NASA ARSET Introduction and Demonstration of STREAM

0:00 Speaker 1: Hello, and welcome everyone to this training on monitoring water

0:05 Speaker 1: quality in lakes and coastal regions using satellite based tool

0:10 Speaker 1: for Rapid Evaluation of Aquatic environments or STREAM. My name

0:15 Speaker 1: is a meta meta from NASA's Applied Remote Sensing Training

0:19 Speaker 1: Program or our SET program, and we have a guest

0:22 Speaker 1: speaker today, William Wainwright from NASA Goddard Spaceword Center, who

0:26 Speaker 1: will be focusing on introduction and demonstration of STREAM. We

0:32 Speaker 1: will start with a brief overview of our SET program.

0:36 Speaker 1: Our SET is part of NASA's Applied Sciences Program, its

0:40 Speaker 1: capacity building program, and provides accessible, relevant and cost free

0:45 Speaker 1: training on remote sensing satellites, sensors, methods, and tools. Trainings

0:51 Speaker 1: include a variety of applications of satellite data and our

0:55 Speaker 1: tailor to audiences with a variety of experience levels. Specific

1:00 Speaker 1: are SET focuses on these thematic areas for applications agriculture,

1:05 Speaker 1: climatory resilience, disasters, ecological conservation, health and air quality, water resources,

1:12 Speaker 1: and wildland fires. SET trainings are offered online or there

1:17 Speaker 1: are in person trainings offered as well. Online trainings can

1:21 Speaker 1: be live and instructor led like this one or They

1:25 Speaker 1: can be asynchronous and self based, available from our set website.

1:29 Speaker 1: The link is provided here. These trainings are cost free

1:33 Speaker 1: as I mentioned earlier, and only use open source software

1:37 Speaker 1: and data. Many of our trainings are offered in multiple languages,

1:42 Speaker 1: especially their material is available in Spanish for many trainings,

1:48 Speaker 1: and they accommodate differing levels of expertise. Please visit ourset

1:52 Speaker 1: website for more information. We'll start with an overview of

1:58 Speaker 1: this training, monitoring what quality in lakes and coastal regions

2:02 Speaker 1: using stream now. It is well recognized that monitoring water

2:08 Speaker 1: quality is vital for planning and managing drinking water treatment,

2:13 Speaker 1: for public health and ecosystem advisories, and also for assessing

2:18 Speaker 1: health and productivity of fresh water and saltwater fisheries. Conventional

2:24 Speaker 1: INC two measurements of water quality parameters are expensive and

2:28 Speaker 1: have limited spatial and temporal coverage. As you can see

2:32 Speaker 1: here there's an example of Lake Aria in the US.

2:36 Speaker 1: Is the western part of the lake which is prone

2:40 Speaker 1: to have harmful alcohol bloom or ABS. And if you

2:44 Speaker 1: can see these dots they represent where INCIT two measurements

2:49 Speaker 1: are taken. As you can see, not the entire lake

2:52 Speaker 1: is covered, just a few locations are covered these. Some

2:56 Speaker 1: of these measurements are weekly. There's buis which provide continuous data,

3:02 Speaker 1: but there are very few. Also, there are so many

3:05 Speaker 1: water bodies in the world and long coastal areas, they

3:09 Speaker 1: all cannot be measured with in water samples and water

3:14 Speaker 1: quality parameters cannot be assessed that way. On the other hand,

3:18 Speaker 1: if you can see this lenset image on August thirteen,

3:22 Speaker 1: twenty thirty four where alcohol bloom occurred in lake around

3:27 Speaker 1: the same area, now you can see a continuous picture

3:31 Speaker 1: and also see pattern where intensity of HEB was low

3:36 Speaker 1: and high in the lake. So that way, remote sensing

3:39 Speaker 1: provides a cost effective way to assess water quality. Also,

3:44 Speaker 1: it can cover thousands of lakes and coastal waters with

3:47 Speaker 1: improved coverage. Currently there are several satellites flying with sensors

3:55 Speaker 1: allowing water quality monitoring. The table here is for your

3:59 Speaker 1: reference and provides information about the satellites, sensors, measurements they do,

4:05 Speaker 1: and their special and temporal resolutions. Just briefly going over

4:11 Speaker 1: the current satellites, it's lens AT eight, lands At nine, Tera,

4:15 Speaker 1: Aqua S and pp GPSS GPSS are Noah satellites. NOAH

4:20 Speaker 1: twenty and twenty one, Sentinel two A, two B, two C,

4:25 Speaker 1: and Sentinel three A and three B. These are European

4:29 Speaker 1: Space Agency satellites and the latest NASA mission base that

4:35 Speaker 1: is flying currently as well. All these satellites are in

4:39 Speaker 1: polar orbits as you can see, and these are the

4:42 Speaker 1: sensors flying on these satellites. All these acronyms are defined

4:48 Speaker 1: at the end of this presentation should be available to

4:50 Speaker 1: you in your dog so this is only only two

4:56 Speaker 1: mods weirs, MSI OLG and o CI. These are the

5:02 Speaker 1: water quality monitoring sensors. They have spectral measurements in optical

5:07 Speaker 1: bands ranging from visible, near infrared, middle infrared, and some

5:11 Speaker 1: also have thermal infrared bands. So all these can be

5:17 Speaker 1: used to monitor water quality parameters. They can be derived

5:21 Speaker 1: from these data top of the atmosphere reflectance in optical bands.

5:27 Speaker 1: As you can see, they all have different temporal and

5:31 Speaker 1: special coverage and resolutions. Thing to note here is that

5:37 Speaker 1: Lenset eight and nine and Sentinel two series they have

5:42 Speaker 1: the highest special resolutions it's intensive meters and the other

5:48 Speaker 1: sensors like motor spheres o g oci, they have low

5:55 Speaker 1: to medium resolutions ranging from two fifty meters all the

5:58 Speaker 1: way to one kilometer. On the other hand, Lancet and

6:03 Speaker 1: Centinal two they have narrow swarths, whereas Terra acquire S

6:09 Speaker 1: and ppgps as Sentinel three and Pace have broad swaths

6:14 Speaker 1: in thousands of kilometers as you can see here, so

6:18 Speaker 1: they all can be used for water quality monitoring at

6:22 Speaker 1: different special and temporal resolutions. Lancet each one has sixteen

6:28 Speaker 1: dairy revisit time, so with both of them it's about

6:32 Speaker 1: eight day, and Sentinal two series has about five day

6:36 Speaker 1: revisit time. All other sensors that we have there one

6:41 Speaker 1: to two day revisit time, so there are trade offs

6:44 Speaker 1: between temporal and special resolutions.

6:47 Speaker 2: As you can.

6:47 Speaker 1: See, our set has several trainings on water quality monitoring

6:54 Speaker 1: using remote sensing, and the links and titles are provided here.

6:59 Speaker 1: These are set trainings focused on water quality parameters derived

7:05 Speaker 1: from sensors like motives, WIRs, OLG and oci with relatively

7:11 Speaker 1: low to medium resolutions from two fifty meters to one kilometer.

7:16 Speaker 1: Before stream higher resolution data that is, at twenty to

7:20 Speaker 1: thirty meter resolution data, water quality parameters were not readily

7:24 Speaker 1: available and had to be derived by using algorithms developed

7:29 Speaker 1: from satellite data and in CITU measurements for the water

7:33 Speaker 1: body of your interest. An example of that can be

7:36 Speaker 1: found here in this training. So here is an example

7:41 Speaker 1: of chlorophyll a concentration from pace oci at one kilometer.

7:46 Speaker 1: I can clearly see that coastal areas and open oceans

7:49 Speaker 1: are resolved and covered nicely in pace oci, and bigger

7:55 Speaker 1: lakes can also be seen like Great Lakes and other

7:59 Speaker 1: bigger water body can be resolved. But there are several

8:02 Speaker 1: small water bodies that are not resolved by this medium

8:07 Speaker 1: to low resolution sensors. And that brings us to today's topic,

8:13 Speaker 1: and that is Stream for water quality monitoring. Stream, as

8:17 Speaker 1: we will see later, uses LENST eight and nine and

8:22 Speaker 1: sentinal two series data and derives water quality parameters at

8:27 Speaker 1: twenty to thirty meters special resolution. It's an interactive web

8:31 Speaker 1: tool and an API enabling monitoring or water quality in

8:36 Speaker 1: near real time since twenty eighteen. Using Stream, you can

8:42 Speaker 1: see that they can resolve smaller water bodies if the

8:47 Speaker 1: lake or water body is one hundred meters square or

8:50 Speaker 1: bigger than STREAM can resolve it. Currently, Stream focuses on

8:56 Speaker 1: inland lakes and coastal areas in the US and some

9:00 Speaker 1: lakes in certain countries are also available from stream. What

9:05 Speaker 1: quality data are derived by using an open source model

9:09 Speaker 1: and we will cover that as well. An example shown

9:13 Speaker 1: here again is pace OCI and stream chlorophile A concentration.

9:18 Speaker 1: This is in Montana, uh This is Lake Canyon Ferry

9:23 Speaker 1: and you can see that this is a fairly large

9:26 Speaker 1: lake and can be seen in both pace OCI data

9:31 Speaker 1: as well as sentinel to data. Thing to notice here

9:35 Speaker 1: is that there are several small lakes that we see

9:38 Speaker 1: here in stream. They are not visible here in pace OCI.

9:42 Speaker 1: Also you can see that being this being high resolution,

9:46 Speaker 1: you can see more details in chlorophil A concentration distribution

9:50 Speaker 1: in the lake. With that, our training learning objectives are

9:58 Speaker 1: that by the end of this training you will be

10:00 Speaker 1: able to identify the purpose, capabilities, and benefits of this

10:04 Speaker 1: stream tool for analyzing inland and coastal water bodies. Identify

10:10 Speaker 1: the process to use stream to monitor chlorophil A concentration,

10:14 Speaker 1: secut disk depths and total suspended solids in lakes and

10:18 Speaker 1: coastal waters. Identify the steps to use stream API to

10:23 Speaker 1: search and download these parameters for a specific time period.

10:29 Speaker 1: Examine time series of chlorophil A concentration SECHI disk steps

10:34 Speaker 1: and total susmcted solids using QGIS and identify how an

10:39 Speaker 1: open source machine learning model based on mixture density network

10:44 Speaker 1: enables users to estimate water quality parameters for any inland

10:49 Speaker 1: or coastal water body which is greater than one hundred

10:52 Speaker 1: meters square or so worldwide. Here are the prerequisites for

10:59 Speaker 1: the strain. First one is Fundamentals of remote sensing. This

11:03 Speaker 1: ur set training provides background about satellite remote sensing principles,

11:08 Speaker 1: different satellite orbits, different data levels that described in here.

11:13 Speaker 1: The second training is monitoring water quality of inline leaks

11:17 Speaker 1: using remote sensing. This training provides information about how to

11:21 Speaker 1: use satellite data together with institute data to develop statistical

11:25 Speaker 1: algorithm and then can be used with satellite data to

11:29 Speaker 1: drive water quality parameters. There will be two parts to

11:34 Speaker 1: this training. Today Part one will focus on introduction and

11:38 Speaker 1: demonstration of stream and part two next week on February seventeenth,

11:42 Speaker 1: will focus on introduction to a machine learning model to

11:46 Speaker 1: estimate water quality parameters based on satellite observations. There will

11:52 Speaker 1: be one homework assignment will be open on February seventeenth

11:56 Speaker 1: and it will be due on March tenth posted on

12:00 Speaker 1: our set website. Our certificate of completion will be awarded

12:04 Speaker 1: to those who attend all live sessions and complete the

12:08 Speaker 1: homework assignment before the given due date. We'll start with

12:14 Speaker 1: today's session on introduction and demonstration of Stream and specific

12:18 Speaker 1: objectives for today are that by the end of part one,

12:22 Speaker 1: you will be able to identify features of stream website

12:25 Speaker 1: and API for selection and display water quality parameters for

12:30 Speaker 1: selected leaks and coastal areas. Recognize how to download water

12:36 Speaker 1: quality parameters using stream API, and recognize how to display

12:41 Speaker 1: maps and time series of Chlorophila concentration, secy disk depth

12:46 Speaker 1: and total suspended solids for selected water body. Outline for

12:52 Speaker 1: today is that our guest speaker, William Wainwright, will start

12:56 Speaker 1: by providing an overview of stream. He will also have

13:00 Speaker 1: demonstration of making maps of water quality parameters using stream

13:04 Speaker 1: web tool and how to use stream API to find

13:10 Speaker 1: water quality parameters multiple days, and then we will use

13:15 Speaker 1: qgis to make time series of these water quality parameters.

13:20 Speaker 1: Just a note about asking questions. Please put your questions

13:24 Speaker 1: in the question box and we will attest them at

13:26 Speaker 1: the end of the webinar.

13:28 Speaker 2: Feel free to enter.

13:29 Speaker 1: Your questions as we go and we'll try to get

13:32 Speaker 1: to all the questions during the Q and A session.

13:34 Speaker 1: After the webinar, the remainder of the questions will be

13:38 Speaker 1: answered in the Q and A document which will be

13:41 Speaker 1: posted to the training website about a week after the training.

13:46 Speaker 1: With that, we introduce our guest speaker for today, William Wainwright.

13:52 Speaker 1: William is a senior scientific programmer with the Freshwater Sensing

13:56 Speaker 1: Group at Science System and Applications in Code with CAR

14:01 Speaker 1: Space Flight Center Code six nineteen, that is, in the

14:04 Speaker 1: Terrestrial Information Lab. William comes from a background of astronomy

14:10 Speaker 1: and physics, with a master's in estrophysics from Rochester Institute

14:14 Speaker 1: of Technology. Now, William applies his skill set to the

14:19 Speaker 1: development of a remote sensing water quality platform called Stream.

14:23 Speaker 1: With that, we invite William to talk about Stream.

14:27 Speaker 3: William, thanks, Amita. My name is will Waynwright. I'm the

14:34 Speaker 3: developer of Stream along with my colleagues Akash, Navide, Arun

14:41 Speaker 3: and Ryan. Together we make up the Freshwater Sensing Group

14:45 Speaker 3: at SSAI and as a team we developed the models

14:49 Speaker 3: behind as well as the platform itself of Stream. Stream

14:55 Speaker 3: is a water quality portal developed by the Freshwater Sensing group.

15:00 Speaker 3: We are motivated by real issues like public health impacts

15:04 Speaker 3: to aquaculture and the larger ecosystem, as well as changing

15:10 Speaker 3: health of water systems from events like harmful alga blooms.

15:15 Speaker 3: Our main goals as a group are to improve the

15:17 Speaker 3: accuracy of remote sensing derived water quality observations in both

15:24 Speaker 3: inland and coastal waters, as well as making those products

15:28 Speaker 3: more accessible. The natural endpoint of those goals is the

15:33 Speaker 3: tool that we call Stream. The data pipeline for Stream

15:38 Speaker 3: starts with the top of atmosphere satellite data. Our team

15:42 Speaker 3: has developed water quality models for many different satellites and sensors,

15:47 Speaker 3: but we chose to begin with Stream, supporting these two

15:50 Speaker 3: groups here lands At eight and nine and the Sentinel

15:55 Speaker 3: two family. These satellites all have the optical band its

16:00 Speaker 3: necessary for us to derive water quality measurements, as well

16:04 Speaker 3: as relatively high spatial resolution compared to some of the

16:08 Speaker 3: other water quality compatible satellites that are available. This increased

16:14 Speaker 3: spatial resolution does come at the cost of temporal resolution,

16:18 Speaker 3: though Lanset features an eight day revisit rate. Between the

16:24 Speaker 3: pair of Lancet eight and nine and Sentinel to A

16:28 Speaker 3: and B operate as a pair, giving us a five

16:32 Speaker 3: day revisit rate. Sentinel two C operates out of phase

16:37 Speaker 3: with A and B and gives you an extra overpass

16:39 Speaker 3: of your site between those five days. In order to

16:44 Speaker 3: retrieve water quality products, we do a few steps to

16:47 Speaker 3: atmospherically correct the top of atmosphere data that Landset and

16:51 Speaker 3: Sentinel two provide. The main engine of this atmospheric correction

16:58 Speaker 3: is a pair of mixture density networks. These are a

17:01 Speaker 3: type of neural network that predict all of the water

17:04 Speaker 3: quality metrics at once, as well as an estimated uncertainty.

17:10 Speaker 3: This methodology is particularly useful in our application, as we're

17:15 Speaker 3: trying to get a sense of how all of the

17:17 Speaker 3: key parameters operate as a connected system rather than just individually.

17:24 Speaker 3: After the first round of atmosphere correction, we're left with

17:27 Speaker 3: the remote sensing reflectants. Then we use a second mixture

17:32 Speaker 3: density network to estimate the water quality parameters, namely chlorophyl a,

17:38 Speaker 3: total suspended solids, and the SECI disk depth. If you'd

17:44 Speaker 3: like to learn more about the methodology and the models

17:47 Speaker 3: that we use, the next r set training will be

17:50 Speaker 3: a deep dive into these models.

17:51 Speaker 4: By Ryan o'she.

17:56 Speaker 3: What really enables Stream to deliver water quality estimates, though,

18:00 Speaker 3: are large training data sets like Gloria and airnet oc.

18:06 Speaker 3: These data sets contain thousands of in situ water samples

18:10 Speaker 3: that were co witnessed by satellite overpass from water bodies

18:14 Speaker 3: of all sizes and all makeups across many different areas

18:18 Speaker 3: around the world and across decades. The matchup between in

18:25 Speaker 3: situ and satellite observations are what allow for us to

18:30 Speaker 3: fine tune our models, and the sheer diversity of water

18:34 Speaker 3: bodies represented are what allows Stream to offer such effective

18:38 Speaker 3: water quality estimates for the wide array of inland and

18:43 Speaker 3: coastal waters across the US and more, and that leads

18:49 Speaker 3: us to the website and API where you can access

18:52 Speaker 3: these water quality maps. The main tool on Stream is

18:56 Speaker 3: the interactive map page, where you can see all of

18:59 Speaker 3: the day for a given day's satellite overpass at once. Normally,

19:05 Speaker 3: the data from these satellites are segmented into square tiles

19:09 Speaker 3: in a grid. Stream stitches these tiles back together so

19:14 Speaker 3: you can see the connected water systems in full context.

19:18 Speaker 3: The webmap has many other features that I'm excited to

19:21 Speaker 3: get into shortly in our demo.

19:25 Speaker 4: Aside from the map.

19:27 Speaker 3: You can search and download scenes from the archive, which

19:31 Speaker 3: stores the results of our daily processing since it began

19:34 Speaker 3: in June of twenty twenty four.

19:38 Speaker 4: These products on.

19:39 Speaker 3: The archive are full resolution and georeferenced, and they represent

19:45 Speaker 3: an individual tile in the satellite's grid. If you'd prefer

19:50 Speaker 3: to do things in a more automated fashion, you can

19:56 Speaker 3: use our Wristful API to perform most of the same

19:59 Speaker 3: fund that you're able to on the website, such as

20:03 Speaker 3: query and downloading product maps, and that takes us to

20:09 Speaker 3: our demo. In this demo, I'll show you more of

20:14 Speaker 3: Stream's features as well as an example of how to

20:18 Speaker 3: search water quality maps. I'll show you how to query

20:22 Speaker 3: and download scenes of an area of interest, in this

20:25 Speaker 3: case Pyramid Lake. Stream does have full Conus coverage though,

20:30 Speaker 3: so the things that I'm about to demo are applicable

20:33 Speaker 3: to many inland and.

20:34 Speaker 4: Coastal water bodies.

20:37 Speaker 3: If you have any questions, please enter them in the

20:39 Speaker 3: chat and I will address as many as I can

20:41 Speaker 3: at the end. All right, so here we are on

20:48 Speaker 3: the home page of Stream, and as you scroll down

20:52 Speaker 3: you'll see a little bit about Stream and some of

20:55 Speaker 3: the things that we talked about in our presentation, you'll

20:59 Speaker 3: also find video demo on using.

21:01 Speaker 4: Streams, map page and.

21:03 Speaker 3: Features, and some of the latest updates that we are

21:07 Speaker 3: excited to announce.

21:10 Speaker 4: Back up at the top.

21:12 Speaker 3: If you go on to the map page, you can

21:14 Speaker 3: get there from either the nav bar or this explore

21:16 Speaker 3: map button that takes us to the map page with

21:25 Speaker 3: nothing loaded in. And if we're looking for an area

21:30 Speaker 3: that we know, we can go ahead and type it

21:32 Speaker 3: into this location search bar. So here we'll look for

21:35 Speaker 3: Pyramid Lake and it's found it and taken us right

21:42 Speaker 3: over to Peermid Lake. Now the easiest way to search

21:46 Speaker 3: for data for Pyramid Lake.

21:48 Speaker 4: We're going to go ahead and zoom out a.

21:50 Speaker 3: Bit here because the satellite tiles are quite large, and

21:54 Speaker 3: then we're going to enable one of our satellite grids.

21:56 Speaker 3: So here we'll look at the Sentinel two grid and

22:02 Speaker 3: you can see those square tiles that the satellite breaks

22:04 Speaker 3: data up into that I was talking about. This lake

22:08 Speaker 3: is a good example because you can see it also

22:11 Speaker 3: falls on the border of two different tiling schemes. So

22:14 Speaker 3: here we've got the one starting with ten, on the right,

22:18 Speaker 3: we've got the one starting with eleven, and in the

22:20 Speaker 3: middle there's a bit of overlap. What that is is

22:24 Speaker 3: just different segments of the satellite grid, so these tiles

22:28 Speaker 3: can sometimes be represented by more than one satellite overpass.

22:35 Speaker 3: When you've identified the area that you're interested in and

22:37 Speaker 3: you've got your satellite grid enabled, you can just go

22:40 Speaker 3: ahead and click on that tile and it will bring

22:44 Speaker 3: up the search results for that tile. So here it's

22:51 Speaker 3: brought us to the search results for the sentinel tile

22:55 Speaker 3: eleven TKE and you can see the results as sorted

22:59 Speaker 3: by date of availability. And we want to look at

23:06 Speaker 3: an example of a harmful alghol bloom affecting pyramid lakes,

23:09 Speaker 3: So we're going to go back to October the tenth

23:12 Speaker 3: of last year. When you click on one of these results,

23:17 Speaker 3: it'll bring you to the archive page where you can

23:19 Speaker 3: download all of the results. So clicking anything in this

23:24 Speaker 3: middle column would download the product represented on the rate.

23:29 Speaker 3: So this would download the full resolution chlorophyl a map

23:33 Speaker 3: as well as some of the other products. The bottom

23:35 Speaker 3: you can also find the compressed archive for all the products.

23:39 Speaker 3: But on the left column we have the map link

23:42 Speaker 3: and so here I'm going to click on the map

23:44 Speaker 3: link for chlorophyl a.

23:47 Speaker 4: And that will take us back to our map.

23:50 Speaker 3: Zoom back to the tile we were looking at, but

23:53 Speaker 3: it will populate all the data for the day that

23:55 Speaker 3: we clicked on. So again we got here by clicking

24:00 Speaker 3: on the data for this tile, but you can see

24:02 Speaker 3: the data for other tiles that were observed on the

24:05 Speaker 3: same day are also loaded in. We can go ahead

24:09 Speaker 3: and turn off the grid now and we're going to

24:12 Speaker 3: zoom in. You can see as we zoom in, the

24:17 Speaker 3: map will increase in resolution, so we can see on

24:25 Speaker 3: October tenth, there's a bit of a powerful happening in

24:31 Speaker 3: the water, but it's it's not extreme and I wouldn't

24:34 Speaker 3: constitute a harmful alglorbim just quite yet. On the right panel,

24:42 Speaker 3: you can add additional layers. So when we were looking

24:46 Speaker 3: in the archive, I saw that the next day that

24:48 Speaker 3: had data was the seventeenth. So if we go ahead

24:52 Speaker 3: and select the seventeenth and add layer, it'll load another

24:57 Speaker 3: layer on top for the seventeenth, And now you can

25:00 Speaker 3: see on the seventeenth there's quite a bit.

25:03 Speaker 4: Of action happening.

25:05 Speaker 3: We want to step back and forth between them. We

25:08 Speaker 3: can scroll down to the active layer menu. We can

25:12 Speaker 3: see the layer for the seventeenth is on top and

25:15 Speaker 3: the layer for the tenth is on the bottom. You

25:17 Speaker 3: can drag to reorder the layers if you'd like, but

25:20 Speaker 3: you can also toggle and hide them to step between

25:23 Speaker 3: them as well as if you click expand on any

25:28 Speaker 3: of the individual layers options, you can change the color

25:32 Speaker 3: palette to.

25:32 Speaker 4: Your suiting.

25:39 Speaker 3: When you are looking at multiple layers, though, as you

25:43 Speaker 3: move your cursor around, you'll see that on the right

25:48 Speaker 3: the coordinates update to where your mouse cursor is hovering,

25:53 Speaker 3: as well as the live redout of the pixel value

25:56 Speaker 3: for each layer. So if we hover a particularly area here,

26:01 Speaker 3: you can see that on the tenth it was a

26:04 Speaker 3: pretty low chlorophyll concentration of two point four seven, but

26:09 Speaker 3: on the seventeenth it was high at fifty five point

26:12 Speaker 3: four to two, and you can get a sense of

26:14 Speaker 3: that as you move your mouth around to different areas.

26:21 Speaker 4: So if you want to access your.

26:24 Speaker 5: Data via the API, we can go back to the

26:28 Speaker 5: homepage and on the rate here we have the API documentation.

26:37 Speaker 3: This is in an open API document for the RESTful

26:42 Speaker 3: API that serves stream, so on the left you can

26:46 Speaker 3: see the various endpoints and details of how to access them.

26:50 Speaker 3: The parameters that they require and everything you need to

26:55 Speaker 3: get off the ground with scripting. But in the top

26:58 Speaker 3: right we also have a button to download the example

27:00 Speaker 3: script that I've created in Python that you can modify

27:04 Speaker 3: to get working with Stream a little easier. So here

27:12 Speaker 3: I've pulled up the API example scripts and at the

27:16 Speaker 3: top you can see some configuration values that you can change.

27:21 Speaker 3: So we've got a tile list, which is just a

27:23 Speaker 3: list of the tiles sentinel to or landst format. There's

27:28 Speaker 3: some examples here, the date range in the specified format

27:34 Speaker 3: for the start and end date of your query. And

27:38 Speaker 3: then the products list that you would like to download,

27:41 Speaker 3: and these are the shorthands that the API expects of.

27:44 Speaker 3: The valid options are specified here, but these are just

27:47 Speaker 3: your products for chlorophyl a, total suspended solids and the

27:51 Speaker 3: SECI disc depth. I've also tried to comment the rest

27:56 Speaker 3: of the script if it's something you'd like to dive

27:58 Speaker 3: into and tweak yourself, but for now let's stick to

28:03 Speaker 3: the examples. So it's very simple to run. You just

28:07 Speaker 3: need to run it with any version of Python, and

28:19 Speaker 3: you can see that it's found eight dates that stream

28:22 Speaker 3: has data for the tile that we've asked for in

28:25 Speaker 3: the date range that we've asked for. So this title

28:28 Speaker 3: here eighteen suj. It's a part of Chesapeake Bay and

28:34 Speaker 3: we're downloading data for most of January of twenty twenty six.

28:43 Speaker 3: It's also gone ahead and created a stream downloads folder

28:46 Speaker 3: in the location that we have this script, and it's

28:50 Speaker 3: downloading those full resolution product maps for the tile for

28:54 Speaker 3: Chlorophyla to that folder. And now that it's completed, we

29:08 Speaker 3: can go ahead and check and see that we have

29:11 Speaker 3: a new folder and it contains those products.

29:14 Speaker 4: That we requested.

29:20 Speaker 3: Now, if we want to look back at Pyramid Lake again,

29:23 Speaker 3: what we would do is I'm going to clear out

29:26 Speaker 3: the folder, going to delete the maps that we downloaded.

29:33 Speaker 3: So here I'm just deleting the folder to get the

29:37 Speaker 3: tiles out of the way so I don't get them

29:39 Speaker 3: confused between the next batch of tiles from Pyramid Lake.

29:43 Speaker 3: But you could rename the folder or even change the

29:47 Speaker 3: location of the download directory in the script. I'm going

29:53 Speaker 3: to go back to editing this script, so instead of

29:58 Speaker 3: eighteen suj, we want eleven tke. That's the tile for

30:04 Speaker 3: Pyramid Lake. If you don't remember or you need help

30:09 Speaker 3: finding the tile that you're interested in, remember that the

30:12 Speaker 3: satellite grid checkbox on the map page, load in the

30:16 Speaker 3: full satellite grid with all the title names in the

30:19 Speaker 3: middle of the grid. And then for the date range

30:24 Speaker 3: we were looking at October, we can download Let's just

30:29 Speaker 3: do the full month of October, and then we're going

30:37 Speaker 3: to go ahead and stick with chlorophyl a for the

30:39 Speaker 3: product because we're interested in looking at chlorophyl for the

30:43 Speaker 3: harmful algoa bloom. So we can save our changes, rerun

30:49 Speaker 3: the script the same way we just did, and it

30:56 Speaker 3: found seven products for Pyramid Week in October, so we

31:17 Speaker 3: can see it successfully downloaded everything, and if we look

31:21 Speaker 3: back at the folder again, we have our colorphal maps.

31:34 Speaker 3: So if you have any questions about the use of

31:38 Speaker 3: the stream website or the API or both, please go

31:43 Speaker 3: ahead and leave any questions that you have in the

31:46 Speaker 3: chat and I'll try to address as many as I can.

31:48 Speaker 3: At the end, I would like to briefly acknowledge the

31:52 Speaker 3: various grants and teams that have supported the continued development

31:56 Speaker 3: of Stream. I've also provided a QR code here for

32:01 Speaker 3: you to try Stream yourself. Going back to our example

32:06 Speaker 3: of Pyramid Lake, a MEETA will now show you how

32:09 Speaker 3: to use the products that we downloaded through the Stream

32:12 Speaker 3: API in order to create a time series.

32:19 Speaker 1: Thank you so much William for the excellent presentation and

32:23 Speaker 1: very informative demonstration of using stream webtool and API both

32:28 Speaker 1: and next moment to continue using stream to make time

32:33 Speaker 1: series of water quality parameters. So we'll download several images

32:38 Speaker 1: using stream API and use qgis.

32:41 Speaker 2: To make time series.

32:43 Speaker 1: We'll pick locations Pyramid Lake and also Chesapeake Bay. In

32:48 Speaker 1: Pyramid Lake, will follow what William showed and we'll have

32:52 Speaker 1: a few images to look at time series of chlorophyll

32:54 Speaker 1: a concentration and then pick a small region in Chisapeake

32:59 Speaker 1: Bay and look at total suspended sediment time series in

33:03 Speaker 1: the bay. So next time I will share my screen

33:10 Speaker 1: and show this demo. So first of all, I'm using

33:16 Speaker 1: the same script that William showed for Pyramid Lake. And

33:22 Speaker 1: as you can see, this is the same tile eleven

33:24 Speaker 1: Tkee from Sentinel two that covers Pyramid Lake. So choosen that,

33:30 Speaker 1: and I'm just choosing a few days so that we

33:33 Speaker 1: can make time series. To see the procedure here October

33:37 Speaker 1: twenty twenty five from fifth through twenty fifth, so just

33:40 Speaker 1: twenty days, and we'll see how many images we have,

33:43 Speaker 1: and then I picked LA as William showed, let's also

33:50 Speaker 1: look at what I chose in Chi's a big bay.

33:53 Speaker 1: So for that I went to the stream web tool

33:58 Speaker 1: and then entered Pig Bay. Here now you can see

34:02 Speaker 1: sentinel two grids, and this is the grid we want

34:06 Speaker 1: to focus on. It's eighteen suh. Here is where Puto

34:10 Speaker 1: mc river goes into the bay and brings a lot

34:13 Speaker 1: of sediment sometimes. And so we've chosen this as a

34:17 Speaker 1: case study that will look at this style, not to

34:20 Speaker 1: look at time series of total suspended sediments.

34:28 Speaker 2: So now I.

34:30 Speaker 1: Have this copy of this script for Chiesapeake Bay in

34:34 Speaker 1: which the tile is eighteen suh. It's a longer time

34:38 Speaker 1: period May twenty four to end of twenty twenty five.

34:43 Speaker 1: And I've picked product lists as TSS and when these

34:48 Speaker 1: both these scripts I executed just as how William demonstrated.

34:54 Speaker 1: I have a number of files in streamline downloads. These

35:01 Speaker 1: are tip files. I've saved files for Pyramid Lake in

35:05 Speaker 1: a directory and all these R files for Pchicipeke Bay.

35:10 Speaker 1: So now we will move on to qgis and see

35:14 Speaker 1: how to make time series. I have downloaded and installed

35:20 Speaker 1: qgs on my computer and once you have it, you

35:24 Speaker 1: can click on this qgis I can and it will

35:28 Speaker 1: open an empty project. Now, just a note here in

35:33 Speaker 1: the appendix of today's presentation there is a procedure and

35:37 Speaker 1: there is a link on how to download qgis and

35:41 Speaker 1: how to install qus, as well as different plugins for

35:45 Speaker 1: doing calculations that has been described in the appendix, so

35:50 Speaker 1: you can refer to that. Here we have opened this

35:55 Speaker 1: new project. First we'll start with go to web and

36:00 Speaker 1: a map through quick map services. This is open streetmap

36:05 Speaker 1: and we just load the standard streetmap so we get

36:10 Speaker 1: the geographical regions here. Now we can add data to this.

36:14 Speaker 1: So I'm going to start with Pyramid Lake and we

36:18 Speaker 1: have raster layers the data that we saved earlier, So

36:23 Speaker 1: say add raster layer and navigate to the directory where

36:28 Speaker 1: the data are. So this is streamed downloads and in

36:32 Speaker 1: Pyramid Lake directory, I have data here. I have four

36:38 Speaker 1: days and I'm going to load pick this to add

36:43 Speaker 1: to the raster and what I'm going to show here

36:49 Speaker 1: is how to let's let's close this and you will

36:54 Speaker 1: see that you can click here and go zoom to

36:58 Speaker 1: the layer. So now you can see Pyramid Lake. Here

37:03 Speaker 1: you can change colors by going to properties and symbology

37:11 Speaker 1: big single band pseudo color. This is maximum, minimum and

37:16 Speaker 1: maximum value, and you can change that to a reasonable range.

37:21 Speaker 2: So I'm just saying one to.

37:25 Speaker 1: See twenty five milligrams per meter cube, and I'm going

37:30 Speaker 1: to have instead of continuous, have equal interval, and I'm

37:37 Speaker 1: going to increase number of intervals here, label precision to

37:45 Speaker 1: two and inward color maps. So when there is red

37:51 Speaker 1: its higher values of chlorophyll. Once you do that, you

37:55 Speaker 1: can say okay, and you will see. This is for

37:58 Speaker 1: seventeenth the image that Williams shoot and this is what

38:03 Speaker 1: you will see. After symbology you can see this. One

38:08 Speaker 1: more thing I want to show before I add rest

38:11 Speaker 1: of the rashers and show you how to make time

38:13 Speaker 1: series is that if you want to find time series

38:19 Speaker 1: at one point, you will have to pick one point.

38:22 Speaker 1: If you want entire lake, you will have to pick

38:25 Speaker 1: entire lake by adding a vector shape file here. What

38:29 Speaker 1: I'm going to do is suppose you want to pick

38:32 Speaker 1: a site for either water atrol or for setting up

38:37 Speaker 1: an a culture site, or for fishing or recreation, and

38:41 Speaker 1: you want to know a particular area how chlorophill a

38:45 Speaker 1: consultation varies over that region.

38:48 Speaker 2: With time.

38:49 Speaker 1: You can add a shape file. So here we're going

38:53 Speaker 1: to go and say create layer. Add new shape file layer.

38:58 Speaker 1: Here you can give.

39:00 Speaker 2: A name, uh, say lake and save.

39:07 Speaker 1: Go to geometry type add polygon because you will be

39:10 Speaker 1: picking an area. And now you can just say okay,

39:15 Speaker 1: and now to add that polygon here, you will click

39:20 Speaker 1: on this big click on this touggle edit and polygon

39:24 Speaker 1: symbol and then start by clicking and adding a polygon.

39:32 Speaker 1: When you want to add the polygon, you can say

39:35 Speaker 1: control right click and you can give it ID, say

39:40 Speaker 1: number one, and say okay. Now you have picked a

39:43 Speaker 1: region over which you want to find time series for

39:46 Speaker 1: a number of days. Now, just to save some time,

39:52 Speaker 1: I've already loaded all the rosters in Project Pyramid time

39:58 Speaker 1: Series and we'll work with this. So we have chlorophyll

40:05 Speaker 1: A rasters for tenth, seventeenth, twenty eighth, and twenty fourth

40:09 Speaker 1: of October twenty twenty five.

40:11 Speaker 2: I've also added.

40:12 Speaker 1: A polygon where we will have time series values and

40:17 Speaker 1: I've named it region of Interest on ROI. So once

40:22 Speaker 1: let's look at the here. This is seventeenth, this is twentieth,

40:28 Speaker 1: and this is twenty fourth October, so you can see

40:31 Speaker 1: how it algal bloom started and it increased and then

40:37 Speaker 1: it went up and then bent down by down again.

40:40 Speaker 1: So we see that cycle here. Now, if you want

40:44 Speaker 1: to find time series of area averaged chlorophyll, you will

40:49 Speaker 1: want to use raster analysis and zonal statistics to find

40:55 Speaker 1: mean value for this polygon. Now, if you use this

41:02 Speaker 1: Zono statistics plug in, you will have to or tool.

41:06 Speaker 1: You will have to do it raster by raster. Here

41:09 Speaker 1: you we only have four rasters, but if you have many,

41:13 Speaker 1: then it's not very convenient to do one by one.

41:16 Speaker 1: So there is another raster. It is also raster analysis

41:23 Speaker 1: Zono statistics. But this is for multiband and this is

41:27 Speaker 1: this comes from an experimental plug in from Dimexian Labs.

41:32 Speaker 1: So you have to install this into your QGIS and

41:37 Speaker 1: procedure to do that is also provided in the appendix.

41:41 Speaker 1: There is a link where you can go and install

41:44 Speaker 1: this experimental plug in Dimexian Lab. So we're going to

41:50 Speaker 1: use this Sono statistics multi band to create time series

41:55 Speaker 1: area average value for this polygon. To do that, the

42:00 Speaker 1: first step is that you create a virtual raster. So

42:03 Speaker 1: you go to raster miscellaneous and click on build virtual roster. Here.

42:09 Speaker 1: What you're going to do is add all your rasters

42:12 Speaker 1: and make one raster in which all the images appear

42:17 Speaker 1: as different bands. So here you can pick your input

42:24 Speaker 1: rosters and you can arrange them by dates and say, okay,

42:33 Speaker 1: you want to place each input file into a separate band.

42:39 Speaker 1: And important thing is that there is no data value

42:42 Speaker 1: that you want to add here, so that when you

42:45 Speaker 1: calculate statistics, those points are.

42:50 Speaker 2: Excluded.

42:51 Speaker 1: And how do you find that. You can click on

42:56 Speaker 1: this layer and go to properties. Sometimes it's given in

43:01 Speaker 1: the information tab, or you here you go to transparency.

43:07 Speaker 1: No data value is minus three two seven six seven,

43:10 Speaker 1: So you can find out what the value is and

43:13 Speaker 1: then you can add that to this virtual raster formation.

43:20 Speaker 1: Then you can run this. And when you run this,

43:24 Speaker 1: it creates a virtual a aster which is already created here.

43:29 Speaker 1: If you click on this virtual a master and look

43:33 Speaker 1: at properties and go to information, you will see that

43:37 Speaker 1: all four images now they're stored in here as bands,

43:42 Speaker 1: different bands, and that will help us in finding statistics

43:47 Speaker 1: in just one go by using multi band zonal statistics.

43:54 Speaker 1: So now once we have that, we can go to

43:58 Speaker 1: the plug in sooner statistics here. This is the astra layer.

44:05 Speaker 1: Virtual layer would be an aster layer, and pick your

44:08 Speaker 1: region of interest polygon for which you want to find

44:12 Speaker 1: mean value and you can say ch L would be

44:19 Speaker 1: a column prefix. The statistics will be stored in attributeable

44:25 Speaker 1: for this polygon and it will have prefix CHL. You

44:33 Speaker 1: can select statistical quantities you want to make time series

44:37 Speaker 1: off here. Accounts is the number of points of pixels

44:42 Speaker 1: within that polygon that we pick. This is some of

44:46 Speaker 1: values mean, mediums and a deviation, et cetera. Let's fix

44:51 Speaker 1: and a deviation and maximum value. These three quantities for chlorophyll,

44:56 Speaker 1: I can say okay, and then you can click here

45:00 Speaker 1: to run this process and at the end you will

45:04 Speaker 1: get statistical quantities for all the days or all the

45:08 Speaker 1: bands in this case in the attribute table that is

45:12 Speaker 1: associated with this shape file. So you can right click

45:16 Speaker 1: and open attribute table. You will see that this is

45:20 Speaker 1: the prefix that we picked. Band one is day one,

45:23 Speaker 1: Day two, three and four means on aviation and mixing.

45:27 Speaker 1: These values are available here. You can save this file

45:32 Speaker 1: as I can click right click on this and export

45:36 Speaker 1: as saved features and then it's comma separated value or

45:41 Speaker 1: a CSV file you can save to the folder of

45:46 Speaker 1: your choice.

45:47 Speaker 2: I've already done.

45:48 Speaker 1: That and save this file as p l ts dot

45:54 Speaker 1: c s V. So these are the numbers that we

45:57 Speaker 1: got for four days, means, deviation and maximum value. I've

46:03 Speaker 1: just rearranged them in two columns so that it's easier

46:06 Speaker 1: to read. These are all the numbers, and you can

46:09 Speaker 1: separate means, deviation and maximum values in separate columns and

46:13 Speaker 1: then plot. But for simplicity, I've just plotted all of

46:19 Speaker 1: them here. Just use insert and then chart, and then

46:24 Speaker 1: you can say bar or here. You can choose from

46:27 Speaker 1: here and you will get a plot here. So what

46:31 Speaker 1: you see here is day one means and deviation and

46:35 Speaker 1: maximum value. This is for day two, day three, and

46:39 Speaker 1: day four. You can see day two and three. So

46:41 Speaker 1: seventeenth and twenty October quite a lot high intensity of

46:46 Speaker 1: fluorophila concentration about three hundred milligrams per meter cube. So

46:50 Speaker 1: this is a quick way to get data from stream,

46:55 Speaker 1: get them in qgis, and make time series. I take

46:59 Speaker 1: the same for Chesapeake Bay suspended solids. So I want

47:06 Speaker 1: to show that to you here. Procedure is the same.

47:18 Speaker 1: I have all a few rasters for May to October

47:24 Speaker 1: of twenty twenty four. Here I made virtual roster by

47:29 Speaker 1: adding all these days as bands in this raster and

47:34 Speaker 1: have a polygon that's sitting here at the where there's

47:38 Speaker 1: potomac coming into the bay. One thing to see here

47:42 Speaker 1: is if you click on different rastas, you will see

47:46 Speaker 1: that this is TSS and this is in grams per

47:53 Speaker 1: meter cube, so it's a scale is the same for everything.

47:58 Speaker 1: But in some cases you will see good coverage. Sometimes

48:03 Speaker 1: you don't see Some areas do not have data, and

48:10 Speaker 1: that is there are two reasons either maybe there is

48:12 Speaker 1: no overpass here now satellite data or sometimes there are

48:16 Speaker 1: clouds and you cannot see the surface anyway. But this

48:20 Speaker 1: is the area for which where I created virtual raster,

48:25 Speaker 1: used solo stakes statistics multi band and created.

48:33 Speaker 2: Time series for that as well. So the time series

48:39 Speaker 2: is here.

48:42 Speaker 1: Now I here I have mean values only of TSS,

48:48 Speaker 1: and this is for different days. You can have dates

48:54 Speaker 1: here and then you can make a proper time series.

48:57 Speaker 1: But this shows how it mean values of TSS varied

49:03 Speaker 1: in that polygon into the big bay with time. So

49:08 Speaker 1: this is a easier way to make time series from

49:14 Speaker 1: stream data and hope you can find it useful and

49:18 Speaker 1: you can explore this on your own as well. So

49:22 Speaker 1: that ends our demonstration for time series. So this brings

49:29 Speaker 1: us to the end of today's session, and just to summarize,

49:34 Speaker 1: we saw background overview and demonstration of Stream webtool and API.

49:40 Speaker 1: We saw that stream is based on lands set eight

49:43 Speaker 1: and nine and Sentinel two, A, B and C optical

49:46 Speaker 1: measurements to obtain water quality parameters including chlorophyl a concentration,

49:52 Speaker 1: total suspended solids, and secy disk depth in coastal estuaries

49:56 Speaker 1: and inland lakes in the US. It provides it's the

50:00 Speaker 1: water quality parameters at twenty to thirty meters special resolution,

50:04 Speaker 1: and the API allows search and download of multiple images

50:09 Speaker 1: of an area of interest. Then we saw a couple

50:12 Speaker 1: of examples of how to select water quality parameters in

50:16 Speaker 1: areas of interest using stream. We focused on Pyramid Lake

50:21 Speaker 1: and then also on a portion of Chesapeake Bay. William

50:25 Speaker 1: showed how to map the water quality parameters and how

50:29 Speaker 1: to access and download multiple water quality data using the

50:32 Speaker 1: Stream API, and then we saw how to make time

50:36 Speaker 1: series of area averaged water quality parameters using qgis so

50:41 Speaker 1: clear advantage of using Stream can be seen here. These

50:45 Speaker 1: two figure show chlorophyl a concentration in Pyramid Lake from

50:49 Speaker 1: a pasoci at one kilometer and sentinel to MSI from

50:53 Speaker 1: stream at twenty meters.

50:55 Speaker 2: So here.

50:56 Speaker 1: If your application requires with all of water for drinking

51:01 Speaker 1: utilities or for aquaculture site selection, you want to use

51:05 Speaker 1: high resolution data because it shows a lot of details

51:10 Speaker 1: and special variation of water quality parameters. Now next week,

51:17 Speaker 1: on seventeenth of February, we will have an introduction to

51:21 Speaker 1: a machine learning model to estimate water quality parameters based

51:24 Speaker 1: on satellite observations. This is the model that's been used

51:28 Speaker 1: to produce stream parameters. As I mentioned earlier, there will

51:34 Speaker 1: be one homework assignment posted on seventeenth of February on

51:38 Speaker 1: the training website and answers must be submitted via Google forms.

51:43 Speaker 1: The homework will be due on tenth of March. Certificate

51:47 Speaker 1: of completion will be awarded to those who attend all

51:51 Speaker 1: live webinars and complete the homework assignment by the deadline.

51:56 Speaker 1: You will receive a certificate by email approximately two months

51:59 Speaker 1: after completion of the course. Once again we thank William

52:05 Speaker 1: Wainwright for his excellent presentation and demonstration of stream and

52:10 Speaker 1: making the webtool and API available to the community. Here

52:16 Speaker 1: are some resources and links to important websites, and acronyms

52:22 Speaker 1: are defined for the satellite and censor table. This is

52:28 Speaker 1: the contact information from William Wainwright and you can always

52:32 Speaker 1: contact us at our set, our set website link and

52:36 Speaker 1: YouTube link they're provided here. For questions, comments, or to

52:40 Speaker 1: share how you have applied our trainings to your work

52:43 Speaker 1: or studies. Please email at NASA rset at gmail dot

52:48 Speaker 1: com and join our quarterly newsletter to stay up to

52:52 Speaker 1: date on our latest trainings. You can send an email

52:57 Speaker 1: here with no subject line and follow the instructions sent

53:01 Speaker 1: in response, and then you can have access to the newsletter.

53:12 Speaker 1: Thank you everyone. We'll start with our question and answer session.

53:17 Speaker 1: We have William wain right here and we'll start with

53:21 Speaker 1: the questions. Question one, is it possible to replicate this

53:26 Speaker 1: on another continent? Also? What processing scale is available?

53:31 Speaker 4: Okay?

53:32 Speaker 3: So I'll do my best to go through and answer

53:35 Speaker 3: as many questions as we saw. If by the end

53:38 Speaker 3: of it you feel like your question hasn't been answered,

53:42 Speaker 3: please ask again or ask what more details? And we

53:47 Speaker 3: have a Q and a document that will be sending

53:49 Speaker 3: out after all the trainings have been completed that will

53:53 Speaker 3: hopefully answer everything. So to get back to question one,

53:58 Speaker 3: is it possible to replicate this on another continent? It

54:02 Speaker 3: definitely is. The models that are behind stream. The aim

54:06 Speaker 3: in developing those models are to have a globally applicable

54:12 Speaker 3: model to any inland and coastal water body. They obviously

54:17 Speaker 3: do have varying performance in different regions, and that will

54:21 Speaker 3: depend some on the amount of insitu samples that we've

54:24 Speaker 3: been able to acquire and train on from that region.

54:28 Speaker 3: So if you want to improve the accuracy of STREAMS models,

54:33 Speaker 3: you can always try to help provide the Freshwater Sensing

54:37 Speaker 3: group with more institute data that we can train on

54:40 Speaker 3: and we can continually improve those models. Question two, What

54:46 Speaker 3: countries are covered by stream?

54:49 Speaker 4: Right now?

54:49 Speaker 3: Stream covers the continental US or Konis, Hawaii, Alaska, and

54:55 Speaker 3: then there's lots of many select water bodies and satellite

54:58 Speaker 3: tiles around the world, including but not limited to, areas

55:04 Speaker 3: in India, South Korea, South Africa, Benin, Ghana, Chile, Uruguay, Peru, Colombia, Mexico,

55:14 Speaker 3: and Cuba. A lot of these select tiles were born

55:22 Speaker 3: from partnerships with other research teams, but We're hoping to

55:26 Speaker 3: eventually be able to cover as much of the world

55:28 Speaker 3: as we can.

55:30 Speaker 2: Thank you, William.

55:32 Speaker 1: The next question is not directly relevant to this topic today.

55:37 Speaker 1: Is there any way to monitor groundwater data?

55:40 Speaker 2: It's TSAR.

55:41 Speaker 1: Is there any way to monitor seawater and groundwater intrusion

55:45 Speaker 1: rates in coastal areas like Sundarbans. Currently incitude data is

55:50 Speaker 1: being collected in very limited areas, but due to the

55:54 Speaker 1: complexity of the region, it's not very effective. I just

55:57 Speaker 1: want to point out here is that ARSET will have

56:00 Speaker 1: a training on groundwater in April, which you may want

56:05 Speaker 1: to check out and join, where we can editor some

56:08 Speaker 1: of these questions. This training is mostly focused on stream

56:13 Speaker 1: and water quality, so right now when we are not

56:16 Speaker 1: addressing this question, but we'll definitely think of it when

56:20 Speaker 1: we do their ground.

56:21 Speaker 2: Or to training.

56:23 Speaker 1: Next question is for William. Does the tool work well

56:26 Speaker 1: with touch screen based devices like iPads and so on?

56:32 Speaker 3: Unfortunately, stream does not work very well right now on

56:35 Speaker 3: mobile devices unless you enable desktop mode. It should work

56:40 Speaker 3: fine with touch screens on desktop mode, but the mobile.

56:43 Speaker 4: Display of the page is not great.

56:45 Speaker 3: That's something I'm working on improving pretty soon hopefully. But yeah,

56:51 Speaker 3: you'll find it works bested on desktops.

56:56 Speaker 2: Great.

56:57 Speaker 1: Can move on to next question as well, what is

56:59 Speaker 1: the special cover your products understream?

57:03 Speaker 3: So Landset eight and nine derived products retain the native

57:09 Speaker 3: spatial resolution of thirty meters. Sentinel two derived products the

57:14 Speaker 3: top of atmosphere level one data comes into between ten

57:17 Speaker 3: and sixty meter resolution. We resample everything to twenty meters,

57:23 Speaker 3: and then we also provide the water quality products at

57:25 Speaker 3: that twenty meter resolution.

57:29 Speaker 4: And that kind of ties into the.

57:33 Speaker 3: Another question of why not use kind of a fusion

57:36 Speaker 3: between those two satellites. If you're referring to like the

57:41 Speaker 3: harmonized Lancet Sentinel products, that unfortunately doesn't include all of

57:46 Speaker 3: the original bands from both sensors, and one of the

57:50 Speaker 3: bands that's missing from that harmonized Sentinel Landset product is

57:54 Speaker 3: actually pretty key in how we derive water quality, right.

58:02 Speaker 1: So I think that also takes care of question six.

58:05 Speaker 1: If I select a Sentinel two file, does that mean

58:08 Speaker 1: the data for that tile is estimated from Sentinel two?

58:12 Speaker 2: And so why not use fusion? So I think you

58:15 Speaker 2: just answered that question.

58:17 Speaker 3: If you select the tile, you will get results just

58:21 Speaker 3: for that satellite. There are other ways and We're in

58:26 Speaker 3: the future looking at something called domain adaptation, which will

58:31 Speaker 3: let us kind of harmonize the results ourselves from different satellites.

58:37 Speaker 4: But that's something that's still.

58:38 Speaker 3: Very much experimental and not ready for the deployment on stream.

58:45 Speaker 1: Also, I just wanted to add to question five when

58:48 Speaker 1: you talk about special coverage, these data are global lends

58:51 Speaker 1: AT eight and Sentinel two. They both have global coverage

58:55 Speaker 1: right now. What quality parameters are produced for the US

58:59 Speaker 1: and for a few selectric ties as William mentioned, but

59:02 Speaker 1: the model that will be presented next week that can

59:05 Speaker 1: be used to get aut quality parameters anywhere where you

59:10 Speaker 1: can see resolve lakes and coastal areas exactly.

59:16 Speaker 3: And next week's training with Ryan O'Shea will talk more

59:19 Speaker 3: about how to apply the models that we use to

59:23 Speaker 3: any tile by not just LANDSTT and Sentinel, but also Pace,

59:30 Speaker 3: Sentinel three and many more. So if there's a satellite

59:34 Speaker 3: that you want covered or a region that you want

59:37 Speaker 3: covered that isn't currently supported on stream, you almost certainly

59:42 Speaker 3: will be able to do it with the models, and

59:44 Speaker 3: you'll learn how to do that next week.

59:48 Speaker 1: Yes, next question is is base data on stream web.

59:54 Speaker 3: Pace is not currently supported on stream. But as I

59:59 Speaker 3: was just saying, the models that.

1:00:04 Speaker 4: Stream uses do support PACE, so we do have water

1:00:10 Speaker 4: quality data four PACE, it's just not currently on stream.

1:00:15 Speaker 2: Yeah.

1:00:15 Speaker 1: The next question is does it extract data from Sentinel

1:00:19 Speaker 1: to or LENST as soon as they are available, or

1:00:22 Speaker 1: are the water quality data uploaded to stream at set

1:00:25 Speaker 1: period when a considerable amount of products are generated.

1:00:31 Speaker 3: We automatically download the scenes in bulk from Copernicus for

1:00:36 Speaker 3: Sentinel two and USGS for LANSA eight and nine pretty

1:00:42 Speaker 3: much as soon as they become available at level one. Now,

1:00:46 Speaker 3: typically they become available at the end of the day

1:00:49 Speaker 3: that they were observed, and given that we're processing between

1:00:53 Speaker 3: six hundred and eight hundred tiles per day, it can

1:00:58 Speaker 3: sometimes take well into the next day to finish downloading

1:01:02 Speaker 3: and processing all you know, eight hundred of those tiles.

1:01:06 Speaker 3: We don't have any specific order that we process them.

1:01:10 Speaker 3: We just asked for data. It becomes available to us

1:01:14 Speaker 3: from the data providers, and then we process it as

1:01:17 Speaker 3: we receive it.

1:01:20 Speaker 2: Great.

1:01:21 Speaker 1: The next question also, you can at risk can I

1:01:23 Speaker 1: collect carring you previous data in the Gulf of America.

1:01:30 Speaker 3: Coastal regions of the Gulf are covered natively natively by stream.

1:01:36 Speaker 3: If you find that your site is not covered, you

1:01:39 Speaker 3: can either apply the models yourself or reach out to

1:01:43 Speaker 3: us about potentially including your region of interest.

1:01:49 Speaker 1: Question tense, stream gives model estimates, how do we get

1:01:53 Speaker 1: these standard errors of these those estimates? How do we

1:01:57 Speaker 1: get these standard errors for the model est mates of

1:02:00 Speaker 1: chlorophyl TSS and SICHI depth.

1:02:04 Speaker 3: Yeah, we're working to add uncertainty as an additional product

1:02:10 Speaker 3: to the archive as far as how we derive that

1:02:14 Speaker 3: air and kind of the training and validation in general.

1:02:18 Speaker 3: That's something that'll be covered a lot more in part

1:02:20 Speaker 3: two of the training next week.

1:02:23 Speaker 4: But you're right that we don't We don't provide.

1:02:26 Speaker 3: Air maps on stream currently, but we would like to

1:02:32 Speaker 3: soon provide uncertainty in the archive page at least.

1:02:37 Speaker 1: Great question eleven, is there any way that I can

1:02:41 Speaker 1: improve the temporal resolution of the data? I mean, if

1:02:44 Speaker 1: I need daily data, how can I get it?

1:02:47 Speaker 4: I'm gonna I'm going to jump back to question kind of.

1:02:49 Speaker 4: I thought of something else as well.

1:02:51 Speaker 3: In the case of training and validation, I think the

1:02:54 Speaker 3: standard air means a little bit more because we're comparing

1:02:59 Speaker 3: against institute, so we actually have a known.

1:03:04 Speaker 4: Thing to compare to.

1:03:06 Speaker 3: When that model is then applied to near real time data,

1:03:10 Speaker 3: we don't have it suits you validation. So the uncertainty

1:03:16 Speaker 3: from the daily derived products on stream is more of

1:03:20 Speaker 3: just the model's best guess at how accurate it thinks

1:03:24 Speaker 3: it is. So it's it's still valuable because it gives

1:03:28 Speaker 3: you a measurement of effectively how close the water quality

1:03:36 Speaker 3: parameters that you're looking at match up with something it

1:03:40 Speaker 3: was trained on. So if you have a high uncertainty

1:03:43 Speaker 3: from streams daily processing, that will mean that we're probably

1:03:48 Speaker 3: giving you water quality estimates for a water body that

1:03:50 Speaker 3: doesn't look very much like anything we've seen in training.

1:03:54 Speaker 3: So that's what an uncertainty map for the daily processing

1:03:59 Speaker 3: would mean.

1:04:02 Speaker 2: Okay, can go to the next question as well.

1:04:05 Speaker 4: Ye, so is there any way I can improve?

1:04:12 Speaker 3: So one of the trade offs of the increased special

1:04:15 Speaker 3: resolution of Landset and Sentinel two relative to the other

1:04:19 Speaker 3: water quality satellites available is that decreased temp resolution. In

1:04:26 Speaker 3: the future, we're gonna try to cover more satellites with

1:04:29 Speaker 3: an aim of increasing temporal resolution though.

1:04:34 Speaker 2: Great.

1:04:34 Speaker 1: Yes, so as we saw earlier, you know, other satellites

1:04:39 Speaker 1: like Taraqua SMPBGPS as base. They do have daily data available.

1:04:46 Speaker 1: The resolutions are not as high as sentinel two in

1:04:49 Speaker 1: lens a day, so there's there's always a trade off

1:04:52 Speaker 1: between temple.

1:04:53 Speaker 2: Special resolutions exactly.

1:04:57 Speaker 1: Next question is is this calibrated? How acurator of the results?

1:05:01 Speaker 1: Is there any way to calibrate the data.

1:05:04 Speaker 3: So the models behind Stream are pretty heavily calibrated and

1:05:10 Speaker 3: validated using data sets like Gloria and aironet oc, but

1:05:15 Speaker 3: they're also supplemented with lots of individual institute data sets

1:05:20 Speaker 3: from research partners. The accuracy of the result can vary

1:05:26 Speaker 3: depending on how many samples that we've had to train

1:05:29 Speaker 3: against in that region. The best way to help calibrate

1:05:34 Speaker 3: data for Stream as a whole would be to help

1:05:36 Speaker 3: us expand our.

1:05:37 Speaker 4: Institute data set.

1:05:38 Speaker 3: If you find the performance is lacking in a region

1:05:41 Speaker 3: of interests, you may be able to help us improve

1:05:45 Speaker 3: the performance in that region by providing insitu data. You

1:05:50 Speaker 3: can also run the models locally to assess performance on

1:05:55 Speaker 3: an area that STREAM doesn't cover.

1:06:00 Speaker 2: Right.

1:06:01 Speaker 1: Next question is are these values of different parameters only

1:06:04 Speaker 1: for surface water or do they represent subsurface water in

1:06:08 Speaker 1: the water bodies.

1:06:12 Speaker 3: That's a question I don't fully know The answer to

1:06:16 Speaker 3: my best understanding, although this may be a better question

1:06:20 Speaker 3: for Ryan o'sh next week, is the water quality parameters

1:06:24 Speaker 3: are derived from the remote sensing reflectance at the water's surface,

1:06:29 Speaker 3: but the properties derived from that reflectance are assumed to

1:06:33 Speaker 3: be indicative of the water the water column as a whole.

1:06:38 Speaker 4: Again, now that's my understanding, but.

1:06:45 Speaker 3: Please if someone knows better in the chat or you know,

1:06:49 Speaker 3: next week, Ryan will probably be better be able to

1:06:51 Speaker 3: better answer that question as well.

1:06:54 Speaker 1: So yeah, we can talk to Ryan more. But these

1:06:57 Speaker 1: parameters are for surface water, as we are using satellite

1:07:01 Speaker 1: would look at top like water, leaving that that's what

1:07:06 Speaker 1: is used, which has information about water column. But the

1:07:13 Speaker 1: the values that you see are for surface water.

1:07:16 Speaker 4: Yeah.

1:07:17 Speaker 1: Question fourteen, what is the model evaluation for the model

1:07:21 Speaker 1: used for estimating chlorophyl A and TSS for SENTILEL and

1:07:25 Speaker 1: that of lendset. Is there a reference that can be

1:07:29 Speaker 1: cited for this information?

1:07:33 Speaker 3: Yeah, The name of the model and specific is aquaverse.

1:07:38 Speaker 3: We have a paper and I'll try to include a

1:07:41 Speaker 3: link to the Q inside the Q and a document. Yeah,

1:07:46 Speaker 3: but if you search for aquaverse you should see a

1:07:51 Speaker 3: paper about our models. And then also Ryan O'shay's GitHub

1:07:57 Speaker 3: has some of the the MDM tutorials.

1:08:04 Speaker 1: So next question is interesting, does stream work in seasonal

1:08:08 Speaker 1: bodies of water around the world?

1:08:13 Speaker 3: If I'm understanding the question correctly, are those water bodies

1:08:17 Speaker 3: where there's only water prevalent during certain seasons and then.

1:08:23 Speaker 4: They dry up?

1:08:24 Speaker 2: Or yes, I think that's what it means.

1:08:28 Speaker 3: Yeah, the way we identify water pixels to process is

1:08:34 Speaker 3: not exclusive of those types of water bodies, which means

1:08:40 Speaker 3: that if there's water in an image, it should be

1:08:45 Speaker 3: processed by the dream and report water quality. But yes,

1:08:50 Speaker 3: stream should work in those seasonal water bodies if you

1:08:52 Speaker 3: have examples. I've never personally looked into it, but I'd

1:08:56 Speaker 3: be really interested to see how those look.

1:09:01 Speaker 2: Great.

1:09:05 Speaker 1: So question sixteen, is it possible to have a quick

1:09:08 Speaker 1: look at the data used in the estimation of glorophile

1:09:11 Speaker 1: A and other presentations?

1:09:16 Speaker 4: So go ahead.

1:09:20 Speaker 3: The Gloria and air onet oc data sets, I believe

1:09:23 Speaker 3: are publicly available, but you can learn a bit more

1:09:29 Speaker 3: about some of the data and products used and the

1:09:34 Speaker 3: estimations and the other predictors.

1:09:36 Speaker 4: And next week's session just.

1:09:39 Speaker 1: To add that Arsett did a training in which Gloria

1:09:44 Speaker 1: observations or measurements were covered, and that link you will

1:09:48 Speaker 1: find on the slides where gloria data was mentioned, so

1:09:54 Speaker 1: you will find link to the training where you have

1:09:56 Speaker 1: more information about gloria data. Next question is can we

1:10:01 Speaker 1: get information about other water quality parameters? For example, if

1:10:05 Speaker 1: need data regarding nutrients, can I get that? So William,

1:10:11 Speaker 1: we can add to this. But to my knowledge, we

1:10:14 Speaker 1: are looking at optical properties of water. So if it

1:10:18 Speaker 1: changes color of the water and if these spectral bands

1:10:23 Speaker 1: can see that, those are the parameters they are derived.

1:10:28 Speaker 1: For nutrients like nitrogen or phosphorus, you'd probably have to

1:10:32 Speaker 1: have some satistical measure model to use that to connect

1:10:37 Speaker 1: institute data with satellite data and then indirectly derived nutrients

1:10:43 Speaker 1: based on that.

1:10:44 Speaker 2: I believe.

1:10:46 Speaker 3: Yeah, the limiting factors really does the parameter you're interested

1:10:50 Speaker 3: in change the optical properties of the water. There are

1:10:54 Speaker 3: some parameters that are not on stream that can be derived,

1:10:58 Speaker 3: like color, dissolved organic.

1:10:59 Speaker 4: Matter re set on.

1:11:01 Speaker 3: That's something that the models can produce, but we don't

1:11:04 Speaker 3: currently offer on stream. But there are many properties that

1:11:08 Speaker 3: people are interested in, like you know, dissolved phosphorus or

1:11:12 Speaker 3: oxygen that can't really be measured from space because they

1:11:17 Speaker 3: don't really measurably change the optical properties of the water. Yeah.

1:11:24 Speaker 1: The next question, please clarify the data range is available

1:11:27 Speaker 1: for these analysis. What is the earliest data available to analyze?

1:11:32 Speaker 1: So online web tool will have data from twenty twenty

1:11:37 Speaker 1: four onwards, correct William.

1:11:40 Speaker 3: Yeah, so stream began daily processing, so downloading data the

1:11:45 Speaker 3: day it became available and then automatically publishing it in

1:11:49 Speaker 3: June of twenty twenty four. You made these themes from

1:11:52 Speaker 3: before then. Most of those were manually processed with collaborators,

1:11:59 Speaker 3: you know, wanted a time series for a specific tile

1:12:02 Speaker 3: or two from twenty eighteen onward, so you'll see individual

1:12:07 Speaker 3: tiles covered across many years before twenty twenty four.

1:12:12 Speaker 4: We're trying to fill in.

1:12:15 Speaker 3: Coverage that began after June twenty twenty four, at least

1:12:19 Speaker 3: back until June of twenty twenty four, to make everything synchronous.

1:12:23 Speaker 3: But right now, full CONUS coverage for Sentinel began at

1:12:29 Speaker 3: the start of twenty twenty five, and we've recently fixed

1:12:34 Speaker 3: our landset downloading, so full CONUS for Landset began.

1:12:39 Speaker 4: Twenty twenty six.

1:12:41 Speaker 3: Oh, okay, select regions for Lancet have been available since

1:12:47 Speaker 3: June of twenty twenty four though, and well I think

1:12:51 Speaker 3: I said full conis for Sentinel has been available since

1:12:54 Speaker 3: the start of twenty twenty five, So right now the

1:12:58 Speaker 3: coverage can dep a bit on what tile you're looking at,

1:13:02 Speaker 3: but that should be permanized pretty soon. Everything should go

1:13:07 Speaker 3: back to at least June of twenty twenty four.

1:13:11 Speaker 1: Great next question is when analyzing tts from satellite data,

1:13:16 Speaker 1: do we focus on the highest value the lowest value

1:13:20 Speaker 1: are a different metric like the average? I would say

1:13:27 Speaker 1: that average, so range and average they both should be

1:13:33 Speaker 1: looked at. But you know we can get more information

1:13:37 Speaker 1: about it. Well, we have a couple of more questions

1:13:45 Speaker 1: million if you can address we have a few minutes. Also,

1:13:49 Speaker 1: does the model only use land set and sentinel or

1:13:54 Speaker 1: there are other hydrological and minological parameters used.

1:14:00 Speaker 4: There are some additional parameters, namely in speed. There's also

1:14:10 Speaker 4: the sensor.

1:14:10 Speaker 3: Angle, but that's still kind of the part of the

1:14:13 Speaker 3: satellite data.

1:14:14 Speaker 4: We have wind speed, water vapor, and there's one more.

1:14:19 Speaker 3: Oh ozone and NO two concentration are also used to

1:14:24 Speaker 3: do the atmosphere correction. The aerosol contribution is taken care

1:14:29 Speaker 3: of by the MDN, but the rest of it, the

1:14:33 Speaker 3: really correction and the gashous absorption, we need the aforementioned parameters.

1:14:41 Speaker 1: Correct question twenty two. Is it possible to use HIMA?

1:14:44 Speaker 1: What did you stationary satellite with stream?

1:14:51 Speaker 4: It's not currently possible to use other satellites with STREAM.

1:14:55 Speaker 3: I would have to check if that satellite is supported

1:15:00 Speaker 3: by the models, namely aquiverse, and I think we skipped

1:15:11 Speaker 3: over twenty If we apply STREAM to a region like

1:15:14 Speaker 3: Jordan that is not explicitly listed, what type of local

1:15:18 Speaker 3: data would most improve model calibration and confidence in the results.

1:15:24 Speaker 3: Any in situ chlorophyl A, tss SECHI, disk depth and

1:15:30 Speaker 3: SETAM measurements would help. The ones that are most likely

1:15:35 Speaker 3: to help the most are institue measurements that coincide pretty

1:15:41 Speaker 3: closely with satellite observations in terms of time, But any

1:15:48 Speaker 3: institu data will help.

1:15:54 Speaker 1: Going down to question twenty three for the nutrient level analysis,

1:15:59 Speaker 1: since these are not optically these are not optically inherent

1:16:03 Speaker 1: properties of water, indirect methods such as empirical method. Emial

1:16:07 Speaker 1: method is the only way.

1:16:13 Speaker 4: If you're able to.

1:16:16 Speaker 3: Like empirically show that it correlates to something that is optically.

1:16:25 Speaker 4: Adherent to the water.

1:16:27 Speaker 3: So, in other words, if you're able to find an

1:16:29 Speaker 3: optical property that's a good tracer for the property you're

1:16:32 Speaker 3: actually interested in, then that's one way to do it.

1:16:37 Speaker 3: If you're not able to demonstrate that something like chlorophyla

1:16:41 Speaker 3: or TSS is a good tracer for your nutrient analysis,

1:16:45 Speaker 3: then to my understanding, you would have to find another

1:16:50 Speaker 3: way to measure it.

1:16:55 Speaker 1: Next question is how soon would the rest of the

1:16:58 Speaker 1: world be covered with this small especially North Africa region.

1:17:04 Speaker 3: We're able to add individual tiles pretty readily, but if

1:17:12 Speaker 3: you're talking about full global, inland and coastal coverage, that's

1:17:20 Speaker 3: an order of magnitude more processing than we already do,

1:17:23 Speaker 3: and we don't have the processing and storage available to

1:17:27 Speaker 3: do that yet.

1:17:30 Speaker 1: So that is one of the reasons why the model

1:17:33 Speaker 1: is open source and made available to anyone who wants

1:17:36 Speaker 1: to apply it in their own.

1:17:39 Speaker 3: Leg and our hope is that the model will at

1:17:44 Speaker 3: least be relevant in globally, which is why we continue

1:17:50 Speaker 3: to retrain and revalidate it with each new institute sample

1:17:55 Speaker 3: that we get. But as far as stream, I think

1:18:01 Speaker 3: global coverage is less of a priority than increasing temporal

1:18:07 Speaker 3: resolution currently for us.

1:18:10 Speaker 2: Right, thanks.

1:18:12 Speaker 1: Next question is can we use stream for monitoring water

1:18:14 Speaker 1: quality in reverse?

1:18:17 Speaker 3: If landst thirty meter resolution or Sentinel at twenty meter

1:18:23 Speaker 3: resolution can resolve your river or other water body, then yeah.

1:18:32 Speaker 1: The next one does the bacteria account like ecoli and

1:18:35 Speaker 1: fecal quliform also be found out through stream?

1:18:43 Speaker 3: I think that's kind of similar to the question about nutrients.

1:18:49 Speaker 3: If you can trace it with a water quality parameter

1:18:54 Speaker 3: like ESS or chlorophyla, then potentially, but it's not something

1:19:02 Speaker 3: we report on directly.

1:19:06 Speaker 1: Next question is what are the top five features for

1:19:09 Speaker 1: the model from the feature importance analysis? Are they the

1:19:13 Speaker 1: same for Lancet and Sentinel.

1:19:17 Speaker 4: Top five features for the model?

1:19:22 Speaker 1: We can address it next week as well.

1:19:24 Speaker 4: Yeah, I'm not sure, so I'll refer that to Ryan probably.

1:19:35 Speaker 1: Next question is is there any training for applying the

1:19:40 Speaker 1: model for individual location?

1:19:43 Speaker 4: There is.

1:19:44 Speaker 3: Next week's training with Ryan O'Shea will show you how

1:19:47 Speaker 3: to run model yourself on a tile of interest.

1:19:57 Speaker 1: Twenty Question twenty nine. When Stream is applying in regions

1:20:01 Speaker 1: without extensive institute data, what level of confidence is considered

1:20:05 Speaker 1: sufficient for decision making or policy applications.

1:20:11 Speaker 3: That is a really good question and I think the

1:20:14 Speaker 3: answer to that does depend somewhat on you know, the

1:20:19 Speaker 3: decision making person or body.

1:20:24 Speaker 4: But I mean, first of all, when you have remote.

1:20:34 Speaker 3: Remote sensing derived water quality, you're never going to be

1:20:38 Speaker 3: as accurate as in situ sampling. So I think the

1:20:41 Speaker 3: best use for remote sensing water quality in a decision

1:20:45 Speaker 3: making capacity is to guide use of your limited insitu resources.

1:20:51 Speaker 3: So if you see things from remote sensing that make

1:20:55 Speaker 3: you want to follow up with in citu data, I

1:20:57 Speaker 3: think that's the best use of the remote sensing for

1:21:02 Speaker 3: like actual decision making and policy applications.

1:21:06 Speaker 1: Yeah, I might add to that that depends first of all,

1:21:09 Speaker 1: if you have institute data, and then if you can

1:21:12 Speaker 1: compare with what remote sensing is showing that gives you

1:21:15 Speaker 1: somewhat some confidence or air bar for remote sensing parameters,

1:21:20 Speaker 1: And then how much tolerance your decision would have. What

1:21:24 Speaker 1: range can your decision tolerate? That also is important to know.

1:21:29 Speaker 4: That's a good point.

1:21:33 Speaker 1: And the next question is can stream be used for

1:21:36 Speaker 1: a river that is less than thirty meters wide but

1:21:39 Speaker 1: very long.

1:21:42 Speaker 4: It can definitely be.

1:21:45 Speaker 3: Used, but you may see a reduction and within them

1:21:53 Speaker 3: because some of the land from the area will have

1:21:57 Speaker 3: been averaged into that pixel.

1:22:01 Speaker 1: So you so accuracy may be less in that case.

1:22:04 Speaker 3: So yeah, and depending on how much land was averaged in,

1:22:08 Speaker 3: it may actually be filtered out as not water.

1:22:14 Speaker 1: I will also add that if it's less than thirty

1:22:19 Speaker 1: meters wide, even thirty meter is you need at least

1:22:23 Speaker 1: three pixels clear in water to get good water quality.

1:22:27 Speaker 2: Estimates from remote sensing.

1:22:29 Speaker 1: So that way, I would say there will be a

1:22:31 Speaker 1: lot of contamination from land if.

1:22:33 Speaker 3: You Yeah, it's not an ideal use case, and if

1:22:37 Speaker 3: you do get results from stream, it won't be nearly

1:22:42 Speaker 3: as reliable as if it were a much wider water body.

1:22:48 Speaker 2: Yeah.

1:22:50 Speaker 1: Next question is do you use Do you use any

1:22:54 Speaker 1: of the water processing software such as eco light CENT

1:22:57 Speaker 1: to corel light, to gen C, to our CEC for

1:23:01 Speaker 1: converting toa surface level reflectance to water living.

1:23:07 Speaker 3: We don't use any of those in specific, but we

1:23:10 Speaker 3: do regularly compare our performance to those. We use our

1:23:15 Speaker 3: own in house atmosphere correction that.

1:23:19 Speaker 4: Is a mixture of.

1:23:23 Speaker 3: Raley correction and Gassess absorption correction, and then we use

1:23:28 Speaker 3: a mixture density network to do the aerosol contribution.

1:23:32 Speaker 4: So as a as a package all of that atmosphere

1:23:34 Speaker 4: correction is done.

1:23:37 Speaker 3: Just by us, but we do compare it pretty regularly

1:23:41 Speaker 3: to some of those other processors.

1:23:44 Speaker 1: Yeah, just to add, we will have some information on

1:23:49 Speaker 1: this next session because md and model will require water

1:23:56 Speaker 1: living reflectancies. So you could start with any level of

1:24:00 Speaker 1: satellite data and use any of this atmospheric correction models

1:24:04 Speaker 1: to convert CoA to surface level reflectancy remote sen collectancies,

1:24:09 Speaker 1: and then to attend to m d N. Next question

1:24:16 Speaker 1: is what model was used, machine learning or physical model?

1:24:20 Speaker 1: Which particular ones were evaluated before the best was chosen.

1:24:27 Speaker 3: So we use machine learning for the aerosol contribution because

1:24:32 Speaker 3: that's something that's been proven to be pretty difficult to

1:24:35 Speaker 3: do with the physical model.

1:24:38 Speaker 4: We do.

1:24:40 Speaker 3: Raley correction with a physical model because we found it

1:24:45 Speaker 3: to be more performant than if we also let the

1:24:49 Speaker 3: make sure density NetWorld take care of the Raley contribution,

1:24:53 Speaker 3: so we did.

1:24:53 Speaker 4: We did evaluate it.

1:24:54 Speaker 3: For for the case of ralely, but for aerosol contribution

1:25:00 Speaker 3: much have to use the mixture density network to get

1:25:03 Speaker 3: good performance.

1:25:11 Speaker 1: Question is what are the most common misinterpretation or misuse

1:25:15 Speaker 1: of stream outputs that users should be aware of.

1:25:21 Speaker 4: My answer to that would probably be.

1:25:25 Speaker 6: Cloud pixels can still frequently show up. We do our

1:25:31 Speaker 6: best to remove cloud and land pixels before we create

1:25:36 Speaker 6: the water quality maps, but it's not hard to find

1:25:40 Speaker 6: examples where you know, some cloud pixels may have made

1:25:45 Speaker 6: it through the cloud masking, and naturally we'll throw off

1:25:49 Speaker 6: the results in that area, so you'll see what might

1:25:53 Speaker 6: appear to be really high chlorophyll a data that it's

1:25:56 Speaker 6: just because there's cloud. You should now for the last

1:26:02 Speaker 6: couple weeks of data be able to find RGB true

1:26:06 Speaker 6: color composite maps that should help you see what the

1:26:10 Speaker 6: scene looked like on that day and be able to

1:26:12 Speaker 6: see for yourself what the cloud cover looks like.

1:26:15 Speaker 4: But we still have to.

1:26:19 Speaker 3: Kind of backwards propagate those true color composite maps for

1:26:23 Speaker 3: the rest of the stream data.

1:26:25 Speaker 2: SEP great. Thank you all.

1:26:30 Speaker 1: I think we are almost at the end of this session.

1:26:34 Speaker 1: If there are any more questions, we will try to

1:26:37 Speaker 1: address them later on, but we hope to see you

1:26:40 Speaker 1: next week at the same time on seventeenth of February.

1:26:44 Speaker 1: Thank you all for attending this session. Thank you to

1:26:47 Speaker 1: our guest speaker once again, and all the r SET

1:26:50 Speaker 1: team for their help with this training. Yeah I think

1:26:59 Speaker 1: there's one more question. Can it monitor turbidity in lakes?

1:27:09 Speaker 4: I think rabidity correlates to.

1:27:14 Speaker 2: Seki depth. Yeah, yeah, so you.

1:27:20 Speaker 1: Can look at sechi depth and then you know that

1:27:24 Speaker 1: will show increase or decrease turbiitity. If you actually want

1:27:27 Speaker 1: to to relate it with turbidity units, then you probably

1:27:32 Speaker 1: need that conversion, a relationship between those two in the

1:27:36 Speaker 1: water body of your interest. Okay, thanks everyone, We'll see

1:27:43 Speaker 1: you next week now, thank you.