NASA ARSET Estimating Biomass using GEDI

0:01 Speaker 1: Welcome to this RSET training series, Estimating Biomass and Change

0:05 Speaker 1: with JEDI and the obi IAPI. Today is session one

0:10 Speaker 1: of our training series and it's focused on estimating biomass

0:15 Speaker 1: using JEDI. My name is Erica Portis. I'm a scientist

0:19 Speaker 1: at NASA's Jet Propulsion Laboratory in Pasadena, California, and I

0:24 Speaker 1: focus on the study of terrestrial ecosystems using SAR. I'm

0:28 Speaker 1: also an instructor with the RSET program and I will

0:32 Speaker 1: be hosting this training. Before we begin today's training, I'll

0:38 Speaker 1: say a few quick words about the r SET program.

0:42 Speaker 1: The NASA Applied Remote Sensing Training or our SET program,

0:46 Speaker 1: offers free trainings on the use of remote sensing and

0:49 Speaker 1: model data, along with methods and tools for the global

0:53 Speaker 1: science and applications communities. Our trainings are tailored to different

0:58 Speaker 1: experience ranging from introductory to intermediate to advanced. We provide

1:05 Speaker 1: trainings in several thematic areas including agriculture, climate and resilience, disasters,

1:13 Speaker 1: eco conservation, health and air quality, water resources, and wildland fires.

1:20 Speaker 1: Our SET trainings are either online or in person, and

1:24 Speaker 1: there are two modalities. One is live and in structure.

1:29 Speaker 1: Lad like today's training, or the other modality is asynchronous

1:34 Speaker 1: and self paced. All trainings are free of charge and

1:39 Speaker 1: utilize open source data and software. Some of our trainings

1:43 Speaker 1: are delivered in both English and Spanish like this one,

1:47 Speaker 1: and all of our trainings have a Spanish presentation available

1:53 Speaker 1: a PDF presentation available. So let's start today by providing

1:58 Speaker 1: an overview of this training series estimating biomass and change

2:04 Speaker 1: with JEDI and the OBIEAPI. What is happening beneath the canopy?

2:11 Speaker 1: One fundamental question that many people are interested in is

2:14 Speaker 1: forest structure and what is happening beneath the canopy. Now,

2:19 Speaker 1: from many traditional satellites, we do not know this because

2:24 Speaker 1: optical remote sensing sensors can tell us things about, for example,

2:29 Speaker 1: the greenness of the vegetation, but not about the understory

2:32 Speaker 1: or the structure. Spaceborne lighter can do this, such as JEDI,

2:38 Speaker 1: which measures forest structure across more than twenty billion locations

2:44 Speaker 1: on Earth, and you'll learn to use this data how

2:47 Speaker 1: to do this in this training. So by the end

2:51 Speaker 1: of this training, attendees will be able to identify the

2:55 Speaker 1: fundamental characteristics of JEDI data and carbon monitoring data structure, uncertainty,

3:02 Speaker 1: and system requirements. Apply methods for estimating biomass and biomass change,

3:09 Speaker 1: including the integration of Jedi with Lancet time series and

3:13 Speaker 1: calibration with forest inventory data. Evaluate the accuracy and uncertainty

3:19 Speaker 1: of biomass estimates using validation data. Use the Obian tools

3:24 Speaker 1: and APIs to generate and analyze biomass change products, including

3:30 Speaker 1: visualizing results for user defined areas and comparing carbon gains

3:36 Speaker 1: to different climate scenarios. And finally, evaluate advantages and disadvantages

3:43 Speaker 1: of the footprint level versus gridded level JEDI biomass products

3:48 Speaker 1: to determine which data are appropriate for a given use case.

3:54 Speaker 1: The prerequisite for this training is the r SET Fundamentals

3:59 Speaker 1: of Remote Sense as well as the Spaceborne Lighter for

4:02 Speaker 1: monitoring vegetation, structure and biomass using Jedi and this was

4:08 Speaker 1: a course from October of twenty twenty five in our

4:11 Speaker 1: Set training and below that is a list of suggested trainings.

4:19 Speaker 1: This training series consists of two sessions. The first session

4:23 Speaker 1: today is focused on estimating biomass using Jedi and the

4:28 Speaker 1: second session will be next Thursday, May twenty eighth and

4:33 Speaker 1: it will be focused on estimating biomass change with Jedi

4:37 Speaker 1: and Thebian API. The homework opens up on May twenty

4:43 Speaker 1: eighth and it will be due on June eighteenth. It

4:47 Speaker 1: will be posted on the training web page. There will

4:50 Speaker 1: be a certificate of completion to participant who attend all

4:54 Speaker 1: live sessions and complete the homework assignment by the do date.

5:01 Speaker 1: So with that, let's begin with Session one Estimating Biomass

5:05 Speaker 1: using Jedi. We have a guest instructor for this training,

5:10 Speaker 1: doctor Sean Healey, who's a research ecologist with the US

5:15 Speaker 1: Forest Service and US Department of Agriculture. So by the

5:21 Speaker 1: end of today's session, participants will be able to identify

5:26 Speaker 1: the fundamental characteristics of Jedi data and carbon monitoring, including

5:33 Speaker 1: data structure, uncertainty, and system requirements. Explain that Jedi structure

5:38 Speaker 1: and how uncertainty is characterized. Demonstrate how Jedi data are

5:42 Speaker 1: fused with Lancet time series and calibrate it with forced

5:47 Speaker 1: inventory data to estimate biomass change, and finally, apply Jedi

5:53 Speaker 1: estimation methods to visualize and generate biomass products for given

5:58 Speaker 1: user defined areas of interest. How to ask questions Locate

6:05 Speaker 1: the three points on the bottom right a sub menu,

6:09 Speaker 1: click on it. A sub many will open and select

6:12 Speaker 1: Q and A so window will open and write your

6:15 Speaker 1: question in that window and we will address your questions

6:18 Speaker 1: at the end of the webinar. Feel free to submit

6:22 Speaker 1: your questions throughout this webinar. We'll do our best to

6:25 Speaker 1: address as many questions as possible during the live Q

6:28 Speaker 1: and A session. Any remaining questions will be answered in

6:32 Speaker 1: a Q and A document which will be posted on

6:36 Speaker 1: the training website approximately one week after the training. So

6:42 Speaker 1: let's get started. This next part is going to be

6:46 Speaker 1: presented by doctor Sean Healey. We're very fortunate to have him,

6:50 Speaker 1: so welcome doctor Healey.

6:56 Speaker 2: Hello, I'm Sean Heally. I am speaking for doctor Jijiang Yang.

7:01 Speaker 2: We both work for the USDA four Service, the Inventory

7:05 Speaker 2: and Monitoring Program. I am very happy to be with

7:08 Speaker 2: you today to talk to you about JEDI. JEDI is

7:12 Speaker 2: an acronym. It stands for Global Ecosystems Dynamics Investigation. It

7:19 Speaker 2: is a NASA mission. It's a light our instrument that

7:23 Speaker 2: is on the International Space Station. I'll talk a little

7:26 Speaker 2: bit more about that, but before I do, I want

7:28 Speaker 2: to clarify. So I will be taught starting with JEDI,

7:33 Speaker 2: but I will also be talking about obi Wan, which

7:36 Speaker 2: is another Star Wars reference. It's another acronym, it doesn't

7:39 Speaker 2: really matter what it stands for. But basically, obi wan

7:43 Speaker 2: is an application that is specifically designed with the requirements

7:49 Speaker 2: of carbon monitoring in mind. So we're going to start

7:53 Speaker 2: with the Jedi mission and we're going to be going

7:56 Speaker 2: we're going to be phasing into the obi Wan tool.

8:01 Speaker 2: And I say phasing because obi wan uses observations from

8:07 Speaker 2: the Jedi lidar, it uses statistics built into Jedi, it

8:13 Speaker 2: uses carbon models built into Jedi, so it's really it's

8:17 Speaker 2: a subsidiary tool of the Jedi mission. We're going to

8:20 Speaker 2: start with the theory and measurements behind Jedi, and between

8:25 Speaker 2: this presentation and the next one there is a second

8:28 Speaker 2: version which you will see.

8:30 Speaker 3: We will get into how obi Wan helps us.

8:37 Speaker 2: Answer real questions in the field of carbon monitoring. So

8:43 Speaker 2: starting with Jedi. It is a NASA Earth Ventures instrument.

8:49 Speaker 2: It launched in twenty eighteen and became operational in twenty nineteen.

8:56 Speaker 2: On the picture on the right there there's not actually

8:58 Speaker 2: a on it that says Jedi. I've added that, but

9:03 Speaker 2: that's basically what it looks like. It's the size of

9:05 Speaker 2: a refrigerator. It is a laser.

9:08 Speaker 4: There.

9:09 Speaker 2: Actually there are three lasers on Jedi, and they help

9:14 Speaker 2: us measure the basically the height of trees around the world,

9:19 Speaker 2: or at least in the latitudes that the International Space

9:24 Speaker 2: Station overpasses between fifty one degrees north and fifty one

9:27 Speaker 2: degrees south. There is another rset training that goes over

9:32 Speaker 2: more of the specifics of Jedi how to and focuses

9:36 Speaker 2: on how to get Jedi data. We're going to be

9:39 Speaker 2: talking today really about how Jedi makes inferences from those

9:44 Speaker 2: measurements related to how much biomass is in particular places.

9:49 Speaker 2: And that's on the way to talking about the Obi

9:53 Speaker 2: Wan tool. I should say that it became operational twenty nineteen,

9:58 Speaker 2: but it took one year nap between March twenty three

10:02 Speaker 2: and April twenty four. It has since resumed operations and

10:10 Speaker 2: the PI is rapped Dubai at the University of Maryland.

10:12 Speaker 2: I should also say that NASA very recently selected the

10:17 Speaker 2: Edge mission, which is similar to Jedi, except it's free flying.

10:22 Speaker 2: It will be free flying. It should launch around twenty thirty.

10:26 Speaker 2: NASA selected that through the Earth System's Explorer program. The

10:30 Speaker 2: pis there are at UCSD.

10:35 Speaker 3: And the University of Maryland.

10:37 Speaker 2: So the technologies that you're seeing here will be available

10:41 Speaker 2: into the future.

10:44 Speaker 3: Okay, so what does JEDI do.

10:47 Speaker 2: It really has three lasers, but those lasers are split

10:51 Speaker 2: and dithered in such a way that it really gives

10:54 Speaker 2: us eight tracks. As the International Space Station goes over

10:58 Speaker 2: an area, it's collecting eight tracks worth of.

11:03 Speaker 3: Height data.

11:04 Speaker 2: And as the mission has continued to go on, you

11:08 Speaker 2: can see that the sample is filling in. So it's

11:11 Speaker 2: not an imager. It is collecting footprints along tracks. And

11:16 Speaker 2: the tracks are not contiguous, they're interrupted. But so there

11:20 Speaker 2: are thirty meters footprints along all of those tracks that

11:24 Speaker 2: have been collected since twenty nineteen.

11:29 Speaker 3: And what do these footprints look like?

11:31 Speaker 2: As I said, they're about twenty five meters in diameter.

11:35 Speaker 2: And really what you get the main observable from JEDI

11:38 Speaker 2: is this representation of the vertical forest structure. So as

11:47 Speaker 2: laser energy comes down from the space station, it bounces

11:51 Speaker 2: off of the surface. Some of that energy bounces off

11:55 Speaker 2: the top of the canopy, some of it makes it

11:58 Speaker 2: all the way down to the ground. And you know,

12:03 Speaker 2: because we know the speed of light and how long

12:06 Speaker 2: it took different parts of that waveform to come back.

12:11 Speaker 3: We infer this waveform.

12:14 Speaker 2: The business end of the waveform is really when you

12:17 Speaker 2: quantize that to things that we can put into a

12:20 Speaker 2: biomass model. So relative height our H is basically the

12:25 Speaker 2: height above where we found the ground a particular percentage

12:31 Speaker 2: of the waveform's energy is returning. So our H one

12:34 Speaker 2: hundred is basically the top of the canopy. URH fifty

12:38 Speaker 2: is the height at which half of the stuff that

12:43 Speaker 2: light our energy is returning from. Half of the stuff

12:45 Speaker 2: is below and half of the stuff is above. And

12:48 Speaker 2: so basically we quantize the waveform into things, into these

12:53 Speaker 2: measures that we can put into a biomass model. There

12:58 Speaker 2: are a lot of jet well, there's four levels of

13:01 Speaker 2: Jedi products. It's not it's not a bewildering array of products.

13:05 Speaker 2: The level one two, especially the level one and two

13:10 Speaker 2: are are sort of more basic, uh physics oriented waveform properties.

13:18 Speaker 2: Level four is really where we're going to focus today.

13:22 Speaker 2: That Level four A is how much biomass for every

13:26 Speaker 2: footprint has a biomass estimate. So level four A is

13:31 Speaker 2: talking about how how we model using those relative height measures,

13:37 Speaker 2: how we model biomass. And Level four b is how

13:42 Speaker 2: we take the sample the Jedi gives us of those

13:47 Speaker 2: predicted biomass values, how we fill in the entire area

13:52 Speaker 2: for one kilometer grid cells. So that's really a sample

13:57 Speaker 2: survey sort of a process.

13:59 Speaker 3: And we.

14:01 Speaker 2: Very similar statistics to those used by the forest inventory

14:05 Speaker 2: that I actually work for.

14:07 Speaker 3: So let's just jump right in.

14:12 Speaker 2: How do we get our predictions for biomass at the

14:15 Speaker 2: footprint level. Basically, the Jedi Mission has asked cooperators from

14:22 Speaker 2: around the world for light our data that's collected directly

14:26 Speaker 2: over inventory plots where biomass has been measured on the ground,

14:31 Speaker 2: and throughout this presentation you will see references here it's

14:36 Speaker 2: Duncanson at All twenty twenty two to the source material

14:40 Speaker 2: if you want more details about the processes. But basically,

14:46 Speaker 2: what the Jedi Mission has done is simulate Jedi waveforms

14:49 Speaker 2: from the airborne light are and used the paired Jedi

14:55 Speaker 2: waveform and field data to make a model, a specific

15:00 Speaker 2: model for every continent, and every continent is split between

15:04 Speaker 2: plant functional types deciduous, evergreen, et cetera. So, and important

15:13 Speaker 2: for our purposes, where we're headed is estimating biomass and inference.

15:20 Speaker 2: Where we're headed is we are keeping the uncertainty that's

15:27 Speaker 2: associated with these models, and we won't get into exactly

15:31 Speaker 2: how that's stored, but there is an uncertainty stored with

15:35 Speaker 2: every model, and we use that in this next slide. Actually,

15:40 Speaker 2: so when we go to L four B JEDI L

15:44 Speaker 2: four B, that's the gridded biomass product, where we are

15:47 Speaker 2: inferring the mean biomass for every one kilometer grid cell.

15:54 Speaker 2: What we're doing here, I mean it looks like here

15:57 Speaker 2: we've got a sample, and that's exactly what JEDI gives us.

16:01 Speaker 2: So every one kilometer cell, not every one kilometer cells.

16:05 Speaker 2: Some one kilometer cells are unlucky, either because of the

16:09 Speaker 2: overpassed pattern of the space station or because they're perpetually cloudy.

16:15 Speaker 2: Some one kilometer grid cells don't have much of a sample,

16:22 Speaker 2: and those grid cells get a different algorithm, which I'll

16:27 Speaker 2: talk about in the next side. But for those cells,

16:29 Speaker 2: which is most cells that do have a good JEDI sample,

16:34 Speaker 2: really what we do is we use an estimator, a

16:37 Speaker 2: statistical estimator that estimates the mean biomass from all the

16:43 Speaker 2: predictions that are in that grid cell, and we also

16:47 Speaker 2: produce an estimate of uncertainty a standard error around the

16:51 Speaker 2: estimated mean. That is a function both of the model uncertainty.

16:58 Speaker 2: If you remember from the previous we saved a lot

17:02 Speaker 2: of not a lot. We saved the relevant model uncertainty information,

17:07 Speaker 2: so that gets propagated through this process and also a

17:12 Speaker 2: sampling uncertainty. As you can see, we're not measuring everything

17:17 Speaker 2: in that one kilometer grid cell, so this estimator Patterson

17:20 Speaker 2: at all uses propagates both sources of uncertainty to get

17:27 Speaker 2: a standard error for four our estimate of mean biomass

17:31 Speaker 2: in one kilometer grid cells around the world. So that's

17:34 Speaker 2: the L four B product. I mentioned that there are

17:39 Speaker 2: some L four B one kilometer grid cells that don't have,

17:44 Speaker 2: for whatever reason, enough.

17:48 Speaker 3: Enough Jedi measurements in there to.

17:51 Speaker 2: Do that process, so this is what we do in

17:55 Speaker 2: those cases.

17:56 Speaker 3: The green squares.

17:58 Speaker 2: Are field plots around the world and that are used

18:03 Speaker 2: by the Jedi team to make our biomass predictions. So

18:06 Speaker 2: we've already talked about the field to Jedi model. So

18:10 Speaker 2: there's one model, but then there's then we predict. We

18:14 Speaker 2: use Jedi's predictions for all the waveforms in a particular area.

18:20 Speaker 2: We're using currently we're using ten kilometer squares ten by

18:24 Speaker 2: ten kilometer squares, and we intersect those predictions of biomass

18:32 Speaker 2: with lance At time series information, So not just a

18:35 Speaker 2: single lance At image, but all of the available clear

18:40 Speaker 2: imagery for a lance AT pixel gets fit into this

18:44 Speaker 2: time series function, and we use properties of that function

18:48 Speaker 2: to make our prediction of biomass with a model that

18:53 Speaker 2: we calibrate with Jedi's predictions, so very local model. We're

18:58 Speaker 2: using Jedi predictions to predict a Lancet model.

19:04 Speaker 3: Okay, So, and we've talked about where that cheese.

19:07 Speaker 2: This is important for obi Wan because this really is

19:11 Speaker 2: the basis for how.

19:12 Speaker 3: Obi Wan works through time.

19:14 Speaker 2: And we'll come to that in just a moment, but

19:17 Speaker 2: before we do, i'd like to show you.

19:21 Speaker 3: What the L four B product looks like.

19:25 Speaker 2: You will be seeing a little bit later in an

19:28 Speaker 2: app that I will be walking through and that you

19:32 Speaker 2: will also have available to you. You will see that

19:36 Speaker 2: there are a lot of one kilometer grid cells in

19:39 Speaker 2: version two, which is what the app is seeing, that

19:44 Speaker 2: do not have an estimate. Version three, which has been

19:48 Speaker 2: submitted to the DAC and will be actually I think

19:51 Speaker 2: it is available currently, fills in all of those gaps with.

19:56 Speaker 3: The process that I just talked about.

19:58 Speaker 2: But basically every place in the world has a prediction

20:02 Speaker 2: at the one kilometer scale of biomass from JEDI, and

20:07 Speaker 2: there's also a corresponding standard error of that estimate map.

20:12 Speaker 2: So this process doesn't have to only apply to one

20:15 Speaker 2: kilometer grid cells.

20:16 Speaker 3: It can apply to entire countries.

20:20 Speaker 2: In this case, what we have done is pointed that

20:24 Speaker 2: process at sixty four thousand hectare hexagons that cover the

20:30 Speaker 2: United States, and in this paper we compared those estimates

20:36 Speaker 2: with estimates from FIA, that's the Forest Inventory, the part

20:41 Speaker 2: of the Forest Service that I work for, and really

20:44 Speaker 2: there's a pretty good match spatially, you know, spacially and

20:49 Speaker 2: looking at the scatterplot. So we'll talk a little bit

20:54 Speaker 2: more about validation with FIA, especially in the next presentation,

21:01 Speaker 2: but in general it lines up. And the important thing

21:03 Speaker 2: is that it doesn't have to be a one kilometer

21:05 Speaker 2: grid cell. It can be you know, any shape, and

21:10 Speaker 2: it can be multiple multiple patches too.

21:15 Speaker 3: So now we're.

21:15 Speaker 2: Getting into obi WAN, which is a tool for estimating

21:21 Speaker 2: biomass change, and change is an operative word in this

21:26 Speaker 2: slide because so far all we've talked about with Jedi

21:31 Speaker 2: is single point in time biomass. But we will see

21:36 Speaker 2: in the next version of this presentation, the next session

21:42 Speaker 2: of this presentation that practical carbon accounting.

21:46 Speaker 3: Really requires a view over time.

21:50 Speaker 2: And so obi Wan is using of the measurements, the

21:54 Speaker 2: statistics and the models that we've already talked about for Jedi,

21:58 Speaker 2: but pointing them directly for the purpose of estimating change. Okay,

22:05 Speaker 2: so you will remember this that this is sort of

22:09 Speaker 2: the alternate algorithm for predicting biomass for a single point

22:14 Speaker 2: in time for the one kilometer grid cells where there

22:20 Speaker 2: is not enough Jedi, not enough Jedi footprints directly in

22:25 Speaker 2: the one kilometer cell. So what we're doing there's two

22:30 Speaker 2: levels of model, one from the field to Jedi and

22:34 Speaker 2: then one fitted very locally at ten kilometer by ten

22:38 Speaker 2: kilometer area Jedi predictions to lance At imagery. And the

22:46 Speaker 2: nice thing is that lance At is consistently processed through time.

22:52 Speaker 2: And so what we do then is we can apply

22:55 Speaker 2: that lance At model not only to the present but

22:58 Speaker 2: to lance At imagery. And again it's lance At time

23:03 Speaker 2: series imagery going back in time to the mid eighties

23:06 Speaker 2: if we want to.

23:09 Speaker 3: So what we end up with is.

23:13 Speaker 2: A time series of predictions through the Lancet record. Lance

23:20 Speaker 2: Modern Lancet started in the mid eighties. The first lance

23:25 Speaker 2: At launched in nineteen seventy two. So what we are

23:29 Speaker 2: doing here in this slide, this is the country of Paraguay.

23:33 Speaker 2: We have fitted our models with JEDI in the current period,

23:39 Speaker 2: and then we are applying that lance AT model back

23:43 Speaker 2: in time. And what we see through time is that

23:47 Speaker 2: there's a lot of reduction of biomass, especially in the

23:50 Speaker 2: western part of fireway through this time period. Okay, so

23:56 Speaker 2: this is how we're getting basically, how we're getting estimates

24:00 Speaker 2: of change. So when we're talking about a company applying

24:09 Speaker 2: for carbon applying to be to sell carbon credits because

24:13 Speaker 2: of forest management, or we're talking about a country complying

24:18 Speaker 2: with its own commitments internationally in terms of mitigating climate

24:23 Speaker 2: change by managing forests in a way that stores more

24:27 Speaker 2: and more carbon, we really need a straightforward way of

24:34 Speaker 2: accounting for the uncertainty of this car of any carbon

24:38 Speaker 2: management credits that we may claim.

24:41 Speaker 3: And so.

24:43 Speaker 2: You might ask, we've got a time series of lance

24:46 Speaker 2: AT predictions, how do we track.

24:49 Speaker 3: The uncertainty of.

24:52 Speaker 2: The predicted change for a particular for an area that

24:55 Speaker 2: we care about, and in this slide, it looks like

24:58 Speaker 2: we care about a square. You know, it could be

25:01 Speaker 2: as I said earlier, it could be any shape, it

25:04 Speaker 2: could be multiple areas from across, from across the region.

25:13 Speaker 2: We do require that it be at least five hundred

25:18 Speaker 2: hectares because the statistics that we're using, some of the

25:22 Speaker 2: assumptions break down for smaller areas than that. But basically

25:28 Speaker 2: the top set is basically what I just showed you.

25:30 Speaker 2: We make two models feel to Jedi Jedi to Lancet.

25:35 Speaker 2: We apply the model through time, and let's say that

25:37 Speaker 2: we care about the change between nineteen ninety and twenty twenty.

25:42 Speaker 2: We get a prediction in ninety of ninety five, and

25:47 Speaker 2: thirty years later it's eighty tons per hectare, indicating there's

25:52 Speaker 2: a loss of fifteen.

25:53 Speaker 3: So what we.

25:54 Speaker 2: Actually do behind the scenes in obi Wan is change

26:01 Speaker 2: both of those models feel the Jedi Jedi to Lancet

26:05 Speaker 2: in ways that across a lot of iterations represent what

26:09 Speaker 2: we know about the uncertainty of those models.

26:12 Speaker 3: And we call that bootstrapping.

26:15 Speaker 2: And I'll talk a little bit about how we sort

26:18 Speaker 2: of jitter those models in a second. But basically we

26:22 Speaker 2: end up making a lot of different time series of predictions,

26:28 Speaker 2: and the first one indicated a loss of fifteen, the

26:32 Speaker 2: second one indicated a loss of thirteen, third a loss

26:35 Speaker 2: of twenty, and across a lot of these iterations or bootstraps,

26:40 Speaker 2: what we will end up with is a distribution of

26:44 Speaker 2: predictions and that's what you see on the right, and

26:49 Speaker 2: the mean of those predictions, that's what we return as

26:54 Speaker 2: an estimated mean, and the standard deviation around that mean

26:59 Speaker 2: is what we.

27:00 Speaker 3: Support as a standard error of our estimate.

27:04 Speaker 2: Just a little bit of detail about how we are

27:08 Speaker 2: jittering these models this you know too.

27:11 Speaker 3: Two models.

27:13 Speaker 2: The first one we have we have because the Jedi

27:17 Speaker 2: mission provides it. We have it's a linear model that

27:21 Speaker 2: goes from the field to field field biomass to Jedi

27:26 Speaker 2: biomass UH, and we have a parameter covariance matrix for

27:31 Speaker 2: for the parameters in that linear model. We can we

27:35 Speaker 2: j We jitter that model on that basis, on the

27:38 Speaker 2: basis of the uncertainty that that is implied by those parameters.

27:44 Speaker 2: And then we also I won't go into the details,

27:46 Speaker 2: but we also re sample the second level of model

27:50 Speaker 2: ah in in a way that gives us a lot

27:53 Speaker 2: of different time series that we need that we query

27:56 Speaker 2: to to return a predicted change and standard around that change.

28:02 Speaker 3: Okay, so here's sort of a.

28:07 Speaker 2: Red flag that is that may be on your mind,

28:11 Speaker 2: especially if you do a lot of remote sensing.

28:13 Speaker 3: So lansat imagery is optical imagery.

28:17 Speaker 2: And as forests grow, let's say they're growing from scratch,

28:21 Speaker 2: you know, from from an open field. In general, the

28:25 Speaker 2: more the forest expands, the greener the imagery gets until

28:32 Speaker 2: the canopy closes. And at that point, the stuff under

28:37 Speaker 2: the canopy continues to grow. The trees get taller, they

28:40 Speaker 2: put on more and more biomass. But there the imagery

28:44 Speaker 2: does not get a lot greener. There's some shadowing the changes.

28:48 Speaker 2: There's still is some signal there. We have shown that

28:52 Speaker 2: if we make our models very locally it helps, but

28:56 Speaker 2: it's still a problem. Basically, that problem manifests itself in

29:01 Speaker 2: our models. Overestimate low values of biomass and underestimate high

29:10 Speaker 2: values of biomass. So that's a bias right there. And then,

29:14 Speaker 2: especially when we're looking at change from either high biomass

29:17 Speaker 2: to low or in the opposite direction, we will systematically

29:21 Speaker 2: be underestimating that change.

29:26 Speaker 3: That's a problem.

29:27 Speaker 2: So what obi Wan does is very explicitly use calibration

29:32 Speaker 2: to address that problem. And we will be talking about

29:37 Speaker 2: two options for calibrating in the next session. The first

29:43 Speaker 2: one is using Jedi waveforms that we didn't use to

29:47 Speaker 2: make the lance At model, and the second is using

29:50 Speaker 2: inventory data that may be available in a particular country.

29:56 Speaker 2: Both of those basically what we do is we look

29:58 Speaker 2: at the predictions that we see in these time series

30:04 Speaker 2: and we develop a really simple calibration function that in essence,

30:10 Speaker 2: most of the time it pushes the high predictions upward

30:14 Speaker 2: and the low predictions downward, because you know, those are

30:17 Speaker 2: generally the what has to happen if we want to

30:21 Speaker 2: remove the biases that integrating that process into the process

30:26 Speaker 2: that I that I just mentioned is really simple and

30:30 Speaker 2: computationally it's very simple, and we'll see a little bit

30:35 Speaker 2: about that with so there's very little additional computing. I

30:38 Speaker 2: just want to make the point that and we'll come

30:42 Speaker 2: back to this in the next session, that there is

30:44 Speaker 2: calibration possible and in fact it is happening.

30:49 Speaker 3: Okay, So before we.

30:51 Speaker 2: Get to sort of a hands on activity that we

30:55 Speaker 2: in which we will be exploring.

30:57 Speaker 3: Jedi data.

31:00 Speaker 2: With a Googleer's Engine app. I just want to point

31:03 Speaker 2: out these resources. There is also, of course, the earlier

31:10 Speaker 2: R set trainings in this series which talked about light

31:14 Speaker 2: R in general and light R specifically from Jedi. Okay,

31:20 Speaker 2: so the hands on activity that we will be pursuing

31:25 Speaker 2: Before we get into it, we'll talk a little bit

31:27 Speaker 2: about what it is. We will be exploring the distribution

31:31 Speaker 2: of Jedi shots. You know what this when you look

31:35 Speaker 2: at Jedi data, how.

31:38 Speaker 3: Dense is it.

31:40 Speaker 2: And how does that density change from place to place.

31:44 Speaker 2: We'll be looking at what Jedi actually measures at each

31:49 Speaker 2: of its footprints.

31:51 Speaker 3: You will see very clearly that they.

31:54 Speaker 2: Line up in paths that echo the overpass pattern of

31:59 Speaker 2: this National Space Station. Lastly, we will be experimenting with

32:07 Speaker 2: the estimator that I mentioned earlier that propagates the uncertainty

32:13 Speaker 2: of the sampling, the model air and the sampling error.

32:20 Speaker 3: So I just want to point out that this.

32:25 Speaker 2: Is possible because Jedi data is stored on Google Earth Engine.

32:29 Speaker 2: Here is the thing that you would search for in

32:33 Speaker 2: the catalog, and here is the link to the Jedi

32:38 Speaker 2: app that we will be exploring. Now, Okay, so here

32:49 Speaker 2: we are. You have registered for Google Earth Engine. If

32:53 Speaker 2: you weren't already registered, you've clicked on the link, and

32:57 Speaker 2: this is a Google Earth Engine application that we built.

33:01 Speaker 2: Really it was Jijiang Yang who built this for this training.

33:05 Speaker 2: And what we're doing here is looking at a number

33:10 Speaker 2: of things. What I don't I think that you can

33:14 Speaker 2: see my cursor here. What we're looking at now is.

33:18 Speaker 3: Jedi L four B.

33:20 Speaker 2: Those are the statistical estimates at the one kilometer grid scale,

33:27 Speaker 2: so you know, just zooming in, you will see that

33:31 Speaker 2: there are places where there's no estimate. Those are places where,

33:35 Speaker 2: at least at the time of this version, which was

33:39 Speaker 2: version two of Jedi, there weren't enough Jedi footprints in

33:44 Speaker 2: there to make an estimate. So if you want to

33:48 Speaker 2: see imagery underneath what we're showing, you can just press

33:53 Speaker 2: the satellite button here, and I'm just zooming into a

33:59 Speaker 2: particular place. So these are the one kilometer estimates. I'm

34:06 Speaker 2: going to press this inspect pixels button and just clicking

34:11 Speaker 2: on a you know, a pixel at random. Pretty high

34:15 Speaker 2: biomass three hundred and fifty seven tons per hectare. We've

34:20 Speaker 2: got a standard air of ninety five tons per hectare.

34:25 Speaker 3: So this is actually fairly uncertain.

34:27 Speaker 2: There's a lot of variability in this because the standard

34:31 Speaker 2: error is maybe thirty percent of or actually it's yeah,

34:37 Speaker 2: well it's about thirty percent.

34:38 Speaker 3: That's a pretty high error.

34:41 Speaker 2: In addition to seeing so you can explore really this

34:45 Speaker 2: is sort of a tool for you to explore Jedi data.

34:48 Speaker 3: It's global.

34:50 Speaker 2: I'm focusing I think it's this is Alabama. But as

34:53 Speaker 2: you zoom in, in addition to these one kilometer estimates,

34:58 Speaker 2: you can.

34:59 Speaker 3: Choose the Jedi product that you would like to see.

35:03 Speaker 3: L two A that's a footprint level.

35:08 Speaker 2: So I'm just zooming in so that you can see

35:12 Speaker 2: the density of the thirty meter footprints. It's really pretty dense.

35:20 Speaker 2: So as we as we zoom in, you can see

35:25 Speaker 2: this looks brown.

35:26 Speaker 3: The forest here looks brown.

35:28 Speaker 2: I have a feeling that this is a sort of

35:29 Speaker 2: wintertime image.

35:31 Speaker 3: When I click.

35:33 Speaker 2: On one of those pixels. This L two A is

35:38 Speaker 2: the product that has those relative height values. So the

35:41 Speaker 2: top of the canopy is about thirty one meters tall.

35:45 Speaker 2: For the for the footprint that I clicked on fifty

35:49 Speaker 2: that is sort of the midpoint of the canopy. Stuff

35:53 Speaker 2: that intersects that intercepts Jedi is about nine meters, so

35:59 Speaker 2: there's a pretty far distance between the midpoint of all

36:03 Speaker 2: the all the photosynthetic material in the top. There is

36:07 Speaker 2: also this L two B product that's also a footprint

36:12 Speaker 2: level product. It is the predicted predicted can it be

36:18 Speaker 2: cover for a particular waveform? And this one is showing

36:23 Speaker 2: about point nine can it be covers? That really tells

36:26 Speaker 2: me this is not brown because it is.

36:30 Speaker 3: Barren.

36:31 Speaker 2: Is brown because this is a wintertime high resolution image.

36:35 Speaker 2: If you want to see what's under there, you can

36:39 Speaker 2: this slider bar opacity can make this well our Jedi

36:45 Speaker 2: data invisible for a moment. I'm just going to switch

36:49 Speaker 2: over to another instance.

36:51 Speaker 3: This is on the coast of California.

36:56 Speaker 2: I'm going here because the redwoods are here, they're very

37:02 Speaker 2: they have very you know, tall heights obviously, so I'm

37:06 Speaker 2: just going to zoom into, you know, a place here

37:11 Speaker 2: and instead of looking at the one kilometer estimates, I'm

37:14 Speaker 2: going to look at the the Jedi footprints. You can

37:18 Speaker 2: see that there's you know, there are i think twenty

37:22 Speaker 2: billion high quality Jedi shots around the world currently. I'm

37:27 Speaker 2: going to display them against forest data. So these are

37:36 Speaker 2: I mean, the redwoods are famous for being tall. Obviously,

37:41 Speaker 2: I'm going to press inspect pixel that's here, I get

37:46 Speaker 2: a cross hair. I'm going to zoom in, making sure

37:52 Speaker 2: that I hit the middle of this pixel. So we're

37:56 Speaker 2: predicting almost five hundred tons of biomass per heck.

37:59 Speaker 3: There, that's a lot of biomass.

38:04 Speaker 2: You can find some that are over a thousand tons.

38:08 Speaker 2: Keep in mind that these footprint level predictions are not

38:13 Speaker 2: necessarily representative of the whole hectare twenty five meters that

38:19 Speaker 2: might just fall luckily or unluckily by chance on one

38:24 Speaker 2: tall tree, and a lot of that thirty meters might

38:27 Speaker 2: be you know, just one column of wood. So for

38:30 Speaker 2: that small area, the density is over one thousand tons

38:34 Speaker 2: per hectare. Over the whole one kilometer or even larger area,

38:41 Speaker 2: you know, it's it's going to be some footprints that

38:45 Speaker 2: fall in between trees. This is the last thing I

38:48 Speaker 2: want to show you here. For an area that you

38:53 Speaker 2: care about, this panel over here polygon rectangle delete analyze

39:01 Speaker 2: this access is the statistical estimator that I talked about

39:06 Speaker 2: earlier that propagates the uncertainty both from the model and

39:14 Speaker 2: from the sample. And so basically I'm just drawing a pixel,

39:18 Speaker 2: and I could draw several pixels, not pixels, polygons, I

39:22 Speaker 2: could draw several of them. When I have the place

39:26 Speaker 2: that I care about, I can press the analyzed button,

39:29 Speaker 2: and over here this panel tells us that the area

39:35 Speaker 2: is six hundred and twenty seven hectares. It's estimating three

39:39 Speaker 2: hundred and fifty five tons per hectare. That's megagrams per hectare.

39:45 Speaker 2: It gives us the standard error this one. It's a pretty.

39:50 Speaker 3: Low standard error.

39:51 Speaker 2: The forest here, I was careful to not include a

39:56 Speaker 2: bunch of different.

39:59 Speaker 3: Forest.

40:00 Speaker 2: If it was really various, the standard error would be higher.

40:04 Speaker 2: There's some other information metion here. There are fifty three clusters,

40:10 Speaker 2: so each one of these lines we treat as a

40:14 Speaker 2: cluster sample. And across those fifty three clusters there's fourteen

40:18 Speaker 2: hundred shots. So the last thing I want to talk about,

40:23 Speaker 2: because this leads into the next session, is that we

40:26 Speaker 2: have built into this app a little button that says,

40:30 Speaker 2: obi Wan report we care about for whatever reason, we

40:34 Speaker 2: care about this place, and we built into this app

40:40 Speaker 2: something that gives some background and records at least what

40:47 Speaker 2: today our estimate for that area was. This could be

40:51 Speaker 2: built into any kind of data system you want. It

40:53 Speaker 2: could look however you want. We'll talk about that in

40:55 Speaker 2: the next session. I just want to point out that

40:58 Speaker 2: even in this fairly simple app, we have options for

41:03 Speaker 2: documenting the out the outputs.

41:07 Speaker 3: That that our estimators are using.

41:10 Speaker 2: This app will be live for the foreseeable future for

41:13 Speaker 2: the purpose of just letting you explore around the world

41:17 Speaker 2: the JEDI data and the capacity for estimating mean biomass

41:22 Speaker 2: that is that is built into obi wan.

41:25 Speaker 3: Thank you.

41:28 Speaker 2: So now we will hand this over to Erica who

41:31 Speaker 2: will lead the Q and a portion of this presentation,

41:35 Speaker 2: and as I said earlier, a lot of what we

41:38 Speaker 2: have hinted at in terms of carbon monitoring, we will

41:42 Speaker 2: explicitly cover in the second session.

41:48 Speaker 1: Thank you very much, Sean. And next we will do

41:51 Speaker 1: a brief summary of the key concepts that were presented

41:55 Speaker 1: during this session. So let's quickly review the key concepts

42:01 Speaker 1: for this session. JEDI is a sensor on the International

42:06 Speaker 1: Space Station. It is an active sensor that measures the

42:10 Speaker 1: distance by timing how long a laser pulse takes to

42:14 Speaker 1: return back to the sensor, and this provides detailed information

42:18 Speaker 1: about elevation canopy heights and three the three dimensional structure

42:23 Speaker 1: of the vegetation. JEDI is a full waveform lightar. Unlike

42:27 Speaker 1: discrete return systems, this sensor records the complete energy profile

42:34 Speaker 1: of each pulse, which captures the entire vertical distribution of

42:38 Speaker 1: vegetation within Jedi's twenty five meter diameter footprint and this

42:44 Speaker 1: as a result, this provides a much richer structural information,

42:51 Speaker 1: so each footprint provides the complete vertical structure of the canopy.

42:58 Speaker 1: And given the as I mentioned, JEDI is on the

43:02 Speaker 1: International Space Station, it doesn't have a regular orbit, so

43:06 Speaker 1: it covers latitudes between plus and minus fifty one degrees

43:11 Speaker 1: and produces both footprint level data at twenty five meters

43:16 Speaker 1: and gridded products at one kilometer resolutions. The relative heights

43:22 Speaker 1: are a property of the waveform that can be plugged

43:24 Speaker 1: into a biomass model, and the Level four A is

43:28 Speaker 1: a biomass density predicted for individual footprints. Level four B

43:34 Speaker 1: are statistical estimates of mean biomass for a one kilometer

43:38 Speaker 1: grid cell using JEDI sample of Level four A biomass predictions.

43:44 Speaker 1: NOWBI one is a forest canopy analysis application that uses

43:50 Speaker 1: JEDI lighter observations together with forced inventory data to estimate

43:55 Speaker 1: biomass and monitor changes in forest carbon over time. JEDI

44:00 Speaker 1: uses sampling theory to estimate mean biomass and uncertainty. Looking

44:08 Speaker 1: ahead to Session two, participants will be able to identify

44:13 Speaker 1: key concepts in carbon monitoring, including system requirements, decision making needs,

44:18 Speaker 1: and the concept of additionality. They'll be able to recognize

44:23 Speaker 1: or you'll be able to recognize how obi Wan estimates

44:26 Speaker 1: biomass change, including its use of JEDI lass time series

44:31 Speaker 1: and under the underlying data infrastructure. Participants will be able

44:37 Speaker 1: to evaluate uncertainty and validation using Forest service inventory data

44:41 Speaker 1: to assess the accuracy and precision of obi wan change estimates.

44:46 Speaker 1: Access open source ob wan API to generate estimates of

44:50 Speaker 1: biomass change and areas of interest, and apply obi Wan

44:54 Speaker 1: tools and APIs to visualize biomass change and compare carbon

44:59 Speaker 1: gains against different climate scenarios. Homework and certificates. There is

45:06 Speaker 1: one homework assignment with this training. It will open on

45:12 Speaker 1: May twenty eighth, that's the next at the end of

45:15 Speaker 1: the next training or the next session next Thursday, and

45:20 Speaker 1: you can access the homework from the training web page.

45:24 Speaker 1: The Answers must be submitted via Google forms and the

45:27 Speaker 1: homework is due by June eighteenth. The certificate of completion

45:33 Speaker 1: will be given to those participants who attend both live

45:39 Speaker 1: webinars and attendance is recorded automatically. Participants complete the homework

45:44 Speaker 1: assignment by the deadline and please expect to receive the

45:48 Speaker 1: certificate via email approximately two months after completion of the training.

45:56 Speaker 1: If you have any questions about the material that was

45:59 Speaker 1: presented to they, please feel free to contact doctor Sean

46:03 Speaker 1: Heally through the email that you see here. We've reached

46:10 Speaker 1: the end of today's session, Thank you very much to

46:14 Speaker 1: doctor Heally, and we are now ready to start the

46:18 Speaker 1: Q and A session.

46:25 Speaker 5: Right, Thanks again, thanks very really thanks.

46:28 Speaker 6: Eric, tag team and one to the spreaders and tag

46:31 Speaker 6: team in here with with Erica because she has to

46:35 Speaker 6: step out, So I'll be leading ast in the quanization

46:39 Speaker 6: here and thanks Bred for putting in there on the

46:43 Speaker 6: on the spring, we can go steck to the to

46:47 Speaker 6: the first question and and then Sean, I'll be I

46:51 Speaker 6: read the question and then feelfree to provide the answer

46:55 Speaker 6: to it. So the first question is are we and

46:59 Speaker 6: some of this questions getting really early in the during

47:02 Speaker 6: the webinars, So I'm assuming that maybe maybe you.

47:05 Speaker 5: Already answered some of them uh D during your clop.

47:10 Speaker 5: But but the first one says, are we measuring bio

47:12 Speaker 5: must change or are we using a sophisticated model based approximation?

47:20 Speaker 3: H Thanks for the question.

47:23 Speaker 2: The answer is, I guess if you give me a

47:26 Speaker 2: choice between those two things, it's a sophisticated model based approximation.

47:30 Speaker 2: I mean, it's almost impossible to measure directly measure you know,

47:35 Speaker 2: a lot of trees over a large area, so there's

47:39 Speaker 2: there probably.

47:40 Speaker 3: Is no direct measurement of biomass change, but there is.

47:44 Speaker 3: Uh what.

47:46 Speaker 2: What we're using is a model based mode of inference

47:48 Speaker 2: where we are accounting for uncertainty that is built into

47:55 Speaker 2: our models. We're also accounting for uncertainty of the introduced

48:00 Speaker 2: by the fact that Jedi is only sampling the landscape

48:03 Speaker 2: and not not collecting an image over the entire population.

48:09 Speaker 2: So the answer, the answer that I put there is

48:12 Speaker 2: that it's you know, it's a model based estimation paradigm,

48:18 Speaker 2: and the uncertainties that we produce are our best understanding

48:23 Speaker 2: of how the errors in our models propagate to the

48:27 Speaker 2: final answer of biomass change.

48:32 Speaker 5: Great, thanks Chub. Question number two, it'sa and space based

48:40 Speaker 5: biomass estimates be accurate and reliable enough to support high

48:44 Speaker 5: stakes carbone decision making. And if not, what monitor approaches

48:48 Speaker 5: would be more appropriate in cases? What comprehensive if ground

48:53 Speaker 5: based monitoring is not peaceable, can space based biomass estimation

48:57 Speaker 5: and service practical do not be complimentary approach? And to

49:02 Speaker 5: what I explain the approach is practically upblet long question there,

49:08 Speaker 5: but jump in.

49:11 Speaker 2: Yeah, two great questions right out of the gate. You know,

49:14 Speaker 2: I think that the gold standard is having a field

49:19 Speaker 2: based inventory, uh that goes back in time and is

49:22 Speaker 2: that is consistently and frequently measured. That is pretty expensive,

49:29 Speaker 2: especially if you know you don't have an inventory for

49:32 Speaker 2: anything else. It's it's hard to imagine, you know, carbon

49:36 Speaker 2: credits paying for that kind of that kind of inventory

49:40 Speaker 2: to be collected all the time. So uh, in in

49:44 Speaker 2: lieu of that, in many places, we don't have any

49:47 Speaker 2: forest inventory. And I believe that what we're talking about

49:52 Speaker 2: is the best estimate that we can get from remote

49:54 Speaker 2: sensing in many parts of the world, and I think

49:59 Speaker 2: you will UH in the next presentation. I believe that

50:04 Speaker 2: in many places that is not good enough. I mean,

50:07 Speaker 2: it's not good it's you know, there are there are

50:09 Speaker 2: definitely biases that creep in. So I although I would

50:15 Speaker 2: say it's the best we have and maybe the best

50:17 Speaker 2: available answer in a lot of cases. There we know

50:22 Speaker 2: that there are biases and that that will be covered

50:25 Speaker 2: in the next the next presentation. I would say, though,

50:28 Speaker 2: and you also see this in the next presentation, that

50:31 Speaker 2: when we in the United States we're using remeasured data

50:36 Speaker 2: from the Forest Inventory and Analysis Unit of the Forest

50:38 Speaker 2: Service to calibrate these estimates, that it's getting much closer

50:44 Speaker 2: to the truth. And I think, you know, in this country,

50:48 Speaker 2: we are entering a space where this can be used

50:52 Speaker 2: for decision support. You know that we we provide probabilistic

50:57 Speaker 2: estimates of uncertainty, which all always includes the possibility that

51:01 Speaker 2: you know, there's you know, there's some bias in there

51:05 Speaker 2: or some errors in there. But I do believe in

51:10 Speaker 2: this country, especially that in the United States, UH, that

51:16 Speaker 2: there is room for using what we're talking about in

51:19 Speaker 2: in operational carbon accounting.

51:22 Speaker 5: Excellent. Big question number three, UH, should temporary force carbon

51:28 Speaker 5: storage be treated as like we willant to permanent fossil

51:32 Speaker 5: free or emission reductions.

51:37 Speaker 2: Well, uh, you know, I I think that the answer

51:40 Speaker 2: depends on the time frame of mitigation that we're talking about.

51:46 Speaker 5: Uh.

51:46 Speaker 3: A lot of.

51:47 Speaker 2: People consider the next few decades as you know, the

51:50 Speaker 2: the real uh you know tip place where tipping points

51:54 Speaker 2: will or will not be passed.

51:56 Speaker 3: Uh.

51:57 Speaker 2: And uh, you know, these upcoming decks decades probably will

52:02 Speaker 2: be you know, see the peak of carbon emissions, hopefully.

52:05 Speaker 3: And you know a lot of people say that.

52:11 Speaker 2: Mitigation in the near future is more important than mitigation

52:15 Speaker 2: four hundred years from now. I don't have a position

52:19 Speaker 2: on that, but I do recognize what the questionnaire is asking. Like,

52:26 Speaker 2: mitigation is not in forest carbon storage, it's not permanent.

52:31 Speaker 2: You know, disturbances are part of life in forest ecosystems.

52:36 Speaker 2: So that question is more of a general forest climate

52:42 Speaker 2: change mitigation question. So I guess the answer to that

52:47 Speaker 2: is it depends on your time frame.

52:51 Speaker 5: Okay, Question number four, is ourh fifty half the elevation

52:56 Speaker 5: of our age one hundred or half the biomass of

53:00 Speaker 5: the column?

53:02 Speaker 3: So the answer is it's neither.

53:06 Speaker 2: As the light from the laser passes from above and

53:10 Speaker 2: eventually bounces off of the ground, some of that is

53:15 Speaker 2: intercepted by the very top of the canopy, some of

53:18 Speaker 2: it's intercepted by the branches and leaves at different levels

53:22 Speaker 2: of the canopy. And RH fifty is really a measure

53:26 Speaker 2: of sort of the midpoint of the density of the

53:30 Speaker 2: stuff that the light passes through and bounces back from

53:34 Speaker 2: on its way to the ground. So if you think

53:36 Speaker 2: of a profile, and I showed a profile of the waveform,

53:43 Speaker 2: that's the point at which half of the energy is

53:46 Speaker 2: returning has already returned back. So yeah, it's really a

53:52 Speaker 2: measure of the waveform. And you know, depending on the

53:57 Speaker 2: distribution of canopy material, that could be half half of

54:01 Speaker 2: the height of the very top, or it might be

54:04 Speaker 2: taller than that or shorter than that, just depending on

54:06 Speaker 2: how that material is distributed.

54:08 Speaker 4: Vertically interesting, Okay, question number five Is it possible to

54:15 Speaker 4: asswer the lighter data using the duncans on at all

54:19 Speaker 4: vapor or is it open source?

54:23 Speaker 2: In general, all NASA sponsored research is open source data. However,

54:32 Speaker 2: the Jedi Mission is collaborating with about one hundred or

54:36 Speaker 2: maybe it's more by now providers from around the world

54:40 Speaker 2: who have put a lot of work into surveying measuring

54:45 Speaker 2: trees and collecting light our data over those trees. So

54:50 Speaker 2: that's proprietary information in many cases shared you know, through goodwill.

54:56 Speaker 2: But part of that goodwill is also about shaped with

55:02 Speaker 2: agreements protecting against sharing. So as much as the mission

55:07 Speaker 2: would like to share that data, it has to respect

55:10 Speaker 2: the providers of that data and it does not share it.

55:16 Speaker 5: Hey. Number six, how do you see the potential of

55:19 Speaker 5: incorporating information from Centinel or higher resolution satellites in addition

55:24 Speaker 5: to the last model?

55:26 Speaker 2: Well, Sentinel too is that's sort of the sensor I

55:31 Speaker 2: addressed in my answer here, that's a fairly easy one.

55:37 Speaker 2: NASA provides harmonized lance at Sentinel data set.

55:42 Speaker 3: I think that globally.

55:45 Speaker 2: It may only go back to twenty eighteen, but basically

55:49 Speaker 2: they make the two sensors sort of interoperable and that

55:53 Speaker 2: would be easy to integrate.

55:54 Speaker 3: We have not done that because.

55:58 Speaker 2: The stuff that I was talking about today, you know,

56:01 Speaker 2: we wanted to go back to two thousand and Sentinel

56:05 Speaker 2: wasn't available at that time. But all of these methods

56:09 Speaker 2: could be adapted. We haven't done it, but they could

56:12 Speaker 2: be adapted to other well, to wall sources of imagery.

56:18 Speaker 3: All right.

56:21 Speaker 5: Number seven here, how do I use Landsat imagery with

56:25 Speaker 5: Jeri data. Which software can I used to do this?

56:28 Speaker 5: Or can I use different coal barriers from landsids such

56:32 Speaker 5: as m dB I e b I. Yes, and additionally

56:37 Speaker 5: kind of a similar other questions are US addresses it's

56:41 Speaker 5: possible to use sending A one or two instead of

56:43 Speaker 5: lanserts for biomass estimation.

56:46 Speaker 3: Yeah, I'll address I just addressed that one. But you

56:50 Speaker 3: know what software would you use?

56:52 Speaker 2: I mean, you could use a lot of different kinds

56:53 Speaker 2: of software, but it would be easy on Google Earth Engine.

56:58 Speaker 2: I think that I mentioned that Google is storing the

57:01 Speaker 2: Jedi data set as an asset. It's very easy to

57:05 Speaker 2: use and it's US also stores lands At and Sentinel

57:09 Speaker 2: and other sources, so fusion with other data sources.

57:14 Speaker 3: One easy answer would be Google Earth Engine.

57:19 Speaker 5: Sober okay. Number eight. Would the bootstrapping time series you

57:25 Speaker 5: mentioned be similar to performing multiple simulations like Carlo simulation?

57:32 Speaker 2: The answer is yes. There's a stochastic element in both

57:35 Speaker 2: of those. I would say that the bootstrapping that we

57:37 Speaker 2: are doing, we're bootstrapping models either Field to Jedi or

57:42 Speaker 2: Jedi to lance At or both. And that's a narrower

57:48 Speaker 2: term and accounts for just the uncertainty involved with those models.

57:55 Speaker 2: Monte Carlo simulation would allow you to incorporate variability from

58:03 Speaker 2: other sources also, so I see Monte Carlo simulations as

58:09 Speaker 2: sort of a broader class of stochastic simulations.

58:16 Speaker 5: Okay, let's do the next couple of questions and and then,

58:23 Speaker 5: just for the benefit of the audience, remember that eventually

58:26 Speaker 5: all the answers will come up in the in the

58:28 Speaker 5: Q and a document uh several days in in a

58:33 Speaker 5: couple of days a week max. Okay. Question nine is

58:39 Speaker 5: estimating biomasses in Jedi applicable and efficient in drylands or grasslands.

58:48 Speaker 2: I I have not personally worked a lot with grasslands.

58:53 Speaker 2: I believe that Jedi's waveforms are.

58:59 Speaker 3: I'm not sure they're great for grasslands, to be honest

59:03 Speaker 3: with you.

59:04 Speaker 2: And and but in dry land forest, I would think

59:07 Speaker 2: that the measurements are are you know, perfectly reasonable to use.

59:16 Speaker 2: We have not tested obi wan so obi Wan brings

59:19 Speaker 2: in the element of modeling Jedi to landsat I'm not

59:27 Speaker 2: sure how well we do in dry land forest. We

59:30 Speaker 2: have not yet tested or validated that in dry land

59:34 Speaker 2: forces specifically. But as I say here, I think we

59:38 Speaker 2: do need to do that. So it's kind of an

59:41 Speaker 2: incomplete answer, but that's that's what I know so far.

59:45 Speaker 5: Yeah, I'd be interested to see how it behaves. All right,

59:51 Speaker 5: Question number ten, I think you mentioned that the assumptions

59:54 Speaker 5: are going to little biomass estimations breakdown for areas less

59:58 Speaker 5: than five hund there? Did I hear that correctly? If so,

1:00:03 Speaker 5: can you speak a little more about that.

1:00:06 Speaker 2: Yes, So, we we're working with model based inference, and

1:00:13 Speaker 2: there is a component of model based inference that involves

1:00:20 Speaker 2: spatial auto correlation of the errors in our models. And uh,

1:00:26 Speaker 2: that is relevant at at very fine scales that we

1:00:30 Speaker 2: and we do not account for that uncertainty, so you know,

1:00:35 Speaker 2: at finer scale, at very fine scales, Uh, you know,

1:00:39 Speaker 2: we simply you know, given the assumptions of our of

1:00:43 Speaker 2: our method, we don't trust those estimates at very fine scales.

1:00:46 Speaker 2: So that's why we recommend at least five hundred hectares

1:00:50 Speaker 2: M makes sense?

1:00:53 Speaker 5: Okay, Yeah, we're gonna We're gonna stop here. And but again,

1:00:58 Speaker 5: as I mentioned with the and a document will eventually

1:01:01 Speaker 5: be available for our participants, uh in the in the

1:01:06 Speaker 5: training web page. Thanks again, Sean and your team for

1:01:10 Speaker 5: for helping with this and for all that valuable information.

1:01:13 Speaker 5: And yes we'll see everyone in Session two next week

1:01:19 Speaker 5: and next Thursday. To have a great day all, and

1:01:23 Speaker 5: we'll see you on Thursday.