NASA Overview Access And Analysis of PACE Ocean Color Data Products

0:02 Speaker 1: Hello and welcome back everyone to the second part of this training on introduction to PACE hyperspectral observations for water quality monitoring. Today we are going to focus on overview, access, and analysis of PACE ocean color data products. Our guest speaker today is Dr. Morgan McKibben. She is the PACE applications lead and I will be introducing her in a few minutes. So here's the training outline. We are here today for the data overview and access and analysis part. And as I mentioned last week, there will be one homework posted on October 9th on our website, and it will be due on October 24th.

0:50 Speaker 1: And a certificate of completion will be awarded to those who attend all live sessions and complete the homework assignment before the due date.

1:01 Speaker 1: So just a brief review of some important points we saw in part one. Dr. Manino talked about space instruments. He described OCI, HAARP2, and SPEX1. Spectral, spatial, and temporal resolutions were discussed. And here is the figure that shows different instruments on the satellite. Also, it was noted that signals from the oceans are small. and differentiating between constituents requires additional information, spectral information, relative to what we have today. So the comparison we saw last week was that this is what Weir's sensor would see.

1:46 Speaker 1: So black is fish food and red is turtle food. So this is the spectrum from Weir's, whereas PACE sees this. So you have many, many more information points, the hyperspectral data. and you can see how this fish food and turtle food signature is differentiated. Here it's just one band where you can see the difference. Here you can consistently see differences in many spectral bands. Dr. Manino also talked about hyperspectral observations. They enable separation of aquatic constituents. As shown here, this is the wavelength and this is the reflectance.

2:31 Speaker 1: And different constituents are found from different parts of the spectrum. And here it is showing spectral signature of colored dissolved organic matter, phytoplankton, and then depigmented particles. All these are shown and based on that, then phytoplankton constituents are derived, composition is derived. Then we also saw that HARP2 and SPEX1 will aid in atmospheric correction useful for deriving water quality parameters. So correcting OCI reflectances, HARP2 and SPEX1 data can help. So one thing to note about PACE is that there is relatively low spatial resolution, about one kilometer.

3:23 Speaker 1: So that constraints use within inland and nearshore waters. hyperspectral algorithms need verification and require hyperspectral field measurements. So that is still going on. And overview of available PACE data products and issues in level two data products were also discussed by Dr. Manino. They're not completely calibrated and there's limited validation so far. So with That will start with today's part. Part two objectives are that by the end of this session, you should be able to explore the current and planned PACE data products for water quality monitoring, identify how to access PACE ocean cover instrument level one, two, and three data, identify applications and the usability of PACE data for monitoring water quality, And finally, analyze and visualize available OCI remote sensing reflectances, level two and level three water quality parameters using NASA's open source software, CDAS.

4:36 Speaker 1: So again, like every other session, you can start putting questions in the questions box and we will address them at the end of the webinar. And while you are listening to the webinar, as you come up with questions, please feel free to enter them in the question and answer question box. And we will get to them in the question and answer session at the end of the webinar.

5:02 Speaker 2: And the remainder of the.

5:03 Speaker 1: Questions will be answered in a question and answer document, which we cannot cover today. Then they will be posted on the training website about a week after the training.

5:17 Speaker 1: With that, I want to introduce Dr. Morgan McEban. Dr. McEban has a background in coastal and estuarine biological oceanography with an emphasis on satellite oceanography, harmful algal bloom ecology, and water quality. Currently, she works in the Ocean Ecology Lab at NASA Goddard Space Flight Center as the base mission applications lead. Dr. McEban has previously conducted applied research at a California nonprofit organization investigating linkages between eutrophication and harmful algal blooms in the San Francisco Bay with the goal of optimizing Bay nutrient management strategies.

6:04 Speaker 1: She has also worked two positions with the Applied Sciences Office of the Earth Science Division at NASA Goddard. First, as a NASA program postdoctoral fellow, where she investigated approaches to derive phytoplankton community composition from hyperspectral information in the Chesapeake Bay. And later, as a research scientist supporting research and development of remote sensing products for water quality applications. So with that, I invite Dr. Morgan McKibben to talk about PACE data products and also examples of PACE applications.

6:43 Speaker 1: Thank you.

6:49 Speaker 2: All right. So, thank you for that introduction, Amita. I'm here to talk about overview, access, and analysis of PACE OceanColor data products. So, today we're going to go over what the PACE Applications Program is and also give you a few use case examples from our program. We're going to be talking about where to access PACE OceanColor data products that are available now. And then we're also going to make sure you have resources and support to help you work with the PACE data, as well as keep up with the changes in PACE data access as they happen.

7:31 Speaker 2: So PACE is an extremely exciting mission. It has two main science goals. which are to advance and extend ocean biological, ecological, and biogeochemical data records, and also cloud, aerosol, and terrestrial data records. Because PACE is hyperspectral, it is the most advanced global ocean color mission to date. So first, let's do a brief review of the observatory. It was described in part one, and I will go over the terms again quickly right now. So PACE is a polar orbiting satellite that provides global observations every one to two days.

8:14 Speaker 2: It is a three-year mission, and it has three instruments aboard. The first one is OCI, or the Ocean Color Instrument. It is a hyperspectral scanning radiometer, and we also have two small but mighty multi-angle polarimeters. They're indicated in the inset there in the upper left-hand corner. One of them is HARP2, which is a wide swath hyperspectral polarimeter contributed by the University of Maryland, Baltimore County. We also have SPECS1, which is a narrow swath hyperspectral polarimeter contributed from a consortium of organizations in the Netherlands.

8:54 Speaker 2: So I do want to pause here and make note of the extreme excitement we have from PACE. PACE has brought a lot to us this year. After many years, many, many years of work, we now have data coming down from the most advanced global ocean color mission to date. And these world-class data products are free and open to everybody. So this is leading a new era of imaging spectroscopy of our Earth's surface. That means looking at our Earth's surface with this hyperspectral information. And this information is providing us with a foundation for really exciting scientific advancements in the coming years, which in turn is also providing us a foundation for really exciting advancement in practical applications of these data.

9:46 Speaker 2: And that brings us to talking about our PACE applications program and giving you a couple of water quality use case examples. So since we have a pretty broad audience here, I want to start at the top with how the PACE Application Program defines applications. So, applications are innovative uses of NASA PACE data products to improve decision-making activities and help provide practical solutions that meet societal needs. Applied research, it bridges the PACE data and the applications together. So this is going to provide us with a fundamental knowledge of how to scale and integrate PACE data products into users' policies, basically meaning putting the data to work for societal benefit.

10:46 Speaker 2: The communities that we work with include individuals and groups. These can be from public, private, or academic sectors. They can be national or international. And our applications can cover local to global scales depending on which project it is.

11:07 Speaker 2: Just like the data products from PACE, this graphic here is showing us that PACE application program, I'm sorry, the PACE applications, they span the land, aquatic, and atmospheric domains. covering multiple applications, thematic areas that are listed here. But today's presentation is focusing on the water quality applications of PACE. For example, how we may use PACE data to look at harmful algal blooms or ecological forecasting. So practical applications of NASA Earth observing data serve society.

11:52 Speaker 2: And this is much needed in our rapidly changing world. So our PACE applications program has the really important job of accelerating and promoting translation of PACE's advanced data into action. And we do this with kind of three main goals. So we build partnerships between PACE data producers and users. We also work to increase the accessibility and actionability of PACE data with training courses like this one to help people get their hands on the data so they can put it to work. We also work to demonstrate the societal value and utility of PACE data.

12:34 Speaker 2: And in recognition of how important the applications part of a mission is, the PACE mission was actually the first mission from NASA to incorporate the applications program and activities into the budget from the development of the mission through the life of the mission, actually.

12:59 Speaker 2: Several activities within the PACE Applications Program help us achieve these goals, such as community engagement. I'll give here three examples of how the PACE Applications Program engages with the community. One of them is our community of practice. So this is for anyone interested in staying up to date on the PACE mission, data, and applications. Basically, if you are here, you may benefit from joining us. And you can do that by simply sending an email to pace-community-join at lists.nasa.gov.

13:38 Speaker 2: Make sure you have join in the subject line. And then you're going to get a confirmation email and Once you confirm it, you will be getting updates from the mission on various events as well as updates to data access and so on. We also have a science and applications team. This is a competed set of NASA-funded scientists that work on algorithm development, applications, and validation and so on that changes out every few years. We also have our early adopters program. So these are researchers and others with applied projects needs that are teamed with specific stakeholders and together they develop and apply advanced PACE applications.

14:29 Speaker 2: We'll go over a few examples of the early adopters in a second, but first I want to go over kind of broadly the scope of the PACE applications program for water quality and water resources. So the new hyperspectral-based PACE ocean color data products, such as hyperspectral color fillet and phytoplankton community composition and pigments that are planned to come in the near future, they're going to advance water quality management and also our understanding of aquatic ecosystems. And they're going to do this by improving several important water quality applications areas.

15:07 Speaker 2: One of them is identification and tracking of harmful algal blooms. Another is assessment of fisheries and aquaculture health, evaluating and maintaining ecosystem health, as well as, you know, kind of looking at discrete events, like identifying oil spills or looking at the kind of water quality impacts we see after disasters like major floods. So as one example, this figure here, the National Oceans and Coastal Information Management Systems fisheries and aquaculture decision support tool. It will be incorporating phytoplankton community information once it's available from PACE in order to improve the decision making capabilities of this tool.

15:55 Speaker 2: And this is just one example of an existing tool that aims to be enriched and expanded through incorporation of PACE data. So these are some pretty broad examples of PACE applications in the water quality realm. Let's go over a couple of specific examples from our early adopters program. So first of all, led by Damian Brady, we have this Brady Aquaculture Site Prospecting Project. And here the application is applying paste products to better choose aquaculture site selection.

16:35 Speaker 2: And the significance here is if you can identify optimal aquaculture sites, you can save prospective shellfish growers both money and time. And they plan to improve their site selection tool by integrating paste products that describe phytoplankton size, which is an important factor in oyster feeding rate. And if you look at this map here on the right, this is an example of their oyster farm site selection tool. You can see that by the colors shown in the water that range from pink up to light green to dark green, it's telling you whether it's a better or worse site for the oyster farm.

17:18 Speaker 2: And the end users of this project include a rotating list of hundreds of aquaculture license holders in Maine. A second example of one of our early adopter projects is led by Dr. Antar Jutla. Antar and his team, they plan to integrate hyperspectral ocean color data into their existing predictive risk assessment model for Vibrio cholerae. So they're going to use the PACE advanced data products to better understand the relationship between phytoplankton, phytoplankton health, and Vibrio cholerae bacteria in both the Chesapeake Bay as well as Florida coasts.

18:02 Speaker 2: This is important because waterborne pathogens pose a significant threat to human health, particularly in developing countries. So accurate risk assessment is really critical for cholera release and mitigation activities. Some of the stakeholders of this product will include the World Health Organization as well as UNICEF, and some others. A third example from our PACE early adopter program is led by Dr. Bingqing Lu. Their application is assessing the potential impact of a changing climate on the water quality of the Northern Gulf of Mexico.

18:44 Speaker 2: So this includes advancing harmful algal bloom identification and forecasting for oyster farms found in the Gulf region. The significance of her early adopter project is that actionable science is needed in this region to understand how the water quality conditions of estuaries in Louisiana are likely to change as the waters warm. So this information is gonna help coastal managers make decisions about the natural resources and restoration practices in the Gulf. And PACE data are gonna help because of the high spectral resolution of PACE, which allows characterization of phytoplankton groups that move beyond the current abilities of multispectral satellites, which they hope will advance the science on harmful algal blooms, hypoxia, and food webs in the coastal Louisiana waters.

19:41 Speaker 2: So their stakeholders include local non-governmental organizations as well as state management agencies. And this emblem down here on the right hand side of the slide is for their Hypercoast tool, which is one part of their project that is actually open to all. I'm going to talk just a little bit about it. So the Hypercoast tool is an open source Python package. And it is for accessing and visualizing PACE data. This is a screenshot, a snapshot of some example output from Hypercoast of their region of study.

20:18 Speaker 2: So I'm going to orient you a bit on this map. Over here we have several lakes, such as Lake Pontchartrain. So this is in the northern Gulf of Mexico in the United States. And if you're familiar, this is Louisiana down here. And we have the large Mississippi River running down this direction.

20:42 Speaker 2: And this is a really great graphic that's demonstrating PACE data. So the green indicates where PACE data are. And it's showing that this PACE data is covering not only these large inland lakes, but it's also going through the estuaries and out into the open oceans. It's also showing us how spectra in the different water bodies compare. So this is an example of what we're talking about with PACE's hyperspectral data. So this is data going from roughly blue all the way through the rainbow into the red and showing us the reflectance values.

21:19 Speaker 2: And each of these is kind of like an optical fingerprint or a color of the water and how they vary in different locations. Let's look at Hypercoast in action in this very region. If you watch this video here, so this is right where we were with Lake Pontchartrain here in the middle, the tip of Louisiana right here. And if we watch what we can do with the Hypercoast tool, this is a live demo. Here you can click on Lake Pontchartrain and you can see that we have kind of bright, you know, brownish kind of colored waters.

21:56 Speaker 2: in the lake, and then if we compare this to another region, such as the open ocean here, you can see that it is bluer. Zooming back in, if we look at an estuary, we again see kind of this, you know, dark, this kind of brightish brownish water, and right there where there's perhaps more phytoplankton, the waters were a bit greener. And so this is giving you a really great visual example of how there are different spectral fingerprints in different water types, particularly when we're looking across the freshwater to coastal continuum.

22:37 Speaker 2: And so these three examples, they're really just a few of a fleet of early adopter projects that we have on board our PACE applications program. And now that we have PACE data coming to us, we really look forward to seeing how these projects are going to evolve over time.

22:59 Speaker 2: So on to the third part of our talk today. This will be the tutorial portion, and this is going to be information that will help you get started with PACE data. So the information in today's presentation is given at a level that's geared for those familiar with downloading, visualizing, and interpreting ocean color satellite data products. And I'm going to note that today's information is current as of this course in September of 2020-24. Make sure you know that data versions, the access points, and the resources will be evolving as the PACE mission evolves.

23:42 Speaker 2: And by the end of this final part of the section today, you will know which data tools provide access to PACE data. You're also going to learn where to find resources that are available to utilize PACE data, such as a couple of software and Python notebook examples. And you're also going to learn how to stay up to date with the data accesses and you're going to learn how to stay up to date with data access and resources as they change over time. So in the next three slides, I'm gonna define three important terms that you're going to need for the rest of today's session.

24:21 Speaker 2: The first term is data level. Let's walk through the different PACE data levels that will be available and their format. So here, data level is referring to the level at which it has been processed. So for example, after collection by the instrument, which is considered level zero format, that level 0 data is then processed through levels 1A through C. And that is essentially converting that raw instrument and telemetry data into the NetCDF4 format. And that data is then calibrated, geolocated, and registered to a common grid.

25:04 Speaker 2: Which brings us to level 2 data. These are derived geophysical science data products. Right here at level two and above, those are the science products that most all of you will be using. The next data processing level is level three. The level two data is temporally and spatially composited. As one example, here on the lower right, I'm showing a chlorophyll-A product that has been temporally composited over an eight-day average. And it has been spatially composited or binned to a four kilometer spatial grid.

25:47 Speaker 2: So finally, we have level four products. There's a range of different kinds of level four products. All of them, what they have in common is they're derived from various lower level data products or output from models. One really good example of this is net primary productivity models created with satellite data. Those net primary productivity data products are derived from inputs that include satellite chlorophyll A, temperature, and light variables in order to make that level four net primary productivity data product.

26:24 Speaker 2: Moving from data levels into our second term that's important to know for today, we have data status. So data status is a descriptor of data maturity. This is different than the data processing steps on the previous slide. I'll explain that a bit in a second. So diagnostic level data is the least mature, whereas standard quality is the most mature. You may hear that also called an operational product or a science quality product. So

27:02 Speaker 2: This table here, it details the names of all four levels of data status that PACE data come in. You're going to see these names in this presentation. And as I mentioned above, this table is basically showing how we move from the least mature diagnostic data to the most mature standard quality data. I won't read through the description of each one of them, but it just basically is describing to what extent the data have been calibrated and validated and otherwise checked for quality assurance. And I do want to note that right now, all of the currently available PACE data products are diagnostic, test, or provisional.

27:51 Speaker 2: PACE as a new mission doesn't yet have standard quality products released, but they are coming in the future. The third important term for you to know today is data product suites. What this means are these are related data products that are all packaged together in one file. The goal of having these data products in one file is to make downloading the data more manageable. So, for example, I'm showing here there's four data product suites in each of these four boxes. And it's four different level two ocean data product suites.

28:36 Speaker 2: So the name of this suite is Apparent Optical Properties. And that's including several data products that describe the apparent optical properties of the water. This one is our Inherent Optical Properties suite, our Biogeochemistry suite. And then finally, here at the end, we have our Photosynthetically Available Radiation suite. So this is grouping like products together into suites. I'm not going to read each of the names of the data products. What's important here is that you understand a data suite is a selection of data products grouped together.

29:13 Speaker 2: These abbreviations that I've added here let you know these will be found in the file name. So when you go look at PACE data, they will say OC underscore AOP, for example, for the apparent optical property suite. So that's how you know which suite you have is by looking at the file name. And there are more data product suites to come and we will be having a phytoplankton community composition suite. A net primary productivity suite is also planned, as well as several others. So please stay tuned for that.

29:54 Speaker 2: And so after going through those three terms, let's get started on talking about accessing the PACE data. Where do we go? A couple of important pages that I want to make sure that you bookmark include, first of all, we have this really useful PACE data access landing page found on our website. You can use the link here in this slide to go to it. So this is a live web page. It is a go-to source for current information. We keep it updated as things evolve. Another page that is also live and that you also should bookmark to keep up with changes is our data products table webpage.

30:37 Speaker 2: So this is listing all of our currently available as well as future data products. Each of the different columns have a description for each data product, its name, Also, the colors are coded according to whether it's available now or whether it's planned in the future. This column, there's also a column to describe the data status that we previously talked about. So, for those of you that have accessed ocean color data from NASA before, you may have this question. I am accustomed to getting ocean color data from the OBDAC, which stands for the Ocean Biology Distributed Active Archives Center.

31:21 Speaker 2: I'm accustomed to getting the ocean color data from the OBDAC via the level one and two, as well as the level three, four browsers that have been available on the ocean colored website. Is accessing PACE any different? And the answer is yes. At this time, access to different data products varies according to the data level, as well as the data status. For example, provisional level one and two data are available through the Earth Data Search, which I will go over in a second. Provisional test and diagnostic data are available from the OBDAC file search and the OBDAC level three and four browsers.

32:10 Speaker 2: We will go over quick examples of each of those as well.

32:16 Speaker 2: And what is available today? So today we do have level one provisional data, from all three instruments on PACE, so from OCI, HARP 2, and SPECS 1. We do have a limited suite of OCI level 2 and level 3 derived products as well. And we want to note that currently we are at our version 2 release, and we want to say this is preliminary data to be used with caution, if you keep in mind those data status levels. There will be frequent updates and reprocessing over time.

32:56 Speaker 2: So now moving to the three sources of PACE data. There's three different tools from NASA that you can use to access our PACE data products. The first one is NASA Worldview. The second one is NASA Earth Data. And the third one is the NASA OBDAC. Let me go over each one of these in a little bit more detail. So NASA Worldview. This is a phenomenal tool for fast, simple data visualization. Currently, there are two PACE products in Worldview. There's both OCI Level 2 Chlorophyll-A as well as OCI Level 2 TrueColor data.

33:40 Speaker 2: If you scan the QR code that I'm showing here, it will provide you a shortcut link to just these two data products in Worldview. If you go to that link, it brings up something that looks exactly like this. So this is worldview showing over here on the left, we have which data layers are viewable. We have chlorophyll A as well as that true color product. Both of them are turned on, both layers, and they can be toggled off by clicking the eyeball like icon over here. So right here on land, we can see that true color reflectance.

34:22 Speaker 2: So that's kind of, that's giving us an example of what it looks like to our eye. And then out here in the water, we have a chlorophyll-A product colored with a false color, color map with red meaning more chlorophyll and blue meaning less. So Worldview is a really simple and intuitive interface. At the bottom, there is a slider here that can help you move through time. You can also zoom in and out on your region of interest using the zoom toggle tools over here. And Worldview is also great for quickly outputting images of satellite data using this camera icon up here.

34:59 Speaker 2: And you can also do icons, I'm sorry, you can also output videos using this video icon here. I encourage you to go to the Worldview main page in order to learn more. There's some excellent learning resources there for learning how to use WorldView. So moving on to a quick example of our second tool, we have NASA Earth Data. So this is a comprehensive site where data from all of the NASA distributed active archive centers is located. This is for all NASA missions, not just PACE. This is cloud-based as well.

35:44 Speaker 2: And you can search across these all in one place. So as for PACE data, there is currently provisional data for all three instruments that can be downloaded at NASA Earth Data. If you click the QR code here, it'll take you directly to PACE data that's currently available on Earth Data. So let's take a quick look at a snapshot of what Earth Data looks like. So here I'm showing an example of looking through available OCI level 1c data granules on Earth Data Search. A granule is each of these kind of green outlined regions.

36:21 Speaker 2: So similar to Worldview, we have location visualization with the data shown here on a map. And then we also have a timeline here at the bottom where you can scroll through different times the data are available. Compared to Worldview, Earthdata is a much more comprehensive interface. It has a steep learning curve with a lot of options. I'm going to give you a few tutorial resources on the next page. So, first of all, if you're going to use Earthdata, importantly, you need to have an Earthdata account.

37:05 Speaker 2: This is required in order for you to download data. You can access it with the link on this slide. And once you've downloaded, or I'm sorry, once you have your EarthData account, I highly suggest going to this link on how to get started with EarthData. This is a great tutorial to help get you using the tool. And we also have available from our PACE OBDAC group, we have a tutorial on using Earth data to get PACE, to access PACE data products. It also includes a tutorial on how to use our third tool, which is going to the NASA OBDAC website itself and using both the file search as well as the level three and four browser search tools.

37:56 Speaker 2: Here at the NASA OBDAC, that's where you're going to find provisional test and diagnostic data. and which data level and also the data status available varies by the instrument and product.

38:14 Speaker 2: Here is a snapshot of the OBDAC Level 3 and 4 browser tool. The purpose of this tool is to visualize, extract, and download Level 3 PACE data products. In order to use this browser, it's important that under the product status menu shown here that you select provisional or test for your data status. And then you're going to be able to select the PACE instrument options here in the instrument pull-down menu. After that, you'll be able to select your product of interest as well as the timeframe and resolution that you're interested in.

38:54 Speaker 2: I want to note also that in order to use this Level 3 and 4 browser, you do need to have the NASA Earth Data account that I mentioned previously. The second tool available on our OBDAC website is the OBDAC File Search Tool. With this one, you can search currently available provisional diagnostic and test data. So this is including data that is not yet available on EarthData, some of the other data maturity levels. There's quite a few useful features in this search tool from basic to advanced. We encourage you to use this help button here to help understand them.

39:40 Speaker 2: For example, it has a great text generator for those of you that use Wget or curl. There's also advanced data search options as well. And again, this file search tool does require a NASA Earth Data account in order to download. And finally, I want to list a few of the resources that are available out there for you to work with PACE data. For those of you that use Python, we have some great data that helps you hit the ground running. You don't have to spend a lot of time getting your code going, because we've already written some of it for you.

40:20 Speaker 2: Follow the link shown here in order to use some of our Learn with OCI tutorial notebooks. We have one on data access, file structure at three processing levels, as well as others. A few of these are gonna be demoed in our third part of this RSAT training course. For those of you that are interested in using PACE data with the cloud, there is NASA cloud support. Namely, there is Earth Access. So this is a Python application programming interface or API that was created by the folks at Earth Data.

41:01 Speaker 2: And it's used to search for and download or stream data from the Earth Data Cloud. With the links here, you can get Earth Access information as well as download Earth Access Also really vital to those of you interested in using PACE data on the cloud is to check out the NASA Openscape Earth Data Cloud Cookbook. They have a fantastic set of resources there to help you understand how Earth data and the cloud work, as well as downloading and streaming data from the cloud. And finally, if you are a GitHub user and you want to share your PACE relevant code on GitHub, With the PACE community, we are encouraging people to tag it with NASA-PACE to help others find your code as well.

41:58 Speaker 2: Another phenomenal resource offered by NASA is CDAS. So this is NASA OBDAC data analysis and visualization software. It was developed and created in-house at the Ocean Ecology Lab. at NASA Goddard. There will be a tutorial later on today using CDAS, and you can download it at the link here. Currently, the newest version of CDAS is version 9.0.1, and what's important is version 9 now supports PACE data. CDAS versions prior to version 9 will not work with the various kinds of CDAS or will not work with PACE data.

42:45 Speaker 2: CDAS works on Mac, Linux, and Windows systems, and it can export paste data to various formats such as GeoTIFF, which is readable by GIS software, as well as KML formats that are readable by Google Earth, among others. I want to direct you to this tutorial video on CDAS 9, which does include... visualizing OCI's hyperspectral data, shown here at the left, and also works with polarimetry data, as shown above.

43:24 Speaker 2: As our final resource for today, if you have data questions, please visit the Earth Data Forum. This first link is for the Earth Data Forum's main webpage. But if you would like to look at PACE tagged questions, that means PACE specific questions, please go to the second link. You can go there and see what questions have been asked and what the various answers are. And the questions are answered by NASA affiliated personnel. So I thank you for attending today. Encourage you to stay up to date with all things PACE by joining our PACE community email list mentioned earlier.

44:14 Speaker 2: Also, keep up with our PACE website, including that data access and data product table webpages I mentioned earlier. We also have a great news and events section. And with that, thank you for having me today.

44:30 Speaker 1: Thank you so much, Morgan, for the very informative presentation about the PACE mission and about all the data products and how to access them. We also learned about PACE applications and early adopters programs, so thanks again for the excellent presentation. Next, we are going to have a short demonstration of how to examine OCI Level 2 and Level 3 data using CDAS. CDAS, as Morgan mentioned, is a NASA software for processing optical remote sensing data. And here is an ARSAT training that we did earlier this year about overview of CDAS 8.4.1.

45:12 Speaker 1: It has a lot of details about processing, analysis, and visualization of optical remote sensing data. And here is a tutorial that Morgan mentioned about CDAS latest version 9.0.1. So we're not going to talk about CDAS in detail, but we're going to learn how to download level three and level two data, OCI data, and then how to examine them using CDAS. So this is the slide that Morgan showed. and we're going to focus on these two sites. So NASA Earth Data for level two OCI data and OB-TAC for level three OCI data.

45:58 Speaker 1: So I'm going to go to this site. First, let's get level three data. And for that, I'm going to share my screen here. This is OB-TAC level three and four browser. And here you can select different instruments, different products, different time periods to select data. So we are going to start with products selection and we're going to go to provisional data. As Morgan mentioned, these are the data which are derived from standard community consensus algorithm for PACE. And let's pick PACE-OCI here and we're going to keep chlorophyll concentration.

46:48 Speaker 1: So for example, we will look at level three chlorophyll concentration from base OCI. Here you can see all the daily maps.

46:58 Speaker 2: For now, we're going to look at.

47:00 Speaker 1: Eight day data so that we have relatively more samples for cloud-free images. For example, I have chosen for level three We're going to focus on a large inland lake. This is Lake Victoria in Africa. And for that, I have chosen this eight day period, 28th August to 4th September, as you can see. It's pretty clear here and you can see Lake Victoria from OCI. So what you can do is extract or download data Here you can pick time period that we just looked at, so 28th of August to 4th of September. And when you say, okay, one more thing, let's pick mapped data and say extract.

48:04 Speaker 1: When you say extract, this window opens and you can now choose your geographical region. So here I'm going to draw a box around Lake Victoria and then say review order. When you review order, it shows the domain that you picked. Here is Lake Victoria and here the period, 8-day period that we chose is shown here. And now you can say extract data. Once you do that, you will be taken to this we have received your request page and you can go down and go here. There is a link provided here. So you have to have an account on NASA Earth Data.

48:51 Speaker 1: So you will also receive an email that your order has been received. If you click here, you will see that the orders that you may have placed they show up here so the latest one you can say confirm once you make sure that this is the one that you ordered and when the data are staged you will get an email or you can see in here it shows status so you can get once you have that you can click on this manifest text file and it is saved on your computer and it gives you URL of either one or several files and you can just copy paste in your browser and download this file to your computer.

49:44 Speaker 1: So I have done that for this eight day period over Lake Victoria for level three chlorophyll A consultation from OCI. And we will look at that in CDAS. Before we do that, let's also download level two data. So for that, I'm going to select NASA Earth Data. To go to NASA Earth Data, to Data and Earth Data Search. This is the site that Morgan shared with us. Here we are going to pick Level 2 OCI reflectances and they are in a package, it's Apparent Optical Property or AOP. So here I'm going to search with base AOP and it tells you that the data are there and there are so many renewals, you can filter it with date.

50:51 Speaker 1: For example, again, I'm going to pick one day in August, say 28th of August. So we're going to pick that applied so these are the granules and for level 2 data we're going to focus on on an estuary say so let's look at Chesapeake Bay you can decide which water body you want to look at I'm just choosing Chesapeake Bay you can go here take this rectangle or any other polygon and you can draw a box and that selects the data for the dates that you have selected. So here now we have PACE, OCI, AOP data, and we have two granules now.

51:46 Speaker 1: When you click on that, you will see which granules you have. And by clicking this on these arrows, you can download these data or you can say download all, then all the files will be downloaded. So I have already done that. So now what we are going to do is go to CDAS and examine these files. So you can open CDAS, which has to be installed on your computer. And once you open CDAS, it opens this window with File Manager. You can go to File, Open Product, and it takes you to your computer desktop And you can choose where you stored the files.

52:35 Speaker 1: Here, this is for Lake Victoria. I've already loaded it, level 3, that we downloaded from OBDAC. I've also uploaded this level 2 AOP file here. So let's start by looking at level 3 data. I'm going to click on this bands, and here is chlorophyll A. We chose that, and we downloaded it. So when you click on that, you can see the chlorophyll concentration map. You can create land mask. I'm going to say create mask by going up here. And then masks appear here. So you can turn the land mask on. So you have this.

53:21 Speaker 1: And you can also see where it is geographically. When you click on this world globe, it shows the map, background map. So here is where you can see this is Lake Victoria. You can change colors if you prefer. Let's just change the range. And here you can see the higher concentration of chlorophyll is in coastal region. here you can select a polygon from the top menu bar just going to take a little square here as an example and you can calculate statistics for that go to analysis and calculate statistics and you can pick geometry and run and and it gives you histogram of chlorophyll A for that box that you chose as the percentile values.

54:34 Speaker 1: You have how many valid pixels are there, mean standard deviation, variance, minimum and maximum values, et cetera, are given here. So again, I recommend that you look at our CDAS training to look at more analysis examples and options. So with that, let's look at

55:00 Speaker 1: level two data now. So next, let us look at this AOP data for remote sensing reflectance. If you click on the file that was downloaded and go to bands, you would see RRS. These are all the band reflectances. And these are the uncertainties. AOT is there. Here we are going to look at RRS. If you go to RRS and just pick any band, you would get a map. Here you can see that Chesapeake Bay is down here. I have used this create a subset tool and have taken focus on just this area. So I'm going to just go to that.

55:56 Speaker 1: So we have a focused area to look at and that is the subsetted area and you can go to bands and here let's turn this off and look at RRS here. I'm going to go down to green region, green wavelengths, and let's see. Okay, here's the focus on Chesapeake Bay. In addition, I have selected a few points. I've placed these pins here, one, two, three, and four, you can see. So you can get analysis done at those specific points as well. As you see, higher reflectance and a little bit of lower reflectance and so forth.

56:45 Speaker 1: Here also you have level two masks, you can mask land here. So next, let us examine RRS spectra at these locations. And for that, let's pick RRS, go to optical and go to spectrum view. Here you can select this, it's a show spectra for all pins. Now, I have already done this before, so the spectra are already calculated. If you are doing this for the first time, it takes a few minutes depending on the file size. So here it looks like we can use a better interval on y-axis. So let's see, from 0 to 0.015 should do it.

57:38 Speaker 1: So here now, if you just go to this plot, it tells you this is for pin 4. It's for which wavelength. So this is 588 nanometer. And it shows the value at this particular wavelength. And this is for pin 3, pin 1, and pin 2. And you can look at where they are and see how spectra vary at different locations. So this is just a simple visualization of the data we have, level two data. And you can go to raster and do math band and see edit expression here. You can use this to do either band ratios, or if you want to calculate spectral indices, you can utilize this tool.

58:43 Speaker 1: So these files, level two and level three files are available to you if you want to explore this in CDAS. This is not mandatory, but if you are interested, you can download the files and work with the data. So with that, we're going to conclude this demonstration. This brings us to the end of today's session, and so let us summarize what we saw. saw today. First, Dr. Morgan McKibben, she talked about PACE's Applications Program, which focuses on decision-making activities in water resources, fisheries, ecosystem areas.

59:23 Speaker 1: She also showed examples of PACE early adopters activities such as aquaculture site selection, enhanced cholera risk models, and also we saw a tool, hypercoast water quality monitoring for lakes and estuaries. Then we had description and access to multiple levels of PACE data, and we also saw how to get data from OBDAC level 3 and level 2 data from Earth data. We saw that in our demonstration. Morgan also showed snapshots from NASA Worldview, which provides near real-time base OCI true color images and chlorophyll A concentration data.

1:00:09 Speaker 1: You can download that data as well. And then we had a short demonstration of how to look at level 3 and level 2 data in CDAS which we downloaded from OBDAG and also from EarthData. We looked at spectra in Chesapeake Bay, and we looked at chlorophyll A concentrations in Lake Victoria in Africa.

1:00:34 Speaker 1: Next week will be our part three. In that, we will focus on access and visualization of OCI remote sensing data, both level two and level three, using open source Python software and Jupyter Notebooks. and also will learn how to customize these codes or notebooks to area of your interest.

1:01:01 Speaker 1: There will be one homework posted next week on 9th October on our website and will be due by 24th of October. And a certificate of completion will be awarded to those who attend all three live webinars and complete the homework assignment by the due date. So we want to thank Dr. Morgan McEwan once again for her presentation and information that she provided. Here's the contact information if you have additional questions. You can always contact.

1:01:36 Speaker 2: Us at RSET.

1:01:37 Speaker 1: Here's our website and our social media presence is there. These are our sister capacity building programs if you are interested.

1:01:47 Speaker 1: And here are the resources that you can learn more about PACE data products and how to access them. And so thank you very much for attending today's session, and we will now move on to our question and answer session. Thank you. Thank you. Let's move to... question and answer session and we'll cover as many questions as we can. We have received several questions and we have Dr. McKibbin here and also Mr. Ian Carroll. They will be helping us with the questions. So I'm going to start with question one and you can unmute and answer the question Morgan.

1:02:40 Speaker 1: So question one Are there more resources we can access to learn more about base data like data access or processing?

1:02:52 Speaker 2: Yeah, hello. So thank you for that question. So this tutorial gave a great overview of where to find things and where to find resources. And then the next training is going to actually do some of the data access visualization and analysis using two different tutorial Jupyter notebooks. So we definitely suggest you check out part three of this RSAT course. And then we've also just launched a new help page. This did not make it into the slides. It wasn't open yet when I made these slides. So I definitely recommend you check out our Help Hub webpage.

1:03:31 Speaker 2: It's new. It's linked there in the question and answer document. And our goal there really is with that is grouping students all of the various resources for ocean color satellite data processing and access under one roof. And so a lot of our Jupiter notebooks are there. We're going to have other tutorials. This course will be listed there. So we very much recommend that you check that website out and also keep an eye on it because it will be, it's going to continue to grow over time as more resources come out.

1:04:07 Speaker 1: Great. Thanks, Morgan. Second question is, do you provide consumable data API for PACE?

1:04:14 Speaker 2: Okay. So, for this one, the answer is that all NASA Earth data is programmatically searchable. So, that means it has an application programming interface or API provided by the common metadata repository. There is a Python wrapper for the API called Earth Access and an R wrapper called Earth Data Login. You will learn in part three about using Earth Access. And in the future, we also expect to provide services like subsetting and downloading as different file formats, such as GeoTIFF or X-Array or through the Harmony API.

1:04:56 Speaker 2: Great.

1:04:57 Speaker 1: Question three is, can someone from outside the USA apply to join the PACE community of practice?

1:05:05 Speaker 2: The answer to that is yes. Sign up for our email listserv and you will stay up to date. I put the instructions to join are both in the slide content from today as well as I've added it here to the question and answer document so you can see how to sign up your email on our list. We send out through that list, we send out updates on events. There's updates on the observatory that come out. So just lots of different mission updates. It's a great way to stay up to date with our mission.

1:05:44 Speaker 1: Question four is, in what way can hyperspectral profile A data contribute to fisheries health?

1:05:54 Speaker 2: So this is a somewhat broad question. The answer really depends on which fishery are you looking at? How do you define health of that fishery? What region are you in? But, you know, broadly, the answer is we anticipate that OCI's hyperspectral capabilities will help with fisheries health assessments. For example, the phytoplankton community composition algorithms that will be coming They're going to be able to inform us what types of phytoplankton are there, which is essentially what types of food are there for the fish if they eat phytoplankton.

1:06:40 Speaker 2: They can inform us what types of phytoplankton are there, which can give you information on the health or status of the ecosystem and the fisheries. So, yeah. So, and then we can go on to question five.

1:06:57 Speaker 1: So what advancements in management practices can result from improved understanding of phytoplankton community composition?

1:07:08 Speaker 2: So this one is a very similar question to the other one in the sense that This also depends on which area you are in, which phytoplankton types you're looking for. But again, with the hyperspectral information from OCI, there's going to be new data products that were not previously available, like phytoplankton community composition, also phytoplankton size distribution. So these types of products allow us to separate the total phytoplankton community into its constituent parts, which is something that wasn't we could not do before.

1:07:43 Speaker 2: And so with that kind of information, you can better understand the food web that's there for instance, if you're looking again at fisheries. We can also, there's gonna be some regions where harmful algal bloom monitoring is anticipated to improve because of PACE's hyperspectral information, meaning we can better identify or track harmful algal blooms in certain areas.

1:08:14 Speaker 1: Thanks, Morgan. If I may add, in our first session, we had a little preview of how different types of phytoplankton, especially harmful ones, toxic ones, they affect human health and different diseases, both in saltwater and freshwater. When you talk about management practices, it's also for health you can use that information for your help. The next question is, is there a link for Hypercoast?

1:08:52 Speaker 2: We did put the links there for Hypercoast for more information. It is open source, and you can go read more about it using the links that we provide.

1:09:00 Speaker 1: Question seven is how can I successfully install and configure the CIDA software on my Windows system and what are the key challenges or requirements during the installation process? So yeah you can If you go through RSAT CDAS training that was presented earlier this year, I think it was in February, and the link is given in the presentation and we'll also provide it here. It talks about how you can install it on a Windows machine and what are some of the challenges that you might face. And here you will also get a website where you can download specific version for windows and then how to proceed from that.

1:09:56 Speaker 1: And the question eight is can OCI separate and quantify non-alcoholic particles in phytoplankton contamination? If we use polarization data from HAARP 2 and SPECS 1, can those be more accurate?

1:10:12 Speaker 2: Yeah, so for this question, the products that I could think of that could be the most relevant are some inherent optical property-related data products that are planned. They aren't available yet, but they will be able to separate out spectral non-algal particle absorption from spectral phytoplankton absorption. There's, of course, going to be or is data products that can also look at particulates overall or sediment overall, but that's not really getting to separating it from phytoplankton. As for your second question about using polarization, that sounds like I'm actually not sure if that's possible.

1:10:58 Speaker 2: Polarization is outside of my wheelhouse, but perhaps that's an interesting research question. to look into.

1:11:08 Speaker 2: And we can go to the next one.

1:11:10 Speaker 1: Yes. How do you differentiate reflectance from algae slash plankton and reflectance from sediments in shallow environments?

1:11:21 Speaker 2: So for this one, it really depends on the concentration of of sediment, how much sediment is there, what type of sediment is there in the water, and how that compares to the concentration of phytoplankton that are there and what type of phytoplankton are there. The reason that's important is because each one creates a different kind of signal that is detected by the satellite, and their signals have different shapes. But if there's a lot of sediment, In the water, it makes it harder to differentiate them.

1:11:57 Speaker 2: If

1:12:00 Speaker 2: you have a lot of phytoplankton in lower sediment, it will be the data products, it'll be easier for those to correctly detect whether sediment or phytoplankton are being seen. This can also be an issue in shallow environments if you can see through the water to the sediment on the bottom because bottom reflectance basically does the same thing as well. Any kind of reflectance from sediments is very bright and it can overwhelm a phytoplankton signal. So the takeaway is it's a bit more difficult to differentiate them with the data products the more sediment there is.

1:12:39 Speaker 1: Question 10 is can you expand upon what steps will be taken to validate the PACE standard data?

1:12:46 Speaker 2: Yes, so we have a funded group of scientists called the PACE Validation Science Team, and they have, they are currently, and over the next couple of years, collecting in-situ data from around the world to help validate and calibrate PACE data. If you, there's a link that I've put in the chat where you can read more about the different teams and what they're doing, and We also, the PACE mission also just completed a month-long extensive calibration and validation effort in Southern California. It was called PACE PACS.

1:13:27 Speaker 2: It was extremely, it was an amazing effort. It included airplanes, two different kinds of airplanes, one of which was an extremely high-altitude airplane. It involved ground stations. It involved ships. It involved sinking all of those up with the satellite, up with PACE. to be able to calibrate and validate the information. And this was for both the ocean data as well as the aerosol data products. I put a link here also that tells you more about PACE Packs as well as other campaigns that are going on to validate the PACE data.

1:14:08 Speaker 1: Great. Question 11 is, could you elaborate on the application of PACE data products for applications of waterborne disease like cholera?

1:14:19 Speaker 2: So, this is related to Dr. Antarjitla's example that we gave of his early adopter project. I personally don't know about the application of PACE data products for cholera or other waterborne disease, but please feel free to go check out his webpage, his early adopter webpage. and maybe look up some of his research to see what they do in that realm. We can go to the next one.

1:14:51 Speaker 1: How accurately do the base data detect a potential phishing zone?

1:14:57 Speaker 2: So this one also really, this question is also pretty broad, also depends really on what type of fish you're looking for and what region you're in. If, for example, the fish you're looking for are attracted to zones or areas with a lot of phytoplankton, high concentration of phytoplankton, PACE data can estimate where the largest concentrations of phytoplankton are in the water. And that would indicate a higher probability of that specific kind of fish being there. That's the kind of example of how pace data may help find fishing zones.

1:15:26 Speaker 2: It all just depends on what the fish are looking for and which data products of paces might relate to that. And next one.

1:15:36 Speaker 1: So next question is, will changing the format of the data affect the quality and projection of the data? For example, I usually convert them either into TIFF files or text arrays.

1:15:58 Speaker 2: So for this one, the answer is also it depends on how you do it. So converting one of the level three mapped products, which come as NetCDF4 file formats, converting that to a GeoTIFF without changing the projection from the plate carry setting it has, that can be done without affecting the data values. Any reprojection will change the data values, but any geospatial data processing software, including CDAS, can do this without reducing the data quality. That's great.

1:16:34 Speaker 1: Question 14 is, where can I find detailed information about the algorithm used in hypercost? to generate Chlorophyll-A products from PACE imagery?

1:16:48 Speaker 2: So Hypercoast is an open source software for visualizing and analyzing the products. Right now, the examples shown of Hypercoast are using PACE data products. So it would be the same algorithms being used right now for PACE Chlorophyll-A. It depends on which. That was the example that we saw. So they are developing a phytoplankton absorption algorithm that will be added at some point. I would check out, it's not there yet, I would check out their website, which we have linked in the question and answer, to see if they have documentation on how that is made.

1:17:37 Speaker 2: I believe they also just published a paper about Hypercoast. I will go look for that and also put it in the question and answer section in case it has some of this information as well.

1:17:50 Speaker 1: Question 15 is, what is the spatial resolution of the OCI level 1C data with the common grid? When projecting the level 1B to a common grid, do I lose spatial resolution since the resolution varies cross-track and changes with scanning?

1:18:11 Speaker 2: So we do have a user guide, which we put a link to in the chat that may be helpful. And it says that for PACE L1C data products, OCI, HARP2, and SPEX1 data, it explains how those are all combined for the L1C products. It also gives the spatial resolution as 5.2 kilometers by 5.2 kilometers. With the reduction in spatial resolution relative to OCI's one kilometer resolution, due to the coarser resolutions of the polarimeters. So OCI has higher spatial resolution. Yeah, so check out that user guide.

1:18:47 Speaker 2: Again, we put the link in the description. the question and answer, and you can read more about that. Next one.

1:18:56 Speaker 1: Next one is data suite is a great concept. If I want to do something similar for L1B to L2, what do you suggest? I see the CDAS command L1b extract OCI only subsets C-gen, not wavelengths. Is it difficult to subset wavelengths?

1:19:18 Speaker 2: So for this one, the answer is if you go from L1b to L2 using the L2-gen processor in CDAS, you can choose the specific product that you want to work with. And in that case, you can choose as many or as few wavelengths as you want to subset out. So yes, it is possible and not difficult. using CDAS. Go to the next one.

1:19:42 Speaker 1: Yes, so just to add that we did not talk about the OCS-SW, which is the processing software in CDAS, but again, I would recommend refer to the earlier ARSA training where we walked through steps of how to install different satellites and sensors in CDAS for L2GEN processing to get data from level 1 to 2. The next question, how do you interpolate the missing value due to cloud coverage? Are there any algorithms that NASA scientists usually use?

1:20:18 Speaker 2: So for the PACE OCI, data products that we serve, we do not interpolate missing values due to cloud coverage per se on an individual swath of data. One way that people do, well, the reason for that is because cloud data means data loss. So the clouds are basically blocking the satellite's view of the Earth's surface. So we don't get data. They get masked out. It's a no data pixel. One way around this that people use in areas, especially areas that are cloudy, is instead of using a data product for one day, you can use a temporally composited data product, such as an eight-day product, which averages all the pixels, all the information seen in all the pixels of your area of interest over eight days.

1:21:11 Speaker 2: So basically, If the satellite flew over once in that eight days, there would be data in that pixel for one day, but it would come out in the average. So you can do this kind of spatial averaging to fill in the gaps. if that works for your use. But otherwise, there's not interpolating per se or kind of creating or modeling data in between the cloud gaps. That's not provided with these data products. And the next.

1:21:47 Speaker 1: Question is, are there any options for image segmentation in the Earth Data Portal?

1:21:53 Speaker 2: So for this one, I'm wondering if you mean... similar to the ability maybe to enter an area of interest, you know, enter four coordinates to have a box that's your study area of interest and extract that out of Earth data and extract and download just that information. That is not possible at this time. I believe that We are looking into figuring out how to make that be possible on Earth data or through the OBDAC tools. So stay tuned on that one if you're looking into, yeah,

1:22:41 Speaker 2: basically cropping the data to your area of interest automatically before you download it.

1:22:47 Speaker 1: So I think ArchDataSearch does have a special filtering you can.

1:22:52 Speaker 2: Oh, does it?

1:22:53 Speaker 1: Yeah. Or you can pick up a point or a polygon or a square. That's what we did for Chesapeake Bay.

1:23:00 Speaker 2: Okay.

1:23:00 Speaker 1: Yeah, from ArchDataSearch. So that is there. You can do spatial and temporal subsetting. Yes.

1:23:07 Speaker 2: Okay. Yeah. Sorry about that. Yeah.

1:23:10 Speaker 1: Thank you.

1:23:11 Speaker 2: Yeah. Never mind what I said. Sorry about that.

1:23:13 Speaker 1: So question 19 is, when does the base data... trace back. How about data before 2024?

1:23:22 Speaker 2: So there isn't PACE data prior to 2024 because PACE launched in 2024. The data go back, I believe, to March.

1:23:32 Speaker 2: There have been several other ocean color missions, however, that have been collecting data for decades all the way back to 1997. Also freely available data. It's different. It's the multispectral data of previous sensors. In contrast, PACE is hyperspectral information. But there isn't a source of continuous hyperspectral information like PACE is prior to 2024.

1:23:57 Speaker 1: Next question is, could it be possible to have a GitHub tutorial on how to upload our code, maybe with a really simple example that is available permanently so that we can follow step by step? That would be very helpful for me, for example, that have never used GitHub but would be happy to share my code. It could also have some best practices so that we get used to doing it in the best way from the first time. Thank you very much in advance.

1:24:36 Speaker 2: Yeah, so for that question, I suspect there's probably a lot of tutorials online, basic GitHub tutorials that can teach you how to use GitHub, kind of the basics of getting your code out there and keeping it updated. And also our team is actively creating tutorials using Jupyter Notebooks. the ones that we keep putting online, the ones that will be presented in the next class. And I can add this to the suggestion box because we are absolutely taking suggestions on what users are interested in. And we can see if maybe it would be, if it's possible to integrate that into any future notebooks.

1:25:17 Speaker 2: But otherwise, I would say until then, see if you can find some tutorials online because we don't provide any currently.

1:25:29 Speaker 1: Next question is, with level 2 data, do we still need to do the atmospheric correction based on local atmospheric conditions to get more accurate values?

1:25:40 Speaker 2: So the answer to this one is level two data downloaded from Earth data is atmospherically corrected already. If someone processes L1b to L2 data using the L2Gen software, then the atmospheric correction can be turned off, but it is included by default in level two data.

1:26:07 Speaker 1: The next question is, to what level will the phytoplankton composition analysis B, a species or a genus or lower?

1:26:18 Speaker 2: So the answer to this is going to be it's going to depend on a lot of variables. This is a good question, and it's basically about taxonomic resolution. of information coming from PACE where, you know, genus or class, especially class, are kind of lower resolution. So an example of a phytoplankton class is diatoms, right? And there's a lot of genus and a lot of species of diatoms under that umbrella. PACE is going to be better generally at seeing more at the class level or by differentiating size classes of phytoplankton, then it will be species.

1:27:07 Speaker 2: So, species is the highest taxonomic resolution and that is going to be very difficult to tease out. That can actually be difficult to tease out even if you have the water sample in front of you and a microscope because phytoplankton species can be so similar to each other. So there may be unique scenarios where you maybe have a very large bloom of a phytoplankton species with a very distinctive signature, where PACE may be able to identify a species of phytoplankton, but in general, it will not be able to detect that high of taxonomic resolution.

1:27:46 Speaker 2: Think more along the lines of phytoplankton classes like diatoms, dinoflagellates, cyanobacteria, or along the size classes of phytoplankton cells.

1:28:01 Speaker 1: So there was a question. I'm an educator for the GLOBE program, and I'm taking AOT measurements. I'm very happy to learn how I can compare the AOT satellite data for a specific location to compare with ground-based data.

1:28:17 Speaker 2: So for this one, you know, the basic mechanics behind it are you pick the pixel that's the closest to your ground measurement in time and space, the closer the better, and you essentially kind of average the satellite data of that pixel and, you know, a certain number of pixels around it to get your satellite value to match up to the ground measurement. That would be kind of the just really basic high-level example of how to do that. We do have, from our recent PACE Hack Week course, we do have a tutorial.

1:28:49 Speaker 2: It's a Jupyter Notebook tutorial that gives a really great example, step-by-step, of doing this. Even if you're not a Python coder, you can still take a look at this notebook because it includes text that explains what each step is doing. And that text can also help you if you don't want to use or don't know how to use the code. That text will still help you understand what the process is that they go through. And you can see how that matches up with what you need to do.

1:29:21 Speaker 1: Next question is, what is the difference between types in OBDAP download parameters?

1:29:32 Speaker 2: This one isn't clear what types. I'm trying to think of what part of the presentation that might have been. I'm not sure what the person is meaning by types.

1:29:49 Speaker 1: So I think that it's provisional or test data or, you know, if.

1:29:53 Speaker 2: Oh, okay. So if you mean that data types, if it's data maturity or data status, that would be the ones that, like, provisional or test data.

1:30:11 Speaker 1: I think you mentioned all that. You described that you have a table in your presentation.

1:30:17 Speaker 2: Right, exactly. In the presentation, there's text that describes each of those in detail, each of those statuses. And then later in the talk, when I talk about where you can find all the data, it explains that. which of those levels of data are available in the different tools. So some of them, the test data, for example, are only available from OBDAC. It's less mature than the provisional data available on Earth Data Search. So if you take a look at the slides and get the definitions of the data maturity levels from the beginning, and then you can check out where those different levels are found in the later part of the talk when we talk about the tools for data access.

1:31:02 Speaker 1: Thank you, Morgan. We still have several questions to go through and we are almost at the end of time for this session. So what we're going to do is address all those questions and post them on our website in the question and answer session document and it will be available in about a week or so. So I think we will conclude Today's session, again, we thank Dr. Morgan McKibbin for her presentation as well as answering all the questions. And we also want to thank our SET team here for helping, our coordinators, Sabin Hudson-Odoi and Brock Levins, Natasha Johnson-Griffin, Sarah Paschall, Jonathan O'Brien.

1:31:48 Speaker 1: We want to thank them all and Sue Monty. for coordinating and helping in various ways in editing, putting things properly so we can all share with you. So thanks again, ARSA team, and thank you, Morgan, again. Thank you all for attending today's session. And we look forward to see you on 9th of October. That's next Wednesday. That will be our session on showing how Jupyter Notebook, Python-based Jupyter Notebooks can be used to access, analyze, and visualize data. So thanks again for attending today's session, and we hope to see you next week.

1:32:29 Speaker 1: Thanks, Morgan.

1:32:32 Speaker 2: Yeah, thank you for having me, and also thank you to the folks from the PACE team, Karina Poland, Antonio Menino, and Ian Carroll that helped me with questions. And thanks, everybody, for coming.

1:32:44 Speaker 1: Yeah, thanks, PACE team.