Data Access and Visualization

0:01 Speaker 1: Welcome back, everyone, to our RSET training series, Introduction to NASA Earth Observations and Tools for Wildfire Monitoring and Management. Today is the final part of our training series, Data Access and Visualization. My name is Melanie Flitcook. I am a researcher at NASA Goddard Space Flight Center right outside Washington, D.C., and project scientist of the RSET program. In Parts 1 and 2, we introduced the fire information for resource management system, or FIRMS, application, learned how fires are detected by satellite sensors, the strengths and limitations of different active fire detection products, and how FIRMS can be used for wildfire monitoring and evaluation, as well as issues that can impact a satellite sensor's ability to detect fire activity.

0:48 Speaker 1: Today, in Part 3, we'll focus on data access and visualization through FIRMS. And a reminder that there will be one homework assignment after this training. It will be available today after part three is complete and due in two weeks on May 14th. A certificate of completion will be awarded to those who attend all live sessions and complete the homework assignment before the given due date. And now we'll begin part three of our training series, Data Access and Visualization. And I'm very happy to introduce our guest instructors, from the firm's team for part three.

1:25 Speaker 1: Atmar Olshina is a principal software engineer, Dylan Mendez is a senior applications developer, and Asen Radov is a software engineer. By the end of today in part three, we hope that you'll be able to access fire detection data through file download, web map feature, web map service, and application programming interface, or API. subscribe to receive automatic email alerts when active fire information is available for a region of your interest, download historical fire detection data, and visualize fire detection data in Google Earth, ArcGIS, and QGIS.

2:08 Speaker 1: And once again, if you have questions during today's training, please put them in the Q & A box within WebEx at any time. At the end of the training, we will try and address as many questions as we can and we'll post them and the written answers to the training page within about a week after the training. And now, I am happy to hand it over to Amar.

2:28 Speaker 2: All right, well, thank you, Melanie. So now let's look at accessing active fire information on the firm site. So in this part, we'll be looking at how to download a simple, say, CSV file, KML, or shapefiles, which can be used in GIS applications or easy to integrate into Excel spreadsheets or download scripts. Then after me, Dylan will go over a web map feature and web map service and also demo how you can integrate mapping services into GIS applications such as ArcGIS and QGIS. And at last, Ascent will go over application programming interface, which is very helpful for developers if they want to integrate their data into their Ingest script.

3:21 Speaker 2: Also in this part, we'll look at downloading archive historical fire hotspot detections. So that's for those who are interested in looking at historical records of fire detections and do some of the additional research on that. We also look at notification through email alerts. So if you are interested in finding out fire detections in area of your interest, that's something that you might be interested in, as well as getting a summary at the end of day or week, how many alerts were generated in your area.

3:54 Speaker 2: And so this part three will be heavily dependent on demos. We'll show you Google Earth, RGIs, and QGIs, and also point you to additional tutorials and resources that FIRM's website provides. So now let's go to.

4:12 Speaker 3: Browser.

4:16 Speaker 2: So you might have seen this front page already in previous parts, but I'll go over it again. So the most important links you can find on our front page, and we'll be looking today at the active FHIR data for global and U.S.

4:30 Speaker 3: Canada.

4:31 Speaker 2: Here's a link to the FHIR alerts, archive download, and web services. which is essentially a summary to all of our services, how you can access our data. So when you get to our web services page, you'll be able to see all of the data access points that our site provides. So from the API to the active fire detections, mapping services, and tutorial and examples. I'm going to first mention these tutorials and examples at the bottom, just so we don't forget about it. Where you can see all of the tutorials, some of them that we provide, And some of this material is covered in this lecture, but there's also links to additional tutorials and examples that you might find helpful when you're doing your research or work.

5:19 Speaker 2: So it's worth to mention that, so you remember we have that. So now we'll focus on looking at the CSV, KML, and shapefiles. This link will provide you with hourly generated data files that provide the most recent active fire detections. Now, for those who are interested in real-time and ultra real-time data, this is not the appropriate service because it's update hourly. So for those, we recommend that you use different services such as our API, KML Fire Footprint service, or our mapping services.

5:59 Speaker 2: So the information is mentioned here for you. and hopefully that will help you to get that data into your system if you need it. So these data sets which are generated hourly are good for those users who don't necessarily need up-to-date information for fires. For added convenience, we have broken down our data files into major geographic regions, and here is the map highlighting them. So this way you don't have to download the entire world to get the data. And as you can see, they're broken down here, and we provide them in three major formats, shapefiles, which are used by the GIS applications, such as RGIS and QGIS, and Google Earth, keyhole markup language, which is generally used by Google Earth, and then text file CSVs, which is the comma-separated files.

6:53 Speaker 2: For those who have never seen one, it looks something like this.

6:56 Speaker 4: Okay.

7:01 Speaker 2: So it's a lot of data files and commas operating. Additionally, we have broken down our data files based on satellites, except for MODIS where Aqua and Terra are included in the same file. Then we provide a temporal breakdown based on 24 hours, 48 hours, and seven days. These, except for Google Earth files, where we only provide 48 hours. Now for Landsat, we only provide Canada and US, actually continental US files. Now, for US Canada version of this map, I mean of our site, we also provide additional data files for Google Earth KML.

7:53 Speaker 2: which are active fire detection footprint and centroids, KMLs, KMZs. And they are generated for US and Canadian provinces. And they provide essentially the area of the detection instead of just the point data. And I will be showing you shortly how it looks in Google Earth. And for Google Earth KMZ bundles, we also have links that provide easy to ingest process into Google Earth. And if I want to do something like that, I can, for example, click on this Conus Latest Firms US Canada, download it, and then use it in Google Earth once I open it.

8:39 Speaker 2: Similarly, I'm going to download, for example, a KML file for South America for MODIS. And I'm going to use it right now in Google Earth. So I have already opened Google Earth. So if I want to ingest data into Google Earth, I click on new, click on import KML file to map project and upload from device. And then I use, for example, we downloaded the South American file. So I click on that and I click open. Now the Google Earth will load the data and we'll see list of the fire detections over South America.

9:30 Speaker 2: And I can zoom in and click on the individual fire detection and get the extra information that's associated with it. Now, similarly, I can go and open the Google Earth KML file and For KML, KMZ files, the bundles, we can again click on new project, import KML file to map project and upload from device. And at this point, I will check the fire point area for continental US and Hawaii. And once that loads, well, I'll receive this type of information. And once I start zooming in, you can start seeing that there are fire detections and their area shown on the map.

10:21 Speaker 2: So that does it for the Google Earth. And now let's look at the archive download. So for downloading our archive data, which you can again access from our front page through archive download or our hamburger menu on the top left corner under download archive data, which will bring you to our archive download application. Now, for this, we recommend that users generate EarthData login, so this way they get access to other EarthData NASA resources. But for, I guess, legacy users, we also allow users just to use their email address, and there will be emailed a six-digit code they need to use for authentication.

11:11 Speaker 2: Now, once you log in, you'll get the application where you can create new requests. So I click on that. And in our download request, I have the option to download the entire world. Select the country of my interest. And for smaller and mid-sized countries, we also allow buffer option. So you can set between 0 to 15 kilometers. So this way you can find fires that are just outside of the country's border, if you're interested in that. Another option is to select custom region, which will provide you with the option to use the map, where I can either zoom in to the area I'm interested in, or I can highlight the area, or draw a custom polygon if I'm, for example, just interested in a much smaller section.

12:11 Speaker 2: And if I save it, it will generate this type of polygon string. So for some of the users that have more complex polygons and they don't wanna manually draw them, they can paste and copy it here as long as it follows the same format that we are showing you here. And in addition, we also provide protected area. And these protected areas are provided by the World Database on Protected Areas. So in this case, I don't know, I can pick, for example, Brazil. And as you can see, there are over 2,000 protected areas in Brazil.

12:52 Speaker 2: So for added convenience, you can just type in the name of national park if you know it, and you can select it. And optionally, again, you can select buffer so you can get information about fires outside of the protected area, just within 15 kilometers. So then next I'll select what fire source I want. So, for example, MODIS versus NPP. And then date range. Now, this is important to notice because the satellites, older missions, for example, MODIS start from 2000, but some of the newer ones, like VIRS NOAA 21, only from 2024.

13:35 Speaker 2: So, it's important to notice that when you want to get in for historical data, so you are aware of how far back you can go to get the relevant information. So now select date range, and then I have the options to generate it as a shape file, comma separated text file, or JSON file. And I can get an email confirmation that I submitted this request. Once your download is processed, which may take anywhere between five minutes to several hours, and sometimes even longer if we have a lot of users submitting download requests, you will get an email notification that your download has finished.

14:24 Speaker 2: So you get a little summary that reminds you what exactly you were downloading with the link for your download. Now, one thing I'd like to know about downloads is that We provide generally standard data if it's available, but not for all of our satellites. So we sometimes have just new real-time data. But since standard data tends to be three months or more behind, then if you request date range from, I don't know, six months back to present, you might receive two files. So once you unzip it, you might receive two files So once you unzip it, you might receive two files that one covers the standard data and the other one covers near real time data.

15:15 Speaker 2: So that's just something to notice so you're not confused why there are two files. But generally, if the entire date range fits within either near real time or standard data, you will just receive one data file. So now let's go and look back at the fire alerts. And fire alerts, again, you can create similarly like the download request. You have the option to put your alert name to remind you what type of alert it is. And just like we had in download request, you can do it by world, country, custom region, or protected area.

16:04 Speaker 2: So in this case, let's just quickly pick one country. And I can pick a fire source. So I can do all four of them if I'm interested, or I can just do a specific satellite if I only need that one. The next important thing is the alert frequency. So we offer daily, weekly, rapid, near real time. So those who are interested in up-to-date information about fire detections in the area of interest, you want to select near real time.

16:37 Speaker 5: So generally, you.

16:38 Speaker 2: Will be notified between one to three hours after fire detection in the area of your interest. It takes about an hour to three hours to process the data, and then an additional maybe five to 10 minutes before the email alert is sent out to you. Then we also have the daily and weekly, which is beneficial for managers We want to keep track of how many fires occurred in their area of interest the previous day or previous week. So this is sort of like a little summary. You can select the email preference, whether you want to receive text and map image or just text only.

17:15 Speaker 2: If you select the map image, you also have the option to select size, how big the image should be. And then we offer English, Spanish, and French version of the email alert. Optionally, you can also decide if you want the CSV file attachment, which will send you the list of the fire detections if you wish to receive that. And once everything is filled, then you click submit and you'll be able to receive email alerts. And now these email alerts, the email alerts will look something like this. So you will get the notifications that there were 36 fire hotspot detections in your area of interest.

18:00 Speaker 2: And you can go click on the map image on the server and access the website from here, just so you can quickly get the information. Now with that, I'm going to pass this to Dylan. We'll go over map services and demo you how to use ArcGIS and QGIS with firms data.

18:28 Speaker 3: Thanks, Atmar, for the handoff. Now I'll be taking you all through firms and separate GIS applications, starting with QGIS on the right, the logo on the right, and then ArcGIS with the logo on the left. I'm going to quickly go over where I'm going to take you all through. We're going to go through the definitions of web mapping service and web feature service, why you would pick a web service over downloaded data, which web service you might want to pick, Then I'll be demonstrating how to use QGIS and ArcGIS Online to ingest downloaded data into the application, how you would generate a MAC key on firms for the web services, and then how you would use the web services in the GIS applications listed QGIS and ArcGIS Online, which are both free.

19:17 Speaker 3: Now I'll show you how to use Atmar's downloaded shapefile in the GIS applications We'll start with ArcGIS Online. Your ArcGIS will look a little bit different. This is just Mouse's Enterprise version, but you'll have the same top bar at the top with all of the gallery and the content and everything. And you'll want to head over to the Content section, and the content will load any content that you've already made. So if you have a shapefile, here mine is on my desktop. up in the top right. And all I have to do for this is just drag and drop it onto here.

19:55 Speaker 3: And it will pop up this window. It can recognize that it's a shapefile. And you'll want to create a hosted feature layer so that the layer can show up on the map.

20:06 Speaker 2: Click Next.

20:08 Speaker 3: Just name it something different. Downloaded prior data. And click Save. And it will make the map and put it into your content. Now that you have your layer on the map, you can open this in the Map Viewer and then view your data in ArcGIS Online on the map. And you can't really tell where this is at first, We're looking at somewhere in the rainforest of Brazil for this data.

20:50 Speaker 3: Now I'm going to show you on QGIS how to add this data in. Also similarly, it's similarly easy to load this data into QGIS. All you do is drag the shapefile that you want onto the layer panel over here. And for a little bit of context, I just have a background layer because if that's not there, then you won't be able to see any of the geographic regions. You'll just see some dots on a white canvas. But I'll take the shapefile, I'll just drag it over here into the layer list, and my data pops up.

21:22 Speaker 3: Now I can zoom in to Brazil over here. It's in the same region. Take a second to load. You can see the shapefile data populate into And that's it for QGIS and ArcGIS for downloaded data. Now, you can use this QR code to take me directly to the firm's MapKey site, but I'll show you a bit of the snippets and how to generate a MapKey for using our API services and our web services for creating a MapKey. This is what the MapKey creation website will look like. And you'll want to click Get a MapKey right here, which will help you generate your MapKey.

22:15 Speaker 3: It will pop up a window just like this, and then build the screen. Behind the box will be shaded out to the rest of the firm's website. Just enter your email in the box, and it will send you an email with your generated MapKey. And there's information about the limits there that you can find on our website. On this email from noreply.nasa.gov, which will look exactly like this, the map key is at the top of the email in big, old black font. This is the text you need for web mapping and feature services, as well as the API.

22:51 Speaker 3: Note the notices right under the map key with the limits on its usage as well. For those that are links to the tutorials and other similar ones, I'm about to get on web mapping and web feature services. Now for the web mapping and web feature services, WMS and WFS respectively. A bit of information first. Since we already looked at downloaded data in shape file format, if you have an isolated event in time, downloading the data will preserve the timeline of its occurrence. If you do not have access to the internet, this would be a good option to download your data that say you're working out in the field.

23:36 Speaker 3: or something where you don't have access to the internet. An example would be wanting to see the difference between also real time versus near real time versus our standard quality data. You would want WFS and WMS often for ongoing usage of our GIS data. WFS and WMS would be preferred. They provide real time access to current data without the need for local storage. So you're just connecting to the internet to view our data. It ensures that you're working with the most up-to-date information, reducing any discrepancies from outdated datasets, eliminating the need for downloading large datasets, and saving storage space.

24:18 Speaker 3: Standardized protocols facilitate seamless integration across various different GIS applications, so you can use this in a variety of platforms. WMS or Web Mapping Service are map images over the internet to visualize spatial data as rendered images. The data is displayed as images in the format of PNG, JPEG, etc. Primarily, it's a visual representation supporting basic panning and zooming, but no manipulation on the data. Web feature services, or WFS, are access to manipulatable vector-based geospatial data over the internet, retrieving the actual feature geometries and attributes.

25:09 Speaker 3: This displays the raw vector, data, point, line, polygon, et cetera, onto the map, and you're able to perform spatial analysis and data editing in the GIS applications and the dual web feature service. This supports operations like querying, updating, inserting, or deleting features of your own type. So the difference is, or the reason you would pick one over the other, but mapping services are ideal for applications that require map visualization without the need for data manipulation. Suitable for when the focus is on rendering maps for the display purposes primarily.

25:51 Speaker 3: And then a web feature service is preferred when an application requires analysis, editing, or complex querying. It enables you to do things such as spatial joins and real-time data integration into other systems. And I think this image that was on the previous slide and the slide shows well that the computer intakes the data from different sources and can deliver it as images or text. features such as polygons and lines and points, which are vector-based, shown at the bottom for web feature services and images at the top for web mapping services.

26:32 Speaker 3: You can scan the QR code to see any additional information like on WFS and WMS from our firm's informational page. I can show that briefly. But now I'll show you how to connect WMS and WFS in ArcGIS Online and QGIS. To start, we'll go back.

27:11 Speaker 3: This is our informational page on WFS and WMS. There is a lot of information that describes all about our services and what you can do to manipulate them and such.

27:25 Speaker 3: But first I will show you how to import web mapping services into ArcGIS. You'll still want to start on the homepage of ArcGIS online. You'll want to go to the content You'll want to create a new item here at the top left. Since we have a URL that we copied from the web mapping service on our firm's website, we'll want.

28:12 Speaker 1: To create a URL.

28:14 Speaker 3: I already have these URLs. You'll be provided, this is what the generic URL with the your map key shows. And then I've replaced it with the map key that we already have. So I'll copy this link here. And I'll paste this link into ArcGIS Online. And it can already recognize that it's a web mapping service. And it will be able to pull in the correct parameters that it needs. I will want to look at all of the web mapping. I actually don't need to look them all. We'll do modus seven days firehide. The title, I don't need to create a new one this time, and we'll click Save like we did before.

29:03 Speaker 3: It loads my web mapping service into the same format that it did before. I'll just want to open it in the Map Viewer. And now you can see the fire hotspots as images all over displayed on the map. This is our web mapping service in ArcGIS. If I'm already on ArcGIS Online, I'll quickly go through how to do the web feature service. It's pretty much very similar to what you would do for a web mapping service. You'll go over here to Content,

29:44 Speaker 3: and you'll want to create another new item. And since we are from the Web Feature Service website, we can copy one of these, and there is a link. Those links are here. Your map key is also in the provided link, and you'll want to replace that with your actual map key, which is replaced in this bottom link. And you can copy this out, click URL, paste your URL in the box, and it can already recognize that it's a web feature service. Pick which layers that you want. We'll do the notice seven-day fire hotspots.

30:27 Speaker 3: It already comes up with the title for the layer, so we don't need to edit anything here, and we can click Save.

30:40 Speaker 3: That loads our web feature service into our GIS item. You can open it in the Map Viewer again. That will also load our web feature service onto the map, and now you can click on any one of these dots and get information about the fire hotspot itself, something that's not available on the web mapping service. Now I'll show you how to do the same, but in QGIS. So once you're on ArcGIS, you want to come over here to layer at the top, and you want to add the layer. And then there's add WMS WMTS layer.

31:27 Speaker 3: And that's what we're going to do first. So we'll hit add WMS WMTS layer. We'll populate a window. You'll want to create a new service, which brings up this window. And then we can copy our WMS link with our map key in it. Place in the URL box. And because I already have the title of this layer, I'll just copy it here, name it there. There's no authentication needed because we have the map key in the URL itself. All I'll have to do is click OK, and then connect. You'll want to select one of these with the name that you recognize.

32:16 Speaker 3: So we'll do fire modus seven days right here. title, BIOS Modus 7 Days Fire Hotspot. Click add, which will add it to the layer or add it to the layer list. But this is the fire hotspots and you can zoom in and see exactly where those hotspots are. And you'll see whether when it's floating right here at the bottom. Now we'll try and add the WFS layer into QGIS. We're going to want to gather the WFS link that is replaced with the API key. Conveniently, I did it here. I'm going to copy that link. You're going to go back to the QGIS window.

32:59 Speaker 3: You're going to come up to layer. In add layer, you're going to want to go all the way down to add WFS OGC API features layer. Click on that. You'll want to make a new connection.

33:12 Speaker 1: We're going to.

33:13 Speaker 3: Call this firms because this is our firms link. Once you've pasted your link in there, you're going to click OK, and then Connect. From there, you're going to want to click Know It 27 Days or whatever layer you'd like.

33:37 Speaker 1: Click Add.

33:38 Speaker 3: That will add it to the map.

33:43 Speaker 4: And click Close.

33:45 Speaker 3: How the data is loading, see it down here with the box, three and a half seconds remaining. And you can see even where the United States outline is from all of the FHIR data, even though there's no map behind it. And that is exactly how you connect a WFS layer to QGIS. And as you can see, the more data that's populating, Thank you, Dylan.

34:24 Speaker 5: Hello, everyone. My name is Osen Radov, and I'm a software engineer at FIRMS. Today, I will do a high-level overview of FIRMS APIs and how to use them. Here's a quick summary of what we're going to cover. We'll go over how to generate a map key again, because this is crucial for using any of our APIs. I will go over the data availability API. Then I will also cover on the missing data table that we have. Then I will go into more details about the KML Fire Footprints API. Then I will also cover the country and the area APIs.

35:01 Speaker 5: And I will finish up by mentioning about our code examples and tutorials that we provide. So let's dive right into it. The APIs are under, so this is the main page of our website. The APIs are under the web services. Today we're going to review the global APIs. The ones for USA Canada are very similar. So here the first option is where all of our APIs are located. First, I wanted to cover the map key because that's very important for using any of our APIs. So the map key, the link to the map key is right here.

35:49 Speaker 5: This is where you will come to generate a map key. You will click on this Get Map Key button, and all you need for generating a map key is an email address. So once you enter your email address over here and click get a map key, you will get an email from us that looks something like this. The email from us will look something like this. The map key will be at the top in bold. So there is a transaction limit that every map key has, which is 5,000 transactions per 10 minutes. If you require more transactions, we can increase that limit.

36:35 Speaker 5: All you have to do is send us an email at support at per data.nasa.gov and we will increase your transaction limit if you think that the one you initially got is not enough. Also in this email, you can see there's some important links through some of the resources that we have. So once you have your map key, you can start using all of our APIs. The first API I wanted to cover is the data availability API. This API is very simple. It only requires a source and your map key. So the source can be either the default one, which is all sources, which is selected right now, or you can select a particular source if you're interested in just a particular satellite.

37:35 Speaker 5: I will do all available sensors and I will enter my map key. And when you click the display results, you can see the results. What this API provides is a date range where we have fire detections for the specific satellite. As you can see here, we have, for some of the satellites, we have standard processing data and near real-time processing data. The difference between those is that the standard processing data is usually three months behind, but that is the best quality data that we have. So it is recommended to use that one when available.

38:22 Speaker 5: It takes a longer process to produce that data. That's why it's three months behind. The near real-time data is produced usually within a few hours of a satellite detection, and that's how we can provide data that is really recent. And over here, with this API, you can see the date ranges for all the satellites and the difference between when For example, let's look at MODIS. The standard data for MODIS, it's only available until the end of last year, 2024. And then from January 2025 until now, we have the near real-time data available for this satellite.

39:10 Speaker 5: Similarly, for NOAA20, we have standard data available until the end of January of this year, 2025. And then from today, going back until February 1st, we have near real-time data. So this is really helpful. For example, if you want to download historical data for, let's say, a month or the year 2020, you can come here and use this API and you can see that there's only three available satellites that will have data from 2020. That will be the MODIS standard product data. That will be also the NOAA 20 standard product data and the SMPP standard data.

39:58 Speaker 5: All of the other satellites have data that is either later than that specific year. So this is very useful for figuring out when data is available and what satellite you want the data from. Next, I wanna review the missing data table that we have. This table has two main columns. The first one is the standard data column and the other column is the near real-time data. And then that's also subtracted into the date and the satellite. So over here in this table, you can see a date and the sensor where a data is missing.

40:51 Speaker 5: And that can happen for a variety of reasons. Sometimes satellites need to do some maneuvers or they need to be recalibrated. So when that happens, there's usually a day or a few days when we don't get any data. If you, for some reason, downloaded data and you're interested in a specific date and that date has no data, you can come over here and double check that we actually have data for that date. Because for one reason or another, we may not have that data. So for example, right here, you can see for the Suomi VIIRS MPP sensor from July 10 of 2024 until July 15 of 2024, there is no data.

41:49 Speaker 5: So next, I want to cover the KML FireFootprint API.

42:00 Speaker 5: This API will produce results in a KMZ file format. And that file format is suitable for display in Google Earth as you have previously seen. This API is subdivided into 12 different regions. So you can select between Canada and Europe and Australia and New Zealand, for example. Then the next subsection is by the the date or the time span. So there are four different options that we provide, either the last 24 hours, the last 48 hours, the last 72 hours, or the last seven days. And finally, the five different sensors that we have.

42:57 Speaker 5: So all this is put into this nice table for you. As you can see, the Landsat product only has data for the Canada and U.S., Also, you can see here in the titles the sensor's resolution. So the resolution, so you can see MODIS has a one kilometer resolution, the VIIRS has 375 kilometer, 275 meter resolution, and LATSAT is 300 meter resolution. So if you're interested in the last seven days fires over Europe, for example, click on that, and that will download that file for you, which then you can go ahead and use however you see fit.

43:52 Speaker 5: The next API I want to cover is the country API. This is one of our more, yeah, so this is one of our APIs that gives you more information. This API will return the fire points into a CSV format, which is like a text file, which is comma separated. This requires you to select a country. So we have a dropdown of countries that you can select. So for example, if I select Thailand, then it requires you to select one of the satellites will go for example with modus near real time data then it the date is optional you can you can leave it blank and it will default to today's date or you can enter a date in in the in this format it has to be four years of four digit year two digit month and two digit date so let's For example, let's do 2025 February 10th.

45:13 Speaker 5: And there's also a date range. So if you select one, it will be just the date that you selected. You can go up to 10 days prior to the date that you have selected. And the last thing you need for this API is the map key. So I'll put my... mat key and when you do display results this is kind of the the format that that you will see

45:50 Speaker 5: all the all the the first uh row is kind of the the title or the what different what each different value is and then you will get all the different points in the in in in this format And just one more thing to cover here. So all this information that I just described is also described here in the bottom. What is the different sensors format that this API accepts? The three letter country code that is needed, the date range and the day range and the date range. So for the country code, if you We also have a countries table, which you can access from here.

46:45 Speaker 5: So when you are using this API, and let's say, for example, you need to find the country code for Denmark, you can come here and see the three-letter country code that this API will accept. And finally, the last API that I want to cover is the area API. So this one is a little more powerful than the country one. In here, it is very similar to the country one, but with one exception that you can do custom area or the world, the whole world in this API. And over here, you can see the world. So by default, if you don't provide anything, it will default to the to giving you results for the whole world, which has these coordinates.

47:43 Speaker 5: So this uses latitude and longitude coordinates, which are subdivided into this format. It has to be the west coordinate, followed by the south, followed by the east, and followed by the north. So if you want to do a custom area, it has, so here is an example with South America. These are kind of the coordinates that South America has. But if you want to do your own custom area, I will show you how, for example, so you can use, for example, the Fire Alert subscription service that we have and that we reviewed earlier.

48:26 Speaker 5: So if you go in the custom region option and then you click use map, you get to this feature that we provide. You can draw a custom box over the area that you like. And let's say you are interested in the Great Lakes region, for example, and you want to do this area. When you click Save, it will give you the area in the format that is needed for the area API. So you can just copy that from here, and you can just use it in the API.

49:14 Speaker 6: And let's say, I will for this example I will just leave this empty so use today's date let's go with 10 days.

49:26 Speaker 5: And let's go with the modus data and the last thing we need is our map key so use the map key and click the display results. So this is an example of what this API will return. So for the last 10 days in our custom area, there were only two fire points detected. Yeah. So this is all the APIs FIRMS provides. The last thing I wanted to mention is that all of this is also covered here with code examples. in our API code examples link. This will take you to our FHIR Data Academy portion of the site.

50:19 Speaker 5: And over here, there's examples of how to use our API with code. So we have some that are in Google Colab, which is a free tool to use. And we also have some examples that are with using the Python programming language. And you can also see there are, you can see the level that is needed to, or what level of knowledge is recommended for these tutorials or these examples. And for example, right here, this is what I covered. There is an example. So that is all I wanted to cover for you today. Thank you.

51:12 Speaker 5: And now let's go back to Melanie.

51:19 Speaker 1: Thanks, Hassan. So once again, a quick review of what we covered today. Active fire detection data can be accessed through several firm services. For data within the past seven days, data can be downloaded directly for specific world regions, time ranges, and sensors in CSV, KML, and shapefile formats. If looking for data older than the past seven days, then data can be downloaded by submitting an archive download request. Recent data can also be ingested into GIS platforms like ArcGIS or QGIS using web map or web feature services.

51:58 Speaker 1: Web map services generate geo-referenced map images for visualization. Web feature services generate vector-based geospatial data for visualization and analysis. Both of these require the user to generate a map key in order to use them, and that can be done on the firm's site. Historical data can also be accessed through the firm's API, although this differs from the archive download in that you're limited to a 10-day time range, and this service also requires a map key. We showed how you can visualize active fire information through download and import into Google Earth or using WMS or WFS services to ingest into GIS platforms like ArcGIS or QGIS.

52:45 Speaker 1: You can also create an email alert to be automatically notified of fires in your area of interest. Like many firm services, this can be customized by sensor, region, and frequency. And now we'll give a short summary of our entire training.

53:04 Speaker 1: In this training series, we were introduced to the FIRMS application, which provides geospatial data, products, and services for detecting, monitoring, and evaluating fires. Active fire detection products are the most frequently used data and are available from several polar and geostationary sensors. In part one, we learned about how active fires are detected by satellites. the strengths and limitations of polar versus geostationary sensors, the impact of sensor spatial resolution, and the location of a pixel within a swath, and how to navigate the global and U.S.-Canada FIRMS map applications.

53:43 Speaker 1: In Part 2, we learned about data latency, or the time between when an observation is made and when the data are available, and how FIRMS provides data at a range of latencies. depending on sensor characteristics like spatial resolution and processing requirements. And we saw how to use the firm's fire map application to monitor and track wildfire growth. We also learned about atmospheric and biophysical factors that can impact a sensor's ability to detect a fire, and strategies that can be used to investigate why a fire might be missing.

54:20 Speaker 1: such as looking for sensor outages or using imagery to check for the presence of clouds or smoke. And finally, today in part three, we heard about the many ways you can access and visualize active fire detection data, such as through download, web map service, web feature service, or using the firm's API. And one final reminder, you can access the homework for this training series from the training webpage starting today or tomorrow And answers should be submitted by May 14th. Participants who attend all three parts and submit the homework before the deadline will receive a certificate of completion.

54:59 Speaker 1: Certificates will be issued by email about two months after the training. And I just want to take another opportunity to acknowledge our fantastic guest presenters, Jenny Hewson, Brad Quayle, Atmar Olshina, Dylan Mendez, Ahsan Rodaf, And another thank you to Diane Davies, who helped with the overall training coordination and Q & A. And a huge thank you to our behind-the-scenes ARSA team, training coordinator Brock Blevins, content editor Jonathan O'Brien, and instructional designer Suzanne Monti.

55:36 Speaker 1: And one last time, I have included our contact information and links to our website, YouTube, and Twitter, or X, and a link to our sister program, Developed. Thank you so much again for attending our training series. We will now transition over to the Q & A. So get started with question one. Oh, and so just a reminder for everybody, this document will be available from the training webpage and you can see we've got a nice direct link to the firm's FAQ there. So question one, how do you obtain and how often update I think this is meant to be protected areas from Brazil and other countries.

56:25 Speaker 1: So I would invite any one of our guests who would like to speak to that answer.

56:30 Speaker 2: We update the protected areas like once every two or three years. And it comes from the World Database on Protected Areas.

56:44 Speaker 1: Okay. Question two. What is the difference between near real time and standard data? How long before the near real time data changes to standard or is updated?

57:01 Speaker 2: Yeah, generally there is a lag of three months, sometimes a little longer, but standard data is only available for MODIS, VIIRS, SNPP, and NOAA 20 currently. And yeah, the information more in detail is.

57:19 Speaker 1: Yeah, it looks like we have included a link to the FAQ there as well. Excellent. And we'll put that link in the chat in case there are any follow-up questions.

57:34 Speaker 1: Is there any, there's a lot of information there. Does anybody want to say anything else about that one?

57:41 Speaker 3: I think a lot of that information is coming from our FAQ, so.

57:46 Speaker 5: Okay.

57:48 Speaker 1: Question three. How can we determine, with the greatest possible degree of certainty, the range of dates from the earliest start date and latest end date during which a fire was active, based on the temporal sequence of hotspot detections within a specific geographic area? Since it seems like this was maybe covered in Part 2, Brad, did you want to speak to this one?

58:14 Speaker 4: As what's been articulated in response here, you can use our time since detection classification and adjust the calendar date settings or use the timeline down at the bottom of your screen and you could slide it down over to where you can see where that fire started and you'd be able to see the origination date there. When you hit the date where the fire has started, those detections and those initial detections will be you know, bright red, if you will, and you'll be able to visualize them that way.

58:47 Speaker 4: The other way you could do it as well is if you have at least a general idea of the window when the fire started on a given day or even within the day, you could use the sub-daily thresholding tools as well and isolate the detections that way. You can select a date and you can select an hour window like, okay, I want to look at everything within six hours of that time. And then you can continue to focus from there to identify when the first detection occurred.

59:20 Speaker 2: Excellent.

59:20 Speaker 1: Thank you. Question four. Given that each hotspot detection represents a pixel with a specific spatial resolution, How can we interpret the potentially burned area associated with each of these pixels reported by MODIS and VIIRT?

59:40 Speaker 3: Okay, so I put that insert in there. That's also coming directly from the FAQ link we put in there. It's not generally suggested to use active fire detections as a way to determine burned area.

59:58 Speaker 4: As Dylan mentioned, it's not recommended to use active fire detection data to estimate burned area. In some cases, it may work pretty well and may be pretty close. If you had another source in burned area, there's a dealing need from imagery and whatnot, but we recommend that you use the MODIS burned area product or any other type of burned area product that's derived using reflectance imagery through a change detection process. The reason why That's better. You have a pre-image and a post-image, let's say, and you can do a difference there and change detection process and identify that area burn based on that surface reflectance outputs there.

1:00:37 Speaker 4: But if you use active fire detection data from the polar orbiters, you're just getting a look a couple times a day. That fire could still be moving along in between those observations, even when you have a lot of polar orbiters up there collecting data. If it's particularly fast-fired in finer fuels and it's a wind-driven event, you would have big gaps in your active fire detection data not representing the area burned. Those are the primary considerations about avoiding using active fire detection data for a burned area.

1:01:16 Speaker 1: Thank you. Okay. Question five. If we could scroll down so we can see. Perfect. Will there be a confidence matrix for all the hotspots that have been generated through the detection? This might be referencing the FHIR confidence metadata.

1:01:43 Speaker 3: Yeah, I have that answer. If you're clicking on an active fire point from the firm's website and the map viewer, it will populate a table with confidence being one of the columns in the table. Also, the FAQ has more information on our confidence themselves.

1:02:07 Speaker 1: Thank you. Okay, question six. My question is, how much data can a person get using the API and the timing of getting that data? And by timing, I'm guessing how long it will take.

1:02:26 Speaker 3: Yes, so you can request various sorts and sources of our data. That answer is coming directly from when you apply for a map key on the firm's website. So that is what you see. The transaction limit is default set to 5,000 transactions per 10 minutes. And on a case-by-case basis, we can increase it if you need that.

1:02:56 Speaker 3: Contact us if you need that.

1:03:02 Speaker 1: Sounds good.

1:03:02 Speaker 3: Okay.

1:03:04 Speaker 1: And question seven. Another thing I want to add is that, is there any way to access data mountain cover, which is, oh, so snow cover data?

1:03:24 Speaker 4: Yeah, Brad here, I'll answer that. Yeah, recently, the snow cover layer has been added to the firm's viewer that's derived from beers. And, you know, so that... data set is updated on a routine basis. And the reason we've added that layer, just as a side note, is, you know, fire managers, you know, typically start to watch areas closer as the snow line, you know, recedes, and they'll pay attention, you know, closer to fire weather and fire weather forecast information and fire danger indices in those areas.

1:04:00 Speaker 4: And so that's why we, you know, provide that product in firms.

1:04:05 Speaker 4: And if you're interested in actually getting that data yourself to work with and download, yeah, you can go to the EarthData site and query that data and download it for use.

1:04:19 Speaker 1: Yeah, and if you ever have trouble finding a data set like that on EarthData, EarthData has a forum with which you can post questions and get answers from the EarthData help. Question eight. Will the area API support JSON or paginated responses in the future?

1:04:45 Speaker 3: Yeah, I put that in that answer. I think our reasoning, we only provide CSV at the moment for CSV responses for our APIs. It's not in our plan yet. And I think the reasoning is that if you're using the API,

1:05:02 Speaker 3: it shouldn't be too hard to translate it into a JSON object. Paginated responses, I'm not really sure what that means in the context, but feel free to email me if you have a more specific question.

1:05:18 Speaker 1: Yeah, or put a follow-up in the Q & A. Let's see if we get to it. Okay, question nine. I was wondering if it's possible to use the API for archive data. I would, for example, like to pull all the FHIR data from the SUMI MPP satellite from 2018 until 2025 for a given country. Can I do this with the API?

1:05:44 Speaker 5: Yeah, hello. Yeah, so you can do this using the API, but I wouldn't recommend it. I would recommend to go with the archive data approach and submit a download request for this particular example, it will be much faster to get the data that way. And it depends, the API is more suitable for programming, So when you want to get your data programmatically and do something with it, getting the data as a big chunk, I think in this example, that's what you want. You can get it as a file and then later on you can decide what you want to do with it.

1:06:51 Speaker 5: So I would recommend not using the API for this particular example.

1:06:57 Speaker 3: Yeah, just to add, the API is intended for like ongoing use of our current data. Our archive download would be sufficient for the.

1:07:10 Speaker 5: Example you put.

1:07:14 Speaker 1: Thank you. Question 10, is the map key for active information time bound? And I agree. I'm not sure exactly what time bound means. I guess maybe does it expire?

1:07:28 Speaker 3: So the map key does not expire itself. You can get our active FHIR information also from our active FHIR download, so you don't need a map key. You'd need a map key if you were going to build some programmatic way to access our most recent FHIR data. But if you would just like, the most recent fire data on a particular day, you can go download that file also

1:07:59 Speaker 3: for whatever region of the world you're in.

1:08:05 Speaker 1: Okay, question 11. Is there a way to extract near real-time fire occurrence or filter imagery to build a near real-time fire alerting system using this product for active fire monitoring in Africa?

1:08:29 Speaker 3: Yeah, so it sounds like our fire alerts would do what you are searching for, I think.

1:08:43 Speaker 1: Yeah, I agree from reading the question. Yeah, and if you need more information about that, we put some links in there, direct links to the firm's fire alerts. Question 12. If For example, I wanted to intersect the firm's data with a settlement layer, like the global human settlement layer, to get fires near buildings, would it be correct to use a 375 meter radius around each fire point as a sort of uncertainty measure?

1:09:16 Speaker 3: So, for the 375-meter product that you're referring to here, that means that under the satellite, exactly, there will be a 375-meter square box projected on the Earth. And then if there's a fire somewhere inside that 375-meter box, then that pixel will turn red and show up on our firm's map viewer. So any red pixel on the firm's map viewer means that there's a fire detected somewhere within that pixel. So you don't need to go outside of the pixel for an uncertainty measure. It's inside of that pixel.

1:09:58 Speaker 3: And yes, it would be correct to use the global human settlement layer to determine if it was near a human settlement layer or human settlement. I don't know if there's any more information Brad.

1:10:17 Speaker 4: Yeah. One thing I may recommend doing, which we've done in some applications before, is, you know, giving it the question part, at least for considering uncertainty there, is you could take the diagonal distance as the buffer instead of the orthogonal distance, you know, half that 375-meter pixel resolution would be that, you know, buffer to the orthogonal distance. direction outward from that centroid. If you take that diagonal distance and use that instead, maybe also apply the positional accuracy that you expect to see with motors and veers.

1:10:58 Speaker 4: I can't remember if it's off the top of my head, but they would both be around somewhere between 50 and 75 meters. And that could be your new radius. So take that diagonal distance and then apply the positional accuracy, and that would be your radius that you could consider using And I can write something up there to put in there on that response, Melanie.

1:11:22 Speaker 1: Sounds perfect.

1:11:24 Speaker 3: So I can– so WebFactor is the choice between WFS and WMS based on– I can write something also to put in there. But the choice between web feature service and web mapping service– a web mapping service really just puts images over a map. And the Web Feature Service puts the exact features, polygons, points, et cetera, onto the map so you can query different spaces that it takes up on the map or things of that nature. If you're really just looking to see where the fires are on the map and you're not looking for any spatial information on them, then Web Mapping Service would suffice.

1:12:16 Speaker 3: A web, by nature, web feature service is a little bit more intensive on the server.

1:12:25 Speaker 1: Okay. Question 14. What's the starting temperature in Kelvin of a flame? Is there a threshold of Kelvin or temperature that can be, that can help define if it's an active flame, a thermal anomaly, or a hot area, for example, just burned? And maybe I would see if Brad wanted to speak to that one.

1:12:56 Speaker 3: Sure. Yeah.

1:12:59 Speaker 4: Degrees Kelvin is, you know, the estimated temperature based on the observed radiance in the thermal bands for the sensor being used. And those estimated temperatures are, you know, factor into the fire detection algorithms that we talked about in section one of this training using.

1:13:20 Speaker 3: The mid.

1:13:22 Speaker 4: The mid-wave infrared and the long-wave infrared bands. And just for reference, you know, nominally, the Earth's surface temperature is about 300 degree Kelvin, which, you know, it's somewhere like around 80 degrees. It's that average temperature there, you know, for a relationship there. But those smoldering fires versus, you know, what we call an active flaming fire, you know, an active flaming fire can be very hot, you know, somewhere around 1,000 degrees. degrees Kelvin or even hotter, you know, 1,200 degrees or even warmer.

1:13:52 Speaker 4: And the smoldering fires, you know, stuff that's, you know, and it's burning low and maybe takes off a little bit every once in a while, but again, it's smoldering, not really active fire. You know, they're in that 600 to 800 degree Kelvin range.

1:14:09 Speaker 1: Okay. Question 15, and it looks like we have two questions left. So, if you have a question, now is the time. As a matter of interest, how does the firm's historical burned area data, and I'm assuming that's the MODIS burned area, compared to that available from Copernicus CDS, fire burned area from 2001 to present derived from satellite observations?

1:14:43 Speaker 4: I haven't done a comparison myself between the Copernicus dataset to the Modus burned area, but we'll take a closer look at that. We have some people on the firm's team who have, and we'll be able to provide you a good answer to that question. So, thanks for that question.

1:15:04 Speaker 1: Okay. Question 16. Is there an animation or movie-like tool in firms, for example, to play an animation of the detection over a day in a subdaily view? It looks like maybe that's a planned development.

1:15:23 Speaker 4: Yeah.

1:15:24 Speaker 1: This is Diane.

1:15:26 Speaker 7: Yeah, unless someone wants to comment, we have something in development, and hopefully it will be released.

1:15:33 Speaker 1: Within the next month or so.

1:15:36 Speaker 3: That's exciting.

1:15:38 Speaker 7: It is.

1:15:43 Speaker 1: Question 17, and this is in reference, I think, to the time-bound question. They were wondering if the key timed out after use or the same key can be used for multiple days or it's just a one-time request scenario.

1:15:58 Speaker 3: Yeah, so the key is good indefinitely.

1:16:04 Speaker 2: Great.

1:16:04 Speaker 1: So save your key. Question 18, will there ever be a current year archive? In other words, an active fire hotspot information for each satellite since January 1st, not only the last seven days. Download option for shapefiles, KMLs, or CSV without having to do the email submission.

1:16:28 Speaker 3: So we like to keep track of who is requesting our data. So no, you're describing what our archive data download does. It has all of our archive data for the past many years before.

1:16:49 Speaker 7: And also just add to that, it's kind of like a practical reason as well, because we just have to keep updating it. If there was enough demand and there was a good reason to do it, we might consider it.

1:17:06 Speaker 1: But currently, we.

1:17:09 Speaker 7: Just do it when a year is complete. Also, another thing to note is that the standard quality data isn't available straight away. So it's usually available with a three- to four-month lag because this is QA'd by the fire science team. So it probably... wouldn't be as current as you'd like it to be. So that's another reason that it's delayed.

1:17:34 Speaker 3: I hope that helps.

1:17:40 Speaker 1: OK. Do we have a question for question 19? How does NASA show PM data on air quality? So I'm assuming this means nighttime data for air quality. I'm not sure what this person might be referring to. Is that PM particulate matter? I was assuming PM nighttime. You're probably right.

1:18:17 Speaker 7: So that isn't something that we currently have in firms, but it is something that we are Definitely interested in looking at, we don't have any immediate plans to do anything, but it is on our radar. And if you have specific use cases, we are always interested to hear what they are. And that may help bump it up the priority list. But we don't have any immediate plans.

1:18:47 Speaker 1: And Janet? offline, I can put some links into satellite-derived PM2. 5 data sets. So I can put some links in there to available data sets.

1:19:08 Speaker 2: OK.

1:19:11 Speaker 1: It looks like that brings us to the end of our questions. So I just want to take a second to once again thank our amazing guest presenters and guest speakers. Thank you so much for these amazing presentations and for a great job answering all of these great questions in the Q & A. Thanks to you all for attending. I'll just invite Brad, Diane, Dylan, Asan, Atmar, any of our presenters online today, if you would like to unmute and say any few words, I'm happy to give you a second to do that before we close today.

1:19:59 Speaker 7: I would just like to say thank you to everybody for showing up for these training sessions. We're always Delighted to see how firms data is being used. So please feel free to get in touch if you have suggestions of what you'd like to see. Then again, please let us know. But I hope you found this training helpful. We're really grateful to the RCEP team and to everybody that's all the firms teams that have presented.

1:20:24 Speaker 1: So, yeah, thanks, Melanie and the rest of you.

1:20:32 Speaker 4: Yeah, Brad here. I echo Diane's sentiments there. Thank you, everyone, for your participation and interest in this training. It's great to see this much interest. And, yes, if you have any questions about the use of the data or suggestions for enhancements, as Diane said, yes, please reach out to us via email, and we look forward to hearing from you. Thank you.

1:21:01 Speaker 1: Okay. Thank you once again, everybody. I hope you have a great rest of your day. Thanks, everyone.