NASA ARSET An Introduction to NISAR

0:01 Speaker 1: Hello, and welcome to this webinar series harnessing NISAR, next Generation Radar Observations for Earth applications. My name is Erica Podest and I'm a scientist at NASA's Jet Propulsion Laboratory, where I study terrestrial ecosystems using ZAR. I'm also an instructor with the RSET program. We're entering a really exciting new era of Earth observation with the launch of NYSAR. This mission will provide an unprecedented view of how Earth's surface is changing, enabling routine monitoring of ecosystems, natural hazards, water resources, agriculture, and the cryosphere at a global scale.

0:45 Speaker 1: Although this webinar series is an introduction to NYSAR, we will also cover KEYSAR fundamentals to provide the context needed to understand the mission's capabilities, data products.

0:57 Speaker 2: And potential applications.

1:00 Speaker 1: The end of this training, you'll have a foundation for understanding how NICE data can be used to address a wide range of Earth science and decision support challenges.

1:14 Speaker 2: Before I begin, I'd like.

1:15 Speaker 1: To provide a bit of background about the RSET program.

1:22 Speaker 1: NASA's Applied Remote Sensing Training Program or r SET, offers free trainings on the use of remote sensing and model data, along with methods and tools for the global science and application communities. Our trainings are tailored to different experienced levels, ranging from introductory to intermediate to advanced. Ourset trainings focus on six thematic areas agriculture, disasters, ecological conservation, health and air quality, water resources, and wildland fires. Ourset trainings are either online or in person.

2:00 Speaker 1: All are free of charge and use open source data and software. Some of our trainings are delivered in both English and Spanish like this one, and all of our training sessions have a Spanish presentation available.

2:17 Speaker 1: Now I'll provide an overview of this webinar series. NISAR marks a new era of Earth observation. A fundamental challenge in Earth observation is monitoring change consistently across the globe, and NYSAR was designed to address this challenge by providing frequent, systematic radar observations that are independent of sunlight and that are largely unaffected by cloud cover. So these are the training learning objectives. By the end of this webinar series, participants will be able to identify the characteristics, capabilities, and limitations of NISAR data, recognize how NISAR data can be applied to decision making related to flooding, earthquakes, and landslides.

3:09 Speaker 1: Differentiate between the different types of NICSAR data products and their applicability for different use cases. Demonstrate how to access NICSAR data and tools via the ASFT platform to search, visualize, and analyze data provided case studies related to floods, earthquakes, and landslides. Visualized flooding with gcov files and generate landslide risk and volcano deformation maps for events of interest using the interfarogram, the GUNW files and qgis. Demonstrate how to access NICARSPAN data via the Israebunity platform and use available tools for basic visualization and analysis.

4:00 Speaker 1: Prerequisites for this training we suggest the fundamentals of remote sensing as well as an introduction to synthetic apeture radar and its applications. So there was one in twenty seventeen and there was also one in twenty twenty four. So these both of these contain background reviews of radar as well as InSAR. We will be focusing on InSAR in this training. The session three we'll discuss InSAR and as you all probably already know, InSAR is all about the face of the star signal.

4:42 Speaker 2: And this is a training outline.

4:44 Speaker 1: There are three sessions associated with this webinar series. Today is session one an introduction to NISAR. The next session will be on July ninth and it will be focused on NISAR data access and tools. That'll be led by the ASF the Alaska Satellite Facility DOC, which is the data archive center that holds all of the NISAR data. And then the third session will be on the sixteenth of July focusing on monitoring earthquakes, volcanoes, and landslides with nisar's InSAR capability, and that'll be led by my colleague, doctor Eric Fielding from JPL.

5:29 Speaker 1: All sessions are at the same time. There is one homework associated with this webinar series and it will open on the last session on the day of the last session on July sixteenth. The due date for the homework is August sixth. A certificate of completion will be awarded to those participants who attend all live sessions and complete the homework assignment by the do date. Now, let's focus on in this session and introduction to NISAR. I will be your instructor for this training. I'm a scientist at NASA's Jet Propulsion Laboratory, where I use microwave remote sensing, particularly SAR, to study wetlands vegetation, growing season in the northern high latitude regions, and land cover and land use change.

6:22 Speaker 1: I also serve as the community Engagement and Training coordinator for the NISAR mission, and I'm a member of nisar's science team supporting the Wetlands Science Area. Through these roles, I help connect the science applications and user communities with NISAR data and capabilities. The objectives for this first session. By the end of this session, participants will be able to identify the characteristics, capabilities, and limitations of NISAR data, recognize how NISAR data can be applied to decision making related to earthquakes and landslides, and identify the different NISAR data products.

7:08 Speaker 1: How to ask questions. To ensure we see your question, please write your question in the Q and A box, which you can locate it by doing a click on the three dots in the bottom right of of the window in the platform, and there you'll see a slider, a Q and A option, and you can also see a standalone slide o, tap or app. Write your questions are all your questions there, and we will answer them during the Q and A session at the end, and we will try to answer all the questions during the Q and A session.

7:49 Speaker 1: The remaining questions will be answered in the Q and A document, which we will post on the training website about a week after the train. Okay, so let's get started providing an overview of the NYSAR mission. So let's start out with why NYSAR. Many of the most important changes on Earth occur continuously and often in areas that are either cloudy or under darkness. And these changes occur at either very small scales on the order of centimeters or very large scales on the order of meters. And some of these changes that occur on the Earth's surface are earthquakes and tectonic deformation, glacier and ice sheet motion, landslides, surface subsidence due to groundwater extraction, and there are other changes to things like forest change and biomass, floods and wetland dynamics, and so to monitor these processes globally, we need a sensor that works day and night.

8:58 Speaker 1: I can see through clouds and can measure small to large changes in the.

9:03 Speaker 2: Earth surface.

9:07 Speaker 1: NASA stands for NASA ISRA which is the Indian Space Research Organization Synthetic Aperture Radar. It's a joint mission between NASA and ISRAE and it uses SAR to measure and understand changes in Earth's land, ice, water, and vegetation. The origins of NYSAR can be traced to CISAT in nineteen seventy eight, which was the first civilian l band T SAR mission in space, and it demonstrated the tremendous potential of radar observations for Earth science and applications. Although the mission operated for only a few months, it transformed our understanding of what TSAR could reveal about Earth surface.

9:51 Speaker 1: In two thousand and seven, the National Research Council de Catal Survey identified NYSAR as one of the highest part already Earth observation missions needed to advance our understanding of the Earth system through systematic global observations of Earth's changing surface and building. Then on decades of scientific and technological advances, discussions between NASA and Israel began in twenty eleven to explore a potential partnership.

10:22 Speaker 2: After several years.

10:23 Speaker 1: Of planning and coordinating, the agencies formalized their collaboration in twenty fourteen, and what followed was more than a decade of joint development, integration, testing, and preparation. Those efforts culminated in the successful launch of NYSAR from India on July thirtieth, twenty twenty five, and the animation that you see here illustrates the deployment sequence following launch. One of the mission's most impressive engineering features is its twelve meter deployable reflector antenna, which enables the collect action of high quality radar observations over a wide swath.

11:03 Speaker 1: The scientific motivation for NISAR spans multiple disciplines. Its observations supports studies of ecosystems and global biomass, I sheet and SIS dynamics, natural hazards, agriculture, hydrology, and many other processes that are critical to understanding our changing planet and supporting societal hazards and resilience. NISAR is the result of a highly complex and collaborative partnership between NASA and ISRAE, requiring close coordination across all aspects of the mission, from instrument development and spacecraft integration to operations and science.

11:47 Speaker 1: The table here shows the contributions from NASA and from Israel. NASA provided the alband star system, operating at a wavelength of twenty four centimeters.

11:59 Speaker 2: The radar system.

12:00 Speaker 1: Includes a twelve meter deployable reflector, a nine meter boom, and an approximately four meter long octagonal cylinder that houses the L band and SPAN radar electronics. And you saw this in the animation from the previous slide. And the two radar systems, the L and S ban radar systems can work simultaneously together. ISRAE provided the SPAN Star, which operates at a wavelength of nine point four centimeters, as well as the spacecraft bus and launch vehicle. NASA's Jet Proportion Laboratory also developed key electronics that enable communication between the spacecraft boss and the radar instruments, along with a high rate downlink system to transmit the very large volume of science data collected by the mission.

12:50 Speaker 1: The L band radar alone generates approximately four point four terabytes of raw data each day. ZO provided a separate high rate downlink system for the S band radar, meaning the mission operates two dedicated telecommunication systems to support the transmission of this very large.

13:13 Speaker 2: Volume of data to the ground.

13:16 Speaker 1: The partnership extends Beyond the hardware, NASA and Israel also share mission operations responsibilities and jointly support the science team and user community. Overall, NYSAR is one of the most capable Earth observing radar missions ever developed. The spacecraft has a total mass of twenty seven hundred kilograms and requires about six point two kilowatts of power, making it a high capacity, high performance system designed to support continuous global observations of the Earth's surface. NISAR has several unique characteristics that distinguish it from other spaceborn star missions.

14:02 Speaker 1: First, NISAR carries both an L band and an S band radar that can operate simultaneously, and because these radars operate at different wavelengths, so it would be twenty four centimeter wavelength at L band and a nine point four centimeter wavelength at S band, this means that they are sensitive. Each sensor is sensitive to different characteristics of the Earth's surface. Another key innovation is nisar's use of the sweepstar technology, and this represents a significant advancement over traditional spaceborn star systems.

14:39 Speaker 1: Sweepsar enables NISAR to achieve a wide two hundred and forty kilometer imaging swath while maintaining high spatial resolution and pleurometric measurement capabilities. NISAR also incorporates highly precise pointing and orbit control, both of which are essential for measuring subtle changes in the Earth's surface through a technique known as interferometric SAR or InSAR. Pointing control ensures that the radar beam is directed at the correct location on the earth surface with the precision required for repeat observations, and orbit control maintains the satellite's flight path so that the same locations can be observed repeatedly from nearly identical viewing geometries.

15:28 Speaker 1: NISAR is capable of repeating its orbit to within one hundred meters throughout the orbits, providing the stability needed for accurate surface deformation measurements. Another distinctive feature is nisar's high observation duty cycle, which refers to the fraction of the orbit during which the radar is actively collecting data. NISAR is designed for systematic global mapping, and the L band radar operates for approximately fifty percent of each orbit, while the SAN operates for about ten percent.

16:02 Speaker 1: This enables NISAR to routinely observe nearly all of Earth's land and ice covered surfaces. Finally, NISAR is a left looking radar system. I'll talk a little bit more about this in the upcoming slides, but this means that the viewing geometry provides comprehensive coverage of Antarctica and complements the observations from many existing SAR missions, which are predominantly right looking. And all of these unique features combined together make NISAR a powerful science and applications tool. NISAR operates at an altitude of seven hundred and forty seven kilometers in a near polar down dusk sun synchronous orbit.

16:51 Speaker 1: The satellite has a twelve day exact repeat cycle. However, by combining observations from both ascending and decent repeat orbits, the same area can be observed twice within each twelve day period. The acending pass crosses the equator at approximately six am local time, while the descending pass crosses at approximately six pm. Depending on the acquisition mode. NISAR provides spatial resolutions ranging from three to ten meters, and this animation illustrates the acquisition plan for the a sending portion of the orbits.

17:28 Speaker 1: The different colors represent distinct acquisition modes, which are selected based on the characteristics of the target being observed, scientific objectives, and practical resource constraints. For example, CIS is shown in pink and is imaged using a lower resolution mode because frequent observations are required and the features of interest generally occur at larger spatial scales. In contrast, North America is observed using the highest resolution mode with full polarimetric capability, enabling the most precise measurements.

18:06 Speaker 2: Most of the rest of the.

18:07 Speaker 1: World is imaged using a moderate resolution dual polarization mode that balances measurement quality with global coverage requirements. So that provides a little bit of background on nisar. However, before I start diving into the data and the different science and applications it can address, I want to provide some context about what radar is actually seeing and measuring and what radar.

18:38 Speaker 2: Is sensitive to.

18:39 Speaker 1: So this next portion will give a brief overview of radar remote sensing. So let's start with the very basics, and this is the electromagnetic spectrum. It's the range of electromagnetic energy that spans from very long wavelengths such as radio waves, which can be the length of a football field, to very short wavelengths such as gamma rays, which are the length of an atomic nucleus. Remote sensing sensors are designed to operate at specific regions of the electromagnetic spectrum according to their intended application.

19:18 Speaker 2: Microwave sensors operate within.

19:20 Speaker 1: The range delineated in this figure in black, and this range is at a much lower frequency range than optical and infrared sensors. So to put things into context, the wavelength of light is about three hundred and ninety to seven hundred nanometers, while for microwaves it's on the order of zero point three to one hundred centimeters. And because of this huge disparity in wavelengths, the features on the Earth's surface appear differently in the microwave range than in the optical range, and so is fundamentally different.

20:01 Speaker 1: The information content is fundamentally different than optical images. So what are the advantages and disadvantages of radar or optical remote sensing? First of all, radar has a nearly all weather capability, so you can see through clouds and most weather conditions. And I do talk about nearly because if you have very heavy thunderstorms that might affect the signal. It has a day or night capability, so it's an active sensor. We're sending out a signal and we can therefore acquire images of the Earth's surface, whether it's day or whether it's night.

20:43 Speaker 1: It can penetrate through the vegetation canopy or through a medium, so that signal that we're sending from the sensor can penetrate through vegetation, through soil, through snow. The radar signal has minimal atmospheric effects, and the most important part is that radar is sensitive to two things structure and moisture.

21:11 Speaker 2: Surface moisture.

21:13 Speaker 1: Optical is sensitive to the chemical properties of the surface, so very different information disadvantages of radar is that the information content is different than optical, so it's not necessarily intuitive, and sometimes it's difficult to interpret. There's the speckle effect that you've all heard about or know about. That's the kind of the salt and pepper effect in the images. It's not noise, it's actually part of the signal, but it is that graininess in the images that you have to deal with, and then the effects of topography that you need to deal with as well.

21:55 Speaker 1: So let's begin with some basic concepts of radar. It stands for radio detection and ranging, and, as the acronym suggests, radar functions as a ranging or distance measuring device. Essentially, a radar sensor emits a signal towards the Earth's surface and measures the time it takes for the signal to return. All imaging radars are side looking. If a radar were to look straight down, as depicted in the figure on the left, it would not be able to differentiate between two points A and B. In this case, the signal would reach points A and B at the same time and return to the sensor at the same time, which is why you wouldn't be able to differentiate them.

22:39 Speaker 1: However, when the radar is side looking, as shown in the figure on the right, the time it takes for the signal to reach points A and B differs along, allowing four points A and B to be differentiated. Synthetic aperture radars are side looking systems with an illumination direction, usually perpendicular to the flight line. So nisar is a left looking radar, and that means it's looking to the left of the satellite's direction of travel, and as a consequence, coverage at high latitudes differs between the Arctic and Antarctic in the northern hemisphere.

23:22 Speaker 1: The left looking geometry creates a larger gap around the North Pole, leaving regions north of approximately seventy seven point five degrees latitude outside the standard imaging swath, and this unobserved area is shown by the red circle in the North Pole coverage map. In contrast, the coverage gap around the South pole is substantially smaller, as illustrated in the Antarctic map. Let's begin by reviewing how radar images are formed. NYSAR measures to fundamental properties of the returned radar signs, the amplitude and the phase, so radar poses travel at the speed of light, and the sensor records only the portion of the transmitted energy that is scattered back toward the antenna.

24:13 Speaker 1: The amplitude represents the strength of the return signal and provides information about the reflective properties of the Earth surface. The signal power, which is proportional to the square of the amplitude, is calibrated and normalized to produce the radar backscattering coefficient or sigma not and this quantity is a measure of radar backscatter and is typically expressed in decibels using a logarithmic scale, so backscatter values can range from approximately minus thirty to minus twenty five deb for surfaces that reflect very little energy back to the radar, and these areas appear as dark tones in the radar images.

25:00 Speaker 1: And for example, the one you see here, there's a river that goes through the middle of the image, and that river has low backscatter because the smooth water surface reflects most of the radar energy away from the satellite. In contrast, surfaces that strongly scatter radar energy back to the sensor have higher backscatter values, often exceeding zero dB, and these areas appear as bright tones in the image and are typically associated with inundated vegetation and urban areas. The second measurement is phase, which describes the position of the radar wave within its cycle.

25:43 Speaker 2: At the moment the signal is received.

25:48 Speaker 1: Phase is usually measured in angular units such as degrees or radiance, and describes, as mentioned, the position of a wave within its waveform cycle. So for nysar, the radar wavelength is approximately twenty four centimeters for L band and nine point four centimeters for S band, and as the radar signal travels from the satellite to the Earth surface and back. It completes as you can imagine millions of wave cycles, and it's just not possible to count all of these cycles. Now viewed on its own, phase image often appears noisy and difficult to interpret, such as the example shown in the upper right here.

26:28 Speaker 2: However, one of.

26:29 Speaker 1: The most powerful applications of sarfase is interferometry or InSAR interferometric synthetic aperture radar, which measures the phase difference between two radar images acquired at different times or different viewing angles. Now, in the case of NYSAR InSAR, it's acquired the images are acquired at different times, and that's also known as repeat pass interferometry, and these phase differences between two different images can be converted it into extremely precise measurements of surface deformation, often detecting displacements of just millimeters to centimeters.

27:09 Speaker 1: Because nysar's radar wavelength is relatively short, as mentioned twenty four centimeters for lband nine point four four S band, even very small changes in the distance between the satellite and the ground produce measurable shift phase shifts, and this sensitivity enables scientists to monitor subtle surface movements associated with for example, earthquakes, volcanic activity, groundwater extraction, landslides, infrastructure deformation, and permafrost thought to name a few. In addition, the consistency or the coherence of the phase over time provides valuable information about surface stability.

27:51 Speaker 1: So if the phase does not change, that means the surface is stable and areas with stable phase behavior tend to remain unchanged, while a loss of phase coherence can indicate some sort of change, whether it's vegetation growth, surface disturbance, tectonic movement, or any other dynamic process. So now let's discuss the radar scattering mechanisms.

28:25 Speaker 1: So we discuss amplitude phase and nisar's interferometric capabilities. In session three of this training, we will take a deeper dive into nisar's insart data products and explore the different applications they enable. For now, let's focus on how the radar signal interacts with the earth surface. So, when the radar transmits a signal towards the ground, the way that signal is reflected or scattered depends on the physical properties of the surface. Different types of surfaces produce distinct scattering mechanisms which are fundamental to interpreting radar imagery the amplitude imagery.

29:07 Speaker 1: Now, if the surface is very smooth, such as a.

29:10 Speaker 2: Calm water body or a paved road.

29:12 Speaker 1: Most of the radar energy is reflected away from the satellite, much like a mirror reflecting light, and this is known as specular reflection and results in very little energy returning to the sensor, and as a result, these areas typically appear dark in radar images. As surface roughness increases, the radar signal is scattered in many different directions. Some of that energy is scattered back toward the satellite, producing a stronger return signal than say, a smooth specular surface, and this type of interaction is often referred to as surface scattering.

29:57 Speaker 1: When the radar signal encounters a volume, such as a forest, canopy, a snowpack, or even soil, the signal can penetrate into the material up to a certain point and interact with multiple components within it. So in a forest, for example, the radar wave may scatter from leaves, branches, stems, trunks, and the ground surface, and the combined contribution of these many interactions is known as volume scattering. A fourth important mechanism is a scattering mechanism is double bound scattering, and this occurs when the radar signal reflects between two surfaces that are roughly perpendicular to one another before returning to the satellite.

30:45 Speaker 1: So common examples include tree trunks standing in flooded areas, or buildings adjacent to a road, or flat ground surfaces. In these cases, a large fraction of the radar energy is redirected back toward the sensor. So the smooth surface causes the radar signal to be reflected away from the sensor, but then that perpendicular feature or surface causes that energy to be redirected back toward the sensor, producing a very strong return that appears very bright in the radar imagery. So in the next light, I'll show you these different backscattering mechanisms how they appear in a radar amplitude image.

31:34 Speaker 1: So this is an example of the different scattering mechanisms that I just discussed and how they look like in a radar image. So this here what you're seeing. This is a SMAP radar mosaic of the Amazon basin. It's all band HH polarization. What I want you to focus on are the different circles colored circles so black.

32:01 Speaker 2: Sorry.

32:01 Speaker 1: The blue circle is an example of a speculat surface. That's the Amazon River, so it's an open water body and the energy is scattered away from the satellite, which is why it appears very dark. There's a rough surface that's the red circle. These are areas where there's probably been cut down of trees due to grazing or due to agriculture, so there is some sort of roughness on the ground and part of the energy is scattered back towards the satellite. So these areas are not as dark as open water bodies, but there's some return to the sensor.

32:45 Speaker 1: And then there's the green circle that's volume scattering, So the signal is penetrating through the forest canopy, interacting with components within the canopy. And then there's double bounce that's the purple circle. And then this case it's inundated vegetation. The signal is penetrating through the canopy, it's hitting the standing water underneath the vegetation. That standing water is a specula reflector. It directs the energy away from the satellite, but then the tree trunks redirect the energy back towards the satellite.

33:20 Speaker 1: So as you can see. The radar can provide valuable information for different different things, different science and application areas. It's great for detecting inundated vegetation, it's great for looking at forest degradation, biomass measures of biomass, looking at areas that have been deforested, as well, looking at flooded areas, and looking at for example, if you have time series, you can look at agricultural growth. So these are some of the things that NICE can address. Now let's talk about radar and surface parameters as related to NYSAUR.

34:10 Speaker 2: So let's start with wavelength.

34:12 Speaker 1: That's kind of the most important parameter upfront, right, So wavelength is the length of the peak of one wave to the peak of the next wave, or really the position of any point in the wave to the same point in the next wave. In radar remote sensing, we often talk about wavelength rather than frequency, and this is because the length of the wave defines the interaction of the signal with the surface or the medium, and it's therefore easier to associate the capability of the sensor. Wavelength is inversely related to frequency.

34:48 Speaker 1: It is the speed of light divided by frequency, and the higher the frequency, the shorter the wavelength and vice versa. So the table on the right lists the common wavelength bands in radar and the letter codes were have been arbitrarily assigned, and these were developed during for military security, during the early stages of radar developments, so they're not in alphabetical order. But nisar operates at L band, so that's L band means there's a range of frequencies that define that band. Nisar operates specifically at a specific frequency.

35:31 Speaker 2: Within that range.

35:33 Speaker 1: That would be one point two five seven gigahertz with L band and three point two gigaherts with S band.

35:47 Speaker 2: There are two important.

35:48 Speaker 1: Things to remember about wavelength. The first is that the longer the wavelength, the greater the penetration through the medium that's whatever that is, vegetation, snow, soil. And the figure on the lower left shows that a shorter wavelength such as X band in this case, which is around three centimeters, will see the top part of a forest, probably the top third or fourth of the canopy, while cband will penetrate halfway through and L band will penetrate all the way through. Now this is limited by a vegetation density, but that's the general idea.

36:24 Speaker 2: So with nysar that has.

36:26 Speaker 1: L band and S band, S band will have less penetration through the canopy than L band, and the same goes for soil. So the second important thing about wavelength is that the length of the wave will determine the interaction with surface objects. In general, if an object or surface roughness is comparable to the size of the wave, then there will be interaction, the surface will appear rough and there will be energy scattered back. For example, little energy is scattered back from a surface with a high fluctuation on the order of nine centimeters if L band is used, and therefore that surface will appear smooth to the radar signal and dark in the radar image.

37:14 Speaker 1: However, that same surface will appear bright to the S band SAR image because it's on the order of of the wavelength the S band wavelength, which is nine point four centimeters. The table on the right shows examples of applications most relevant to specific bands.

37:39 Speaker 2: So if you're.

37:39 Speaker 1: Interested in forest type studies and you want deep penetration, L band is better than S band. S band can still be used for forest studies, but it's better for forests that are not as donce, so L band and P band are much more suited for, say, tropical forests where you have high biomass high density. The other radar parameter is polarization, and it refers to the plane of propagation of the electric field of the signal. So, irrespective of wavelength, radar signals can be transmitted and or received in different polarizations.

38:20 Speaker 1: The figure illustrates these planes of propagation. If you had a long string attied to an object and you stand at the very end of the string and move it up and down in a vertical manner, such as the figure in the top, then the signal is vertically polarized. If you move the string horizontally from side to the side, then that signal is horizontally polarized. And similarly, the antenna can receive either the horizontally or vertically polarized backscattered energy. By applying specific filters, NYSAR can receive both, so it can transmit and receive both.

38:57 Speaker 1: So there can be four combinations of both transmit and received polarizations. HH that means horizontally transmitted and horizontally received, VV, vertically transmitted and vertically received HV horizontally transmitted and vertically received and VH vertically transmitted and horizontally received. So nisar's polarimetric capability is important because it allows the radar to do more than just see the surface. It helps understand the orientation of the structures on the surface, and NISAR has different acquisition modes.

39:35 Speaker 1: Depending on the acquisition mode, either all polarizations are collected or just a subset of these polarizations.

39:46 Speaker 1: So we discussed two important radar parameters in the context of NISAR, that's wavelength and polarization, and the radar backscatter is in part driven by these two characteristics. The reflection of the signal is also driven by the characteristics of the surface, so this is super important. Radar is sensitive to two things. One is structure and the other one is moisture. So let's start with structure. In general, there are three parameters related to structure that's density, size relative to wavelength, and size and orientation.

40:20 Speaker 1: So we discuss size relative to wavelength. Here we have an example of an Austrian pine and that wavelength will or that wave that signal will interact with components in the surface that are approximately the size of the wave. Right, So in this example, at X band, which is about three centimeters. We're primarily seeing the needles in the Austrian pine at the top of the canopy. And with L band we're seeing more of the longer branches. So lband for nisar twenty four centimeters, we're looking at the longer branches, the trunk and so.

41:01 Speaker 1: And with PEA band even the longer structures that are more similar to the wavelength at PEA band.

41:09 Speaker 2: Also, the longer the wavelength.

41:11 Speaker 1: The greater the penetration through the medium. Then there's density that signal will penetrate through the forest canopy all the way down, depending on how dense your vegetation is. And there are many different studies that have looked at the saturation point in terms of the above ground biomass tons per hector, and this value differs where that saturation point is. It differs between different types of forests, tropical forests and boil forests, but the signal will definitely saturate at a larger tonnage per hector the longer the wavelength as opposed to the shorter wavelengths so at lband.

41:53 Speaker 1: L band is more suitable to look at dense forest s band forests that are like less dense and cband tends to be a better sensor for agriculture. And then their size and orientation we discuss polarization, So polarization provides information on the orientation of the components on the ground. The other surface parameter is the dielectric constant, which is controlled by the amount of moisture in the surface, and radar backscatter is influenced by the amount of moisture in the vegetation and the soil and the snowpack, and the larger the water content in the medium, the less the penetration.

42:44 Speaker 1: So the magnitude of the radar backscatter is proportional to the dielectric constant of the surface, So for dry surfaces it's on the order of three to eight the dielectric constant for liquid water it's around forty to eighty. This is frequency dependent, so the amount of moisture in the soil or in the medium can greatly influence an increase.

43:08 Speaker 2: In radar reflectivity.

43:10 Speaker 1: And so what you're seeing here on the right is an example of an L band. This is a JRS one image time series near Fairbanks, Alaska, in a place called Bonanza Creek, and you're seeing the transition of the land surface for the same area from early from mid September through late from mid February to late September. And what I want you to note here is the change in reflectivity when things are frozen in this part of the world. In Alaska, things are fully frozen in February, and then you've got a spring transition and then you've got summer when things are thought.

43:53 Speaker 1: So the radar backscatter increases when things go in this case from frozen, which is it has a very low dielectric to springtras when there is a large amount of water in the soil and things are thought. Now, let's focus on the science and applications that NYSAR can address, as well as some early results. Summarizing some of the key characteristics of NISAR include its dual frequency radar system operating in both L and S band. The satellite provides near global coverage with the exception of the high latitude regions, especially the Arctic.

44:36 Speaker 1: NISAR has a twelve day exact repeat cycle. However, it acquires two observations within each cycle, one in a sending orbit and the other one in d sending orbit. The spatial resolution of the data varies between three and ten meters depending on the acquisition mode. The mission has a baseline duration of three years for lband and five years for S band, and all data are or will be freely and openly available. L band data is distributed through the Alaska Satellite Facility Distributed Active Archive Center, also known as the ASF DOC, while the S band data, along with some L band data products, will be available through isroe's Bounity platform.

45:20 Speaker 1: Now given these capabilities, NISAR will support a wide range of science and application areas, including solid Earth processes, ecosystems, the cryosphere, soil moisture, and applied remote sensing applications. The mission has key science objectives organized into three main areas. The first is the cryosphere, which focuses on the dynamics of Earth, cis and major ice sheets in Greenland and Antarctica, as well as glaciers, and this includes examining how these components contribute to sea level rise.

46:00 Speaker 1: The second is ecosystems and this includes understanding the dynamics of carbon storage, landscape disturbance, wetland inundation dynamics, and agricultural systems.

46:10 Speaker 2: And it also provides a.

46:12 Speaker 1: Unique capability for estimating standing biomass and carbon stocks as they are distributed across the globe. The third is solid Earth science, which focuses on processes such as earthquakes, plate tectonics, volcanic activity and landslides, as well as related geohazards and together, these three science domains are supported by dedicated science teams working across each area. In addition, the mission includes important applications objectives, and these include understanding societal impacts related to hazards, water dynamics, ecosystem restoration and degradation, as well as their links to broader societal needs.

46:52 Speaker 1: In agriculture, the mission supports food security applications by monitoring land use and crop activity, distinct fishing between actively managed farmland, fallow areas and crops intended for food production, as well as monitoring soil moisture. There's also an urgent response capability enabling rapid tasking for events such as oil spills, storms, earthquakes and other natural hazards, and this allows the mission to acquire additional targeted observations and deliver data quickly to the user community, which is a particularly powerful aspect of the mission and in particular for India.

47:33 Speaker 1: There is a very strong focus on coastal processes.

47:40 Speaker 1: This is an example of a complete twelve day cycle at lband AHH polarization. The data used to create this mosaic have not yet been fully calibrated, and as a result, some radio frequency interference interference is present in certain areas. However, this example deserves as an illustration of nysar's L band coverage over land and ice, and it also provides a sense of the mission's ability to generate and manage a very large data value on the order of approximately thirteen hundred frames per day, about sixteen thousand frames per twelve day cycle and roughly half a million frames per year.

48:26 Speaker 1: This is a preliminary result of biomass with NISAR, and I mentioned previously that biomass dynamics are extremely important and these results illustrate the relationship between polarized backscatter and biomass. This information can be used at the calibration and validation sites shown on the left to generate biomass maps as illustrated on the right, and as you can see, the system is able to estimate biomass across a wide range of conditions, including very high biomass levels exceeding to one hundred metric tons per hector, and this capability is also important for improving climate cycle modeling.

49:08 Speaker 1: This is an example of wetland inundation, So this is a false color composite, and what you're seeing here in purple are inundated areas vegetated inundated area. So whatever is dark, that's river, that's open water, and all of the pink areas are inundated vegetation. And this inundated vegetation is something you would not see with optical data. Another example is a striking image related to ice sheets, and on the left you can see a series of ice streams flowing across Antarctica. The feature highlighted in the red box is shown in greater detail on the right, and this area corresponds to a topographic mound beneath the outsheet, and as ice flows through the region, it's diverted around the obstruction, which causes the formation and fracturing of the ice, and this process produces extensive crevass fields sometimes referred to as hairline crevasses, although they are far from small, spanning as you can see in the scale here several kilometers and so these features are clearly visible in the radar imagery and NYSAR, and they arise in part because the radar signal penetrates to some depth into the ice and it scatters from internal and surface structures and returns to the sensor.

50:39 Speaker 1: So measurements like these are used to study ice dynamics, including ice flow behavior and its interactions with both oceanic and atmospheric processes. So here's another ice related example where we directly measure velocity using interferometric techniques. On the left is a rate image of an ice field and you can see some areas appear slightly to focus and that's due to signal penetration effects, but the rougher regions clearly reveal the location of the ice stream and this example is from Jakobsham Glacier in Greenland.

51:19 Speaker 1: On the right, you can see a time series that's been generated by combining one image with another acquired twelve days later and forming an interferometric pair. So from this we can estimate ice surface displacement and derive velocity. In this case, the NYSAR data have also been combined with sentinel one data from a different viewing geometry to generate a full velocity vector field that shows ice motion of approximately three kilometers per year in this region.

51:52 Speaker 2: That's very fast flowing ice.

51:55 Speaker 1: Tracking the variability of this dynamic behavior over time is one of the primary objectives of NYSAR. The mission enables seasonal scale monitoring, whereas other systems have required up to a decade to produce a single velocity map. So this represents a significant advance in our ability to understand the ice dynamics and their relationship to climate. So here's a solid Earth example. Another important application of interferometry is tracking earthquakes, volcanic activity, and other forms of surface deformation to better understand processes that are these dynamic processes occurring within the Earth's crust.

52:38 Speaker 1: And this example shows a volcanic event in Ethiopia captured using.

52:44 Speaker 2: A time series.

52:45 Speaker 1: So in the lower right you can see the difference between two images acquired November ten and twenty second of last year, and the subtle color variations reveal surface deformation. So these are areas of subsidence and uplift, indicating subsurface magma movements, and so in some areas the magma chambers are losing material, while in other areas they're being recharged and may be approaching eruption. So in the subsequent twelve day interview, the data showed that an eruption did occur, did in fact occur, the magma that accumulated in the system was released and the diet that dyke effectively closed.

53:31 Speaker 2: So the colored.

53:32 Speaker 1: Fringes that you see are also known as interferometric fringes, and they represent the magnitude of ground displacement and in this case it's on the order of half a meter to a meter. And remarkably, we're able to measure this level of information deformation from space with centimeter scale precision using InSAR, and this capability is repeated every twelve days over the lifetime of the mission, providing, as you can imagine, a very powerful tool for monitoring these dynamic geophysical processes in near real time.

54:08 Speaker 1: Soil moisture is another key variable that is being measured by NISSAR. It's critically important for things like agriculture, hydrological models of epablic transporation, drought monitoring, weather forecasting, and related applications. And here you can see an example of the sole moisture product at field scale. Previous satellite based soil moisture measurements have generally been at relatively core spatial resolutions on the order of nine kilometers or more. In contrast, these observations that are approximately two hundred meter spatial resolution, and this represents field scale variability in soil moisture, which is highly valuable for the agricultural community.

54:57 Speaker 1: And of course, there are a number of applications that NISSAR and address. Here are some examples and you can access these white papers through the.

55:05 Speaker 2: Link on this slide.

55:10 Speaker 1: Now let's focus on the science and applications that NYSAR can address, as well as some early results. So this figure illustrates nisar's observation plan, which defines how the satellite acquires radar data around the globe. So rather than using a single imaging mode everywhere, the mission employs multiple radar acquisition modes optimized for different science objectives. And these modes also provide different spatial resolutions and polarizations, allowing NISAR to balance detailed observations with efficient global coverage.

55:48 Speaker 1: L band observations provide near global coverage averts land surface including islands, as well as the major ice sheets and cis. In contrast, the s BAN acquisition strategy is more targeted. Israel's primary focus is on the Indian subcontinent other regions of South Asia Antarctica, so S band data are not collected globally, just over targeted over specific areas, and in addition to those areas, S band observations are also acquired over global network of calibration and validation sites, and these locations provide them unique opportunities to combine L band and S band observations, helping scientists better understand the complementary information provided by the two frequencies and the benefits of using dual frequency radar measurements.

56:45 Speaker 1: The observation plan is reviewed and updated approximately every six months based on science team member recommendations, available mission resources, and other programmatic considerations to ensure that the mission contain unities to meet its evolving science objectives. These are the different NISAR data product levels, and like most NASA Earth observing missions, NISAR organizes its data into a series of processing levels, so level zero consists of the raw telemetry and signal data. Level one data products have undergone some initial processing and calibration, but remain in the radar's native or in the satellites native geometry also known as range Doppler rather than geographic coordinates.

57:37 Speaker 1: Level two products are geocoded, meaning they have been projected into geographic coordinates latitude and longitude, making them easier to integrate with other geospatial data sets. These products include calibrated radar backscatter, interferometric phase, unwrapped interferograms, and other geophysical measurements, so these are considered an ready Then Level three products are the higher level geocoded derived science products expressed in physical units, so the only global level three products generated by NYSAR is soul moisture.

58:16 Speaker 1: There are algorithms that are known as atbds in which stands for an algorithm theoretical basis document. These are algorithms that have been generated by the different disciplines and they have been validated over calibration validation sites and so these algorithms are focused on generating biomass vegetation disturbance, wetlands, in undation, crop area, surface deformation damage proxy, I sheet glacier velocity and c ice velocity and all of these atbds are openly available. So NISAR data products because system of backscatter and phase as well as interferograms that contain information on surface deformation as well as coherence, which is the stability of a surface on the order of centimeters.

59:12 Speaker 1: There are also pixel offset products generating using the backscattered data, and these identify large shifts in the surface on the order of meters. And finally there's the sole moisture product, which is a two hundred meter spatial resolution. You'll notice that many of these data products are derived only over Chrispear regions and the rest is global.

59:37 Speaker 2: Or near global.

59:39 Speaker 1: Now, if you're new to radar and you're doing, for example, ecosystem type studies lamp cover classification, or you're looking at fire burned scars, or you're looking at agricultural growth through time, you'll want to use the g COVE data products. So that is the geocoded covariance matrix. So that's basically the backscattered data at multiple polarizations, and that dataset has been radiometrically and terrain corrected, so it's analysis ready. If you're new to instar, then you'll want to use the gun W which is the geocoded unwrapped and it contains also the wrapped interferogram.

1:00:30 Speaker 1: So these are both level two data products that are produced globally. Nice art data status. In February of twenty twenty six, a large number of pre calibrated global data products were made available, and this release included over one hundred thousand Level one to three products totaling over five hundred terabytes of science data. In July day of twenty twenty six, the calibrated forward processing data will be released, so after the calibration phase concludes and planned software improvements are completed, the nice OUR project will begin forward processing of the newly acquired data.

1:01:18 Speaker 1: So forward processing is targeted to begin in sometime in July of twenty twenty six. Sometime during this month, the fully calibrated level zero to three products will be continuously generated from all data acquired from the start of forward processing into the future, and new data will be continuously published to ASF and made publicly available. Level one to three products will have a nominal latency of thirty six to seventy two hours from data acquisition to availability the Q three to four, so there's a calibrated back process.

1:02:00 Speaker 1: So after the start of the forward processing, the nice OAR project will produce calibrated level zero to three products from the backlog of all data collected from September twenty third, twenty twenty five, through the start of the forward processing in July twenty twenty six, and this processing will be completed sometime before the end of calendar year twenty twenty six.

1:02:31 Speaker 1: So the Alaska Satellite Facility is the data archives center for the L band data and Israel's Bounty Platform is the data archives center for the S band data as well as for overlapping.

1:02:46 Speaker 2: L band data.

1:02:48 Speaker 1: If you like to learn more, you can scan thequre code on this slide. Data can be downloaded through ASF's web based graphical user interface, or it can be accessed programmatically using an API, making it easy to incorporate nice our data into automated workflows. All nice our data are freely and openly available and are hosted in the Amazon Web Services Cloud, providing fast and scalable access for users around the world. Now, I'll show you a short demo on how to use the gcoff product to visualize flooding between two acquisitions, one during a high flood period and another one during the low flood period.

1:03:37 Speaker 1: All right, so the first thing we're going to do is we are going to go to the Alaska Satellite Facility asf Vertex tool. So go to Google and do a search on your web browser and we'll select. Here are our first results. Now, this is the Verse text platform and the default is Sentinel one. What we want is NICAR data. Note that this is uncalibrated data. So in February of twenty twenty six, there was a release of over one hundred thousand Nicar data products.

1:04:14 Speaker 3: These are all.

1:04:15 Speaker 1: Uncalibrated, so today we are working with the uncalibrated data. There are issues with the data, and I'll.

1:04:21 Speaker 3: Show you what some of those are.

1:04:24 Speaker 1: The plan, as mentioned is that sometime in July of twenty twenty six, late July twenty twenty six, there will be a release of the calibrated data NICEAR data.

1:04:37 Speaker 3: So let's just select NISAR.

1:04:39 Speaker 1: And again there are over one hundred thousand data products available.

1:04:46 Speaker 3: Let's do a specific search.

1:04:48 Speaker 1: This example is a flood visualization example. So let's just look at Columbia because I know there was a period of high flood and low flood there. And let's select specifically the GKOV product, so that is the radiometrically terrain corrected backscattered data in different polarizations in the case of overland in at least in this region that I specified, it's just AGH and HV primarily.

1:05:25 Speaker 4: So let's.

1:05:27 Speaker 3: Oops, let's do this.

1:05:34 Speaker 1: So I'm going to draw my area of interest, and under filters, I'm going to select the GCOV data product.

1:05:46 Speaker 3: And I will do search.

1:05:51 Speaker 1: And so there are two hundred and thirty six nice art files over my area of interest. So let's just zoom in. And I know I've already located an area where there's interesting flooding. So it's this frame right here,

1:06:10 Speaker 1: and that frame is this one. Whichever frame I select, you can see a quick browse image right here, but we're specifically interested in this one. So let's just click on that browse image and open a separate window where you can actually take a closer look at that browse image. So I'm seeing a lot of flooding here along the river. This is a double bounce to all of these purple areas. As mentioned, these are uncalibrated data. So what you see up here, this bright green area here, that's a radio frequency interference.

1:06:57 Speaker 1: So that's noise that's RIFY and that is going to be resolved in the release of calibrated data. That we will be doing later in July twenty twenty six. All right, So this is an interesting image because of the flooding, and so one thing I can say is I want to see images more like this. So I want to see the time series of images over this my specific area of interest. Okay, And so these are the images the exact repeat, exact area, and let's take a look. So the nice thing about Vertex is you can actually take a look at these browse images.

1:07:52 Speaker 1: Take a quick look. And what I'm seeing here, what I'm most interested in is the flooding, the right purple areas along the river that's inundated vegetation. Okay, So I'm gonna select, and you see this one has a lot of that inundated vegetation here along the river, while this one doesn't, and neither does this one. So I'm just gonna select October twenty ninth and December fourth. And in order to download the data, you go to the HDF five file. So there are a series of files here. This is the actual data file, and you download the file it's pretty large, or you can add it to your cart.

1:08:40 Speaker 1: So I can add these two images to my cart and download both of them at the same time, or I can individually download it. So I've downloaded the images already, and let's just go to qgis and take a look at the images and create an RGB. Okay, so I have the latest version of QGIS three point four four,

1:09:12 Speaker 1: the stable version, and what I do is I go to layer at layer at raster layer and then I select the file of interest.

1:09:26 Speaker 3: So I need to go here.

1:09:29 Speaker 1: I actually downloaded three different dates, but we're going to use two dates for this example. So we're going to use the October twenty ninth and December fourth. You can see it in the name right here, twenty twenty five twelve O four and this one is twenty twenty five ten twenty nine. Okay, now let's start with the ten twenty nine. That's the October that's where we saw the flooding. So let's call that the high flood image and we'll open it. Now what happens is qgis does not have a reader to recognize this version of HDF five that NISAR is using.

1:10:14 Speaker 1: That's still in development. We're developing the GDL reader still. So what you need to do in QGIS is you need to trick it so it thinks it's a net CDF five net CDF file. And the way you do that is you go to the front, to the very beginning and just type net cdf colon Okay. The other way to do this is just to change the extension in the name of the file instead of HDF at the end, just change it to dot nc. So there are two ways, either specifying that it's a net CDF in the path beginning or changing the extension from HDF to dot n C. So let's just go this way and we'll say add and there these are HDF five files, so they have a bunch of different files within it, and the one that we want is this one the frequency A. There are two polarizations HH and HP.

1:11:27 Speaker 1: In this case, we're just going to work with the HH data frequency A. So that's about that's a ten meter spatial resolution.

1:11:36 Speaker 3: And we'll say at layer.

1:11:39 Speaker 2: Close.

1:11:40 Speaker 1: So this is the file and it's always good to view it. And what we will do is, so it's loading right now. Since it's a it's a very heavy file, okay, and this is you see that it doesn't it looks very dark or very bright, So that has to do with how we're this qgis is doing the histogram stretch to visualize the image. So what we do is you double click on the file name. You go to symbology and just set the maximum. So this is a cumulative count cut based on percentile, and just select that.

1:12:34 Speaker 3: And that's our image. Okay. So let's do the same thing.

1:12:40 Speaker 1: We'll open the this is the October Let's open the December file same way at raster layer and here we'll just select the these are very long file names, but this is the December fourth. Will open it again. We add net CDF at the beginning, and we select.

1:13:12 Speaker 3: The frequency a.

1:13:15 Speaker 1: Hh band okay, and we say at layer okay. So here we have our two images. It takes a little while to load, and it's always really good practice to look at your images and just make sure that everything looks okay. So again we'll do the accumulative count cut.

1:13:50 Speaker 3: Okay.

1:13:53 Speaker 1: So the next thing we do is we will sub.

1:14:00 Speaker 3: At these images.

1:14:01 Speaker 1: And in order to do that, you go to processing TOOLBOXOPS Processing toolbox and clip clip raster by extent. So I recently used that, but you can go to raster extraction and select clipped raster by extent. Okay, and so we want to clip. Let's see, this is the October image.

1:14:30 Speaker 3: So let's start with that.

1:14:32 Speaker 1: And let's say that the extent that we want to clip is what we draw.

1:14:37 Speaker 3: Will draw the extent.

1:14:39 Speaker 1: So I'm going to go here and this area this is where the flooding occurs along the river here. So we'll clip this area and we'll say run, great, close, will rename. So you double click on the file name the clipped extent and you write click and you select the rename the layer. Okay, so this is October twenty nine, twenty twenty five.

1:15:17 Speaker 3: Aheh, clipped.

1:15:22 Speaker 1: Okay. Now we want to do the same thing for the next image. So again we say clip raster by extent,

1:15:33 Speaker 1: and we want to clip the December file, so it's this one you can see at the date here, and then clipping extent.

1:15:44 Speaker 3: We want to.

1:15:45 Speaker 1: Clip it according to the same extent from the first image. So we say calculate from layer and select the image that we just clipped, so it will clip according to the extent of that extent, and we say run. Okay.

1:16:05 Speaker 3: So let's just.

1:16:09 Speaker 1: Visualize this first of all, the October twenty nine Say how about it symbology and max cumulative concorde. Okay, that's the October twenty ninth, and then let's visualize the Let's rename this one.

1:16:48 Speaker 3: This is.

1:16:51 Speaker 1: December fourth, so we'll say December four, twenty twenty five.

1:17:03 Speaker 2: H h.

1:17:06 Speaker 3: Clipped and we will.

1:17:11 Speaker 1: Visualize it again there. Okay, so that's my clipped image. So we have two clipped images from different dates. The next thing we want to do is we want to convert these These are digital numbers. We want to convert them to decibels. And again these images, the terrain has been corrected and they're radiometrically corrected, so when we convert them to decibel, they're actually in sigma in gamma knat sorry. So in order to convert them to decibels, you go to raster raster calculator.

1:17:48 Speaker 3: And we'll say.

1:17:52 Speaker 1: We input the expression here which is ten times log ten and in parenthesi you then specify the file that you want to convert to dB. So we'll select this one and we need to close the parentheses and the output layer. We want to say where we want to put this file and give it a name. So this is December four, twenty twenty five. Agehe clipped, and we'll say it's dB okay.

1:18:28 Speaker 2: So.

1:18:31 Speaker 1: We press okay, and this is our dB image. In this case, I'm just going to set the values for visualization. So usually the values go, say from minus twenty five to let's say minus two, or let's say is zero,

1:19:00 Speaker 1: and you can play around with that for a better visualization. So that's the December twenty five. We'll do the same thing with the October file.

1:19:12 Speaker 3: So we'll say.

1:19:13 Speaker 1: This ten times log ten and the October twenty nine is the one that we want to convert to dB. So you double click and it fills it down here, and then the output layer again you specify where and you specify the name, and in this case it's October twenty nine, twenty twenty five.

1:19:44 Speaker 3: HH clipped dB.

1:19:49 Speaker 1: Okay, and we say okay. Again, we will stretch it from minus twenty five to zero and all of these bright areas that you're seeing again, that's in undated vegetation. So one thing we want to do is let's just create an RGB, and in order to do that, we will build start. We need to build a virtual raster because these images are independent. So let's go ahead and build a virtual raster, and we say place each input file into a separate band. So let's just select our input bands. So we'll select December four and the clipped dB, and then we'll select October twenty nine the clipped dB.

1:20:56 Speaker 3: Okay, so we have.

1:21:01 Speaker 1: Two bands, and what we want to do now is visualize that as an an RGB. So you double click, you select multi band color band one and let's just say.

1:21:20 Speaker 3: Minus eighteen and all of them men.

1:21:24 Speaker 1: So that's how we're stretching the visualization and max.

1:21:29 Speaker 3: Let's just say it's one one.

1:21:33 Speaker 1: So in the red band we have band one, in the green band we have band two, and in the blue band will have band one. Okay, so what we're seeing here is that all of the green areas those are areas where whatever is in the green channel has a higher value. So the green channel has the high flood the inundated vegetation. So all of these green areas that you're seeing, all of that is flooded vegetation that's there in October on the October twenty nine scene and not in the December fourth scene. So this is just a simple RGB with just two images.

1:22:27 Speaker 1: So a couple of important points to remember NYSAR is designed to monitor continuous Earth surface changes on the order of centimeter to meter scale globally. It has a capability to image the Earth surface day and night and to almost all weather or cloud conditions. The key focus processes include tectonic deformation, iet movement, ecosystem change, and agricultural monitoring. NISAR has an L band and an S band radar so L band twenty four centimeters S band nine point four centimeters. The satellite has an exact twelve day repeat cycle.

1:23:11 Speaker 1: However, there are two observations within every twelve days. One is in the ascending orbit, the other one is in descending orbit.

1:23:20 Speaker 2: It uses a two.

1:23:21 Speaker 1: Hundred and forty kilometer imaging swath using the sweep star technique, and the spatial resolutions vary between three and thirty meters depending on the radar mode. NISAR has a repeat pass interferometric capability and that allows for measurements of surface deformation on the order of centimeters. The polarization acquisitions vary depending on acquisition mode, and the core science disciplines cover cryosphere, ecosystems, and solid earth. NYSAR can also dress different applications and support disaster and hazard management.

1:24:04 Speaker 1: NISAR products are openly and freely available. The L band data is available through the ASF DOC That's Global data. The S band data, along with coincident L band data, are available through israe's bounity platform. The s band acquisitions are over India and targeted locations around the world. Data processing levels span from level zero, which is the raw data, to level three, which are the geocoded derived products, with soil moisture being the only Level three global product being generated. Level one and level two products contain information about amplitude phase.

1:24:45 Speaker 1: There are interferometric products, there's coherence, and there are pixel offsets. The Level one products are in the satellite native coordinates range Doppler, and the level two products are geocoded and analysis ready. The data are in HDF five formats. For those interested in doing ecosystem type studies, probably the best data sets are the g COVE and the GSLC data in terms of data status. In July and this month twenty twenty six, the fully calibrated forward process data will be released with a data availability latency of thirty six to seventy two hours, and then in late twenty twenty six, the back processing of the archived global historical data will be released, so that will be the data acquired by NISAR, spanning from September twenty twenty five through the beginning of this forward processed data release and looking ahead to session two.

1:25:54 Speaker 1: NISAR Data Access and Tools. You will learn learn how to identify where to access NICSAR data, recognize the data format and tools for reading the data, and apply tools for visualizing and analyzing NISAR data.

1:26:14 Speaker 2: Homework and certificates.

1:26:15 Speaker 1: There is one homework assignment associated with this training. It opens on July sixteenth and the due date is August six You can access the homework through the training web page. A certificate of completion will be awarded to those participants that attend all three live webinars and complete the homework assignment by the deadline. You'll receive a certificate via email approximately two months after completion of the course. If you have any questions, feel free to contact me via my email posted here.

1:26:51 Speaker 2: And here are two great resources for those.

1:26:54 Speaker 1: That want to take a deeper dive into NISAR. One is the NISAR Handbook the second version of it, and the other one is the nice Art Data User Guide by the Alaska Satellite facility.

1:27:08 Speaker 2: And the QR codes.

1:27:09 Speaker 1: The one on top is for the handbook, the one on the bottom is for the user guide that concludes the first session of this training. Thank you very much to everyone participating. Thank you for all the questions. We are now ready to start the q and A session. Great. So, there have been a lot of really great questions coming in a lot of conversation on the chat. We have been compiling your questions on a Google doc and we will be sharing that in just a few moments and we will start answering your questions.

1:27:51 Speaker 1: I don't think we'll be able we'll have enough time to get to all of your questions, but we will be responding to all of the questions on the Google doc and we will be posting this document on the web page. So let's start with question number one. Do NASA and Israel plan to launch other nice our satellites in the future to have nice our constellation increase the temporal resolution of the acquisitions. That would be absolutely fantastic. I am not aware of, however, I will consult if there are any sort of conversations and I will update this, but not that I'm aware of.

1:28:40 Speaker 1: Question number two. If we're interested only on structure, should we try to obtain imagery during dry periods of the year. And the response to that is that if you are studying vegetation structure, you should really try to acquire imagery during both the dry and wet seasons, because the structural characteristics of the vegetation can differ significantly between these two seasons. So I wouldn't just try to use imagery from the dry season. Okay, Question number three, how deep in the soil can sarro or insart penetrate?

1:29:21 Speaker 1: This is totally dependent on the frequency the polarization as well as the soil wetness. So again, the longer the wavelength, the greater the penetration. So L band will penetrate further into the soil than s band, and the wetter the soil, the less the penetration. But the nice or soil moisture product that I discuss represents the average volumetric soil moisture of the upper five centimeters of the soil. Question number four, you set nicsar can penetrate through soil? Can it be used to detect and study the heavy metal contamination in the soil?

1:30:04 Speaker 1: So this is an interesting question. It likely cannot be used to study heavy metals in the soil because again radar is not sensitive to the chemical properties of the surface, but rather to structure and to moisture.

1:30:24 Speaker 3: Now, if there's.

1:30:27 Speaker 1: I'd say a large area that is contaminated with heavy metals where the dielectric is different, right, so the wetness is different because of those heavy metals, then you might be able to see differences. You can't correlate it necessarily why one area is wetter or drier than another area, but you would be able to see those differences if it's large enough, right, So indirectly there might be some correlations. Question number five, can interferometry be used to study vegetation changes? How can nicser contribute to mapping deforestation and tropical forests?

1:31:15 Speaker 1: So, just for your information, we will be diving deep into that interferometric component of NICs are in session three, So my colleague colleague from JPL, Eric Fueling, will be leading that session. He'll be discussing providing a brief background on InSAR and discussing the ns A interferometric products and where they work well and where they don't, and the type of corrections that you need to do. However, to answer your question here, you know, the best nicer product to study vegetation changes is the backscatter product.

1:31:51 Speaker 1: So that's that's the amplitude and that's the GKOV the GKOV product the level took at different polarizations. So with that, you remember when I was discussing the fundamentals of SAR. You know how you have the scattering mechanisms and within a forest the signal bounces from many different components. So so the backscatter signal can detect biomass. You can use that to for for doing estimates, not detect, but you can do estimates because you need an algorithm uh.

1:32:34 Speaker 3: OF to do estimates of biomass.

1:32:37 Speaker 1: And you can also look at forest disturbance as well as uh wetland inundation, agriculture, agricultural growth through time.

1:32:50 Speaker 4: So to do those.

1:32:51 Speaker 1: So for ecosystem type studies, the best product is the g CoV product. Okay, And and deforestation is pretty straightforward in the sense that the change in the signal in the backscatter signal is pretty significant from an area where you have forest, you know, where you have a high back scatter relatively high backscattered to an area where you do not have forests.

1:33:15 Speaker 3: Where you have very low backscatter.

1:33:19 Speaker 1: Okay, the next question, question number six.

1:33:24 Speaker 3: Can nights are.

1:33:25 Speaker 1: Effectively distinguished between open water emergent vegetation and invasive aquatic vegetation in complex wetland ecosystems. So elban can very well distinguish open water because smooth water surfaces produce very low backscatter. Obviously, sometimes there's some sources of confusion. So in the open water, for example, if you've got wind, it's going to look a little rougher, it's not going to look as dark.

1:33:58 Speaker 1: So sometimes you have some sources of confus But L and S band can very well. Sea band can very well like distinguish calm open water. Now, L band is less effective at detecting low emergent vegetation, especially if that vegetation is sparse or has a very simple structure. And by that I mean, you know, if it's emergent and it's a couple of centimeters. Remember how I said that the signal interaction depends on the size of the component. So L band is about twenty four centimeters, and if that emergent vegetation is just a couple of centimeters four or five centimeters that it's going to look like an open water.

1:34:41 Speaker 1: Like open water, it's not going to detect it. But S band would be better suited for detecting that emergent vegetation as well as sea band. And then this person also asks about invasive aquatic vegetation. Hard to respond to that, I'm not entirely sure. You know, if it's underwater, obviously we can't. The signal is not penetrating through the water. But if it's above the water, and that invasive vegetation has enough density and is structurally different than the surrounding vegetation, then you might be able to detect it with radar.

1:35:27 Speaker 1: All right, the next question number seven, I've been studying the principles of weather radar, specifically how we manage signal clutter. Given that nisar is an L and S band system, how do the clutter removal and ground processing algorithms in NICs are compared to the techniques used in traditional Doppler weather radar networks like Opera.

1:35:50 Speaker 3: Okay, so.

1:35:52 Speaker 1: I'll elaborate on this response. I need to consult with some of my colleagues. However, radars are different than nice er weather radars operate at much smaller wavelengths because at that point you're looking at the interaction of the radar signal with with with water droplets, so you need much much the much smaller wavelengths and networks radar networks like opera.

1:36:27 Speaker 4: So opera is a so opera is a.

1:36:40 Speaker 3: Platform.

1:36:41 Speaker 1: It's an observational products for end users. So these are products that are being derived from remote sensing data and these opera products, and the next session number two will discuss some of the opera products they use. They don't use the weather radar data, they use Sentinel one data, they use optical data, and they'll be using NICs our data to generate these end user products which are related to surface water extent, surface disturbance, displacement, vertical land motion. Okay, the next question, how will nisar's dual frequency L and S bands are observations improve earth monitoring compared to existing missions and what applications do you expect to benefit the most.

1:37:44 Speaker 1: I think there are many different things.

1:37:47 Speaker 4: So for.

1:37:49 Speaker 3: L band sees.

1:37:50 Speaker 1: Deeper and larger structures, S band sees shallower and finer structures, So combining them will provide and more thorough characteristic of an ecosystem so L band will work well, say, or works well in areas where you have dense forest. S band works better in areas where you have less dense forest. So when you combine these unique capabilities or information content from these different bands, then you can have a more thorough characterization of your ecosystem. And what applications do you expect.

1:38:27 Speaker 1: I think that that's all. It's also up to the end user like what they want to explore in terms of the use of the complementary use of L and S band and even cband. You know we have there's Sentinel one data. So how can you exploit the unique information content in each of these different bands to come up with more thorough or generate more thorough either land cover maps or just different sorts of products. I think this is super exciting, this era of multi frequency sardin.

1:39:12 Speaker 1: All right, we have a short question number nine. Swath range, pixel spacing and resolution per mote. Well, okay, so the swath is two hundred and forty kilometers for most areas. Some areas are half swathed. The pixel spacing that varies between three and ten meters for the lower level products, and that totally depends on the acquisition mode number ten. I would like to ask how NICSAR has proved to be better in flood mapping and climate change analysis than the optical remote sensing satellites, for example Lancet.

1:39:53 Speaker 4: Okay, so NICSAR, so.

1:39:55 Speaker 1: We haven't yet proved to be better. It will be better, you know, because we're still working on the calibrated data. So the data that we have generated and released is pre calibrated, so we really haven't started that period of like really rigorous science analysis with calibrated data by the community. However, we expect this to be huge, the data to be huge in terms of supporting flood mapping because of that unique capability that nice AR has and radar and l band and penetrating through the vegetation canopy and detecting standing water underneath the vegetation.

1:40:40 Speaker 1: And this is something that just simply cannot be done with optical data because optical is really just seeing the top of the canopy. And in areas where you have very dense vegetation, like in tropical forests, yeah, you're not getting a sense of the extent of flooded in a day inundated vegetation. So and given nicsar's high relatively high spatial resolution and the temple repeat exact temple repeat every twelve days, but really two observations within those twelve days.

1:41:16 Speaker 3: Every six days. I think this is going.

1:41:19 Speaker 1: To be game changing really for the study of wetland ecosystems, looking at wetland inundation dynamics and extent all. Right, question number eleven, can NICER data be downscaled with multi input neural networks? Yeah, that's a great question. I don't have an answer to that. It might be I think this There are many, many, many things that are worth exploring with the NISAR data, and this is one definitely worth investigating to see what sort of results you can come up with. Okay, so please, I didn't understand the left side looking and the consequences on the North Pole.

1:42:14 Speaker 1: So because of the the the look geometry and and the orbit, so it's in the the the it's a Sun synchronous polar orbit. The north pole has larger gaps with just the way the orbit and and the fact that it's left looking. And so with NISAR we've really prioritized the Antarctica observations in Antarctica, which in the past I think most radar sensors at least openly available.

1:42:56 Speaker 3: Radar data out there.

1:42:59 Speaker 1: They're right looking, so they've prioritized observations over the Arctic, but less so over Antarctica. In this case, it's the other way around, so observations from other radar sensors fill in the Arctic, while nisar really fills that gap in Antarctica.

1:43:24 Speaker 3: And the South Pool.

1:43:26 Speaker 1: All right, the next question when will nice our data be available in Google Earth Engine. That's a really good question. I get that question a lot, and the answer is that we're not sure if it's going to be available in Google Earth Engine. However, we are discussing this possibility, and yeah, we hope in sometime in the future to have an update on whether that will be possible or not.

1:44:00 Speaker 1: Question number fourteen, which DM is used for the radiometric correction, So the DM that's been used is the Copernicus Glow thirty meter DM. Question number fifteen, how can we use the dual poll HH and HV available globally to differentiate in speculator DOBO bounds and volume scattering phase during different crop growth stages? So that's really that's a great question. I think there are different ways, right, So HV tends to be much more sensitive to the presence of vegetation than hh And one thing you can do is you can look at time series of HV to look at crop growth.

1:44:49 Speaker 3: Another now, if you want to.

1:44:50 Speaker 1: Look at the dominant scattering mechanisms, So if you want to understand whether within a pixel, so the dominant scattering mechanism is double bounce or surface scattering or speculator scattering. You can do quasi polarimetric decompositions.

1:45:14 Speaker 3: And this is a great point.

1:45:15 Speaker 1: One of the things I do want to do as we move forward into more intermediate type nice OUR trainings is to have some tutorials showing how to do these decompositions and what sort of algorithms are out there. Okay, question number sixteen, this is a great question. How accurately can NICs our data generate a DM in a mountainous region. So one thing to clarify is that NICSAR has this interferometric capability. It's a repeat pass interferometry. And so with repeat pass interferometry where within one hundred meters of the lap of the same point, so we're very close, we're always, you know, whenever we have that exact repeat, we're very close within one hundred meters from the last exact acquisition, and so with repeat past interferometry, this is much much more more suitable for measuring small changes in the surface.

1:46:25 Speaker 1: These are changes on the order of centimeters, but not suitable for generating a d M. You'll have errors on the order of many, many, many meters. So yeah, this is this is just to look at small changes in the surface, the type of interferometry that NYSAR can do. And this is something that my colleague Eric Fielding will discuss further in session three. Okay, question number seventeen, How can the polarizations inform on the orientation of an object? So it's you know, the best way to understand the orientation is to do an RGB.

1:47:19 Speaker 1: So if you have two polarizations, do a false color composite where you do HH in the red channel, HV in the green channel, and then HH in the blue channel, just to start out with just something very simple and that way you can understand the unique information content in the in each polarization. So things that are brighter in h H that means you know, you have you have a large you have horizontal components there than in the VV, say, for example, if you had VV. So, yeah, the best way is just to do an RGB of your different polarizations.

1:48:15 Speaker 1: Is there documentation? Question eighteen? Is there documentation for the different applications? Yeah, there is, and I'll share a link here, But there was that one slide I showed of the different applications that NYSA can address, and there are white papers associated with each of those, and they're very informative and interesting and they're very short, but they explain the capability of NISAR for addressing these different applications. Question number nineteen, How current is the current observation plan or rather how far into the future is it applicable?

1:48:59 Speaker 1: For mighty areas of priority acquisitions like the US Midwest change in the next year so I.

1:49:10 Speaker 3: So that's another great question.

1:49:12 Speaker 1: I believe, as mentioned, this acquisition plan is going to be revised every six years depending on the needs of the UH science team and the community that the science, you know, the science to address the science objectives.

1:49:32 Speaker 3: And I'll update when.

1:49:34 Speaker 1: It's going to be revised again, I believe it's later, yeah, later this year, but there might be if there are priority areas that new priority areas identified, say in the US Midwest, where maybe four polarizations would be of great benefit. Then that would be something definitely to consider the next time it's being updated. Question number twenty, can we analyze images of first past and second past jointly? So I think this question refers to can we analyze descending and images from descending and ascending orbits?

1:50:26 Speaker 3: And that's that's really good. That's a really good question.

1:50:30 Speaker 1: So the viewing geometry is slightly different between a sending and descending, you know, because we're looking towards the side. And so if you have a mountainous region, you should never combine ascending and descending, at least, you know, with the backscattered data. If your area is flat, then it should be okay. I personally do not like to combine ascending and descend.

1:51:00 Speaker 3: I deal with them separately.

1:51:03 Speaker 1: I'll generate a product with you know, one stack or the other, and then maybe compare or combine once the product is generated.

1:51:15 Speaker 3: All right, which band is good for soul moisture?

1:51:19 Speaker 1: So this is an interesting one, and and part of the plans for our set is to have a nicer

1:51:29 Speaker 1: training on the use of the uh SO moisture data products. So there are actually three different algorithms as part of the sole moisture product, and the I think these algorithms are using different polarizations or combination of the polarizations, but the best one will be h age. HV is sensitive to is more sensitive to the presence of vegetation and will have less penetration through the soil, so hh will have the greatest penetration through the soil, I think, followed by eight by vb.

1:52:22 Speaker 3: Okay.

1:52:23 Speaker 1: The next question, all right, question twenty two. Considering nysar's dual frequency capability, how do we plan to combine its deep penetrating L band data with cannopy sensitive S band data to solve a specific measurement challenge for research?

1:52:50 Speaker 3: That's a great question.

1:52:51 Speaker 1: As mentioned, you know, the data is there. The only level three product that is being generated globally is so moture. So there's a lot for the community to explore, and these are certainly things to explore. Is that complementary nature of these two different bands and how they can be.

1:53:15 Speaker 3: Used to solve.

1:53:18 Speaker 1: Something very specific to your research?

1:53:30 Speaker 4: All right?

1:53:31 Speaker 1: Question number twenty three, how are slope parallel and slope perpendicular motion components projected into a sending NYSAR line of sight displacement measurements for a west facing landslide, for an east facing slope with similar downslope and vertical motion, how would the line of site display the magnitude, sign and sensitivity differ compared to the west facing slope. Okay, so that's that's a.

1:54:04 Speaker 2: That's a great question.

1:54:05 Speaker 1: So in the ascending mode, the satellite is looking left, so it's looking towards the west, and so this is so I will respond to this question in the in the document, but as mentioned the third session, we will dive will take a deep dive into the instart products and specifically these sorts of questions that you're asking about, like related to the line of sight of the observation. You know, what sort of movement are we actually looking at with the interferrograms that are generated? Number twenty four.

1:54:52 Speaker 1: What are the most promising applications of NYSAR data for atmospheric science, climate research, and natural have a monitoring in the coming years. I think there are many So atmospheric science, as mentioned, you know, NYSAR is not a sensor to do atmospheric science. However, you know when you have measurements like soil moisture that can feed into eve apple transpiration models, into weather models, and so I think you know there's a lot of stuff that different sorts of research that the products that can be generated will support natural hazard monitoring.

1:55:44 Speaker 4: There's a whole.

1:55:47 Speaker 1: Area and you know within applications that addresses disasters and so looking at for example, with the instart products, you can measure small all changes in the land surface. So if you see an area, for example, landslides, if you're looking an area where you're seeing small movements through time, that might be indicative of an area that might be at risk for a landslide, right if it's in a sloped area. If you're seeing, for example, you can also detect floods, which is as mentioned, it's ideal, so natural flooding, flooded ecosystems, but floods related to disasters, and usually you can do assessments of the extent of the flood.

1:56:43 Speaker 1: When these sorts of things happens, there's a lot of cloud covers, so it's very beneficial that we have a sensor, a radar sensor that can see through clouds and provide those assessments. Obviously, those assessments can only be provided as long as there's an observation. But yeah, there are many many natural hazard monitoring looking at volcano movements around volcanoes which might be indicative of volcanoes becoming active, movement around tectonic plates and faults which might indicate areas where there's a higher likelihood of earthquakes.

1:57:27 Speaker 1: So there's a lot that NYSAR can support.

1:57:41 Speaker 1: How does nisar's question number twenty five? So I will answer that one separately in the Google doc. Let's carry on to question number twenty six. Usually it's a smap provides meter cubic meter over cubic meter, but here you show it as a percent Are these comparable?

1:58:09 Speaker 3: These?

1:58:09 Speaker 1: This percentage represents the amount of soil moisture per volume. What is the maximum depth that nicar can reach? Okay, so that's a great question. It is comparable. It's volumetrics, so it is the volume of water within a volume.

1:58:26 Speaker 3: Of soil and.

1:58:30 Speaker 1: The maximum depth that nicar can reach. As mentioned at the top, nice are the nice are soul moisture kind of represents so moition the top five centimeters. But you know, actually that's going to totally vary depending on many, many, many, many different things. Depending on the texture of the soil, depending on the amount of water. So basically, the wetter the soil, the less the penetration. But there have been studies from way back showing that L band can penetrate soils. And this was a study like in the desert somewhere where there was no vegetation and super dry soils.

1:59:18 Speaker 1: It showed that the L band signal could penetrate through that dry soil on the order of one and a half.

1:59:25 Speaker 3: To two meters.

1:59:27 Speaker 1: So in the case of NYSA, again it just represents the top five centimeters.

1:59:39 Speaker 3: All right.

1:59:40 Speaker 1: The next question number twenty seven, when you mentioned interferometry measurements combining NISA and sent and wood data, are you describing bistatic collections. No, this is not bistatic at all.

1:59:56 Speaker 1: This is a combination of NICs our data which is more contemporaneous, you know, measurements from some pre calibrated measurements since middle of September twenty twenty five, and the Sentinel one data in that example was used to go back in time. So for that historical context question, how is it possible to combine L and S band if their resolutions are different? So you know, the s BAND products have the same they're the same level products as as l band. It doesn't matter if their resolutions are slightly different.

2:00:55 Speaker 1: You can interpolate or extrapolate one to the other and you will be learning a little more about the spand data and its characteristics and how to access it. Even though ISRAE has not yet made available the spand data, there will be a short video about the Bunity platform that's where you can access the spand data through ISRAE, and part of that will discuss the characteristics of the spand data. Okay, Question number twenty nine does each level have different spatial resolutions? Yes, so the different levels have different spatial resolutions, and within each level there are different acquisition modes.

2:01:57 Speaker 1: So for example, level two you have acquisitions.

2:02:01 Speaker 4: Over ice and over land and.

2:02:06 Speaker 1: Those acquisitions are at different spatial resolutions. Question number thirty what products are built off NISA right now? Are there partnerships with hyper scalers like Google for Google's initiatives initiatives using AI to improve Earth, or other AI for good initiatives by others? So I will have to I will respond to this question in the document, but right now, as mentioned and as you know, was the plan for the mission. There's the different level products. That's what we're concentrating on, and the only level three global product is the sole moisture product.

2:02:56 Speaker 1: I think there are plans or discussions about developing other products, but I will respond to that in terms of what those plans are. Okay, so we are three minutes past the hour. Let me see how many more questions we have.

2:03:21 Speaker 3: Okay, so we.

2:03:22 Speaker 1: Have quite quite a bit more number of questions.

2:03:30 Speaker 3: I will respond to all of the questions.

2:03:32 Speaker 1: It will take me a little bit of time, so give me about a week to polish and respond to the questions and this will be posted on the uh ON on the page for this training. So with that, I want to thank everyone really for all the enthusiasm and all of the great questions coming in. And remember that there are two more sessions. Great sessions. Data Access will be session number two, that will be next Thursday at the same time, and then session number three will be in two weeks and that will be on InSAR.

2:04:15 Speaker 1: So wishing everyone a great day, great weekend, and until next week.