question number 24, it's relevant to the previous one. In the case that I've downloaded information in the Python script, is there a guide or extra material to work with in a better way?

Speaker 3: So the only prerequisite for, if this is just for the download script, the only thing that you need is Python installed on your computer. And it can be any, we've used very basic libraries, so you don't need anything fancy.

Speaker 3: You can check the Vertex help if you want some additional information but it's a pretty straightforward operation. You just download the script, double-click it, and as long as Python is installed, you can launch that script.

Speaker 1: Okay, great. So we're coming up here to the end of our session. The questions, we have a number of questions that we didn't get to. I think we have a total of over 50 questions. So we will respond to questions. We will answer all of them on the Google Doc. And once they're reviewed, we will post the Google Doc on the training webpage. So this brings us to the end of this session. I would like to thank all of the participants from all over the world for your interest and enthusiasm for SAR. I'd like to thank the RSAT team, Natasha Johnson-Griffin, Brock Blevins, Sawun Hudson-Odoi, Jonathan O'Brien, Sue Monty, Sarah Kutschall.

Speaker 1: And more than anything, I would like to thank all of our guest speakers throughout this webinar series, Dr. Eric Fielding, Dr. Franz Meyer, Dr. Forrest Williams, Heidi Christensen. So before I close, I'd like to offer the forum for some closing words from Heidi and Franz. So Heidi, please go ahead.

Speaker 3: I would just like to thank you all for attending and do please get in touch, check out our tutorials and let us.

Speaker 1: Know if you have any questions.

Speaker 2: Thanks.

Speaker 1: Super. Thank you so much. And Franz?

Speaker 2: Yeah, just to reiterate, thank you for your interest in this imaging modality. This is a super exciting time for SAR. We are all sort of massively excited for the upcoming NISA launch. There is ALOS-4, a Japanese satellite just launched. ESA is preparing a number of launches. So this is a really exciting time for you to learn how to utilize these resources. And please be in touch if you have questions about the tools and services we provide. And we have a dedicated support staff that will quickly answer your questions to the best of their abilities.

Speaker 1: Thank you so much, Franz and Heidi. And yes, please make sure you check out the Alaska Satellite Facility. In addition to what was presented today, they have so many resources there that are of value. And with so much SAR data now available and what's coming up, I mean, it's just going to be a a new golden era of SAR. So really looking forward to that. With that, then, we've reached an end to this webinar series. There will be another SAR webinar series probably in 2025, around September, August, September.

Speaker 1: Hopefully we'll have NISAR data by then. So I hope everyone can follow and start using NISAR data Once NYSAR becomes available, it's going to be such an amazing data set. So thank you again. Wishing you all a great day. And until next time, bye-bye, everyone.

The Story