Subscribe
Share
Share
Embed
Join the Ecobot team as they talk with environmental experts about their work and how they’re influencing the environmental industry.
CAMERON: Hey, everyone. Welcome to the Ecobite podcast, where we will be diving
into topics around the environmental industry. Join me, Camera Davis, and Egobots product manager and seasoned environmental scientist
Liv Haney, in a deeper conversation with our guest.
If you'd like more context to our conversation
and or a crash course on the topic at hand, please view the Ecobyte video recording before getting started. Either
way, enjoy.
JESSIE: I'm Jesse Meyer. I'm the Director of Technology
at the Environmental Policy Innovation Center,
or Epic. And this is Ego Byte. I was just going to share
a bit about my background. My whole career primarily
has been around wetlands, wetland monitoring,
wetland science, wetland ecosystem benefits. And
I had a lot of road gaps or like, stop gaps throughout
that process. And the first, I would say is, like,
in grad school, I was really focused on where
should we cite wetland restoration projects for
the maximum benefit? Like, what is the scientifically
proven best way to do it? If I'm a coastal community
or if I'm inland, or if I'm in a really urban
area or rural. And to do that, I needed really
good data and I needed to know where are wetlands,
what are the impacts, what is the current land
use? And each of those data sets were at least
ten to 15 years out of date. And if we look larger
beyond just Massachusetts, where I was working
in the US. Some of the wetland data sets are like
50 years old. Alaska is not even mapped. And that
was a really AHA moment for me of like, we don't
have good data to inform and scientifically drive
where we should be doing some of the most impactful
work when it comes to wetland restoration and
conservation. So in grad school, I studied water
resource engineering and environmental policy.
And my thesis really focused on where we should
cite wetland restoration projects and how successful
are we in the restoration that we're doing. So
as part of that, I did kind of two things. One was a large GIS model on prime locations for wetland
restoration based on where wetlands currently
are, where there's been impacts, where we need
more stormwater management, or we need more support
for coastal flooding. And the second part of that
was doing field work on different restoration
projects that had happened in Massachusetts to
say, all right, six months out, a year out, three
years out, what are we seeing, and how quickly
does wildlife and the benefits of restoration
start? How does it come to fruition? I then, during
grad school, was trying to do a bit of field work
on wetland science and what can we track in terms
of benefits following a restoration project? First,
I was thwarted by my fear of snakes, terrified,
was not a field scientist person, but I tried,
and.
LIV: We.
JESSIE: Put all these stream gauges throughout
because there was a river running through the
wetland restoration. And we wanted to know how
the stream flow changed. And I had to carry an
umbrella with my laptop, with a USB cord. Sometimes
the monitors that we put in were just cut and
stolen. And sometimes it was like so hard to physically
get to the location to try to plug in my laptop
and extract the data. This was about five years
ago, maybe a little bit longer, maybe everything's
changed and things are in the cloud. But our monitoring
and sampling plan really depended on where we
could easily get to. And I couldn't get that data
on a regular basis because it depended on when
I could get out to the field and extract the information.
And sometimes there's like, weird technology errors.
I want to say maybe we fix that, but who knows?
But that led me to really think, one, I can't
do fieldwork terrified of snakes. Like, there
has to be a better way to do this or a way that
can complement the folks that want to be in the
field. So after grad school, I landed at a technology
company called Upstream Tech that was using satellite
imagery to monitor large landscapes. And I was
really excited to think about the application
for wetlands. We quickly realized, right, satellites
might make sense in some scenarios. There's daily
imagery, there's weekly imagery, there's biweekly
imagery, but also we might need drones because
they have much higher quality information. And
so I was really thinking about this and educating
myself and the stakeholders that might be able
to use this on what type of information and technology
makes sense for wherever you're at in your restoration
project or design. And quickly hit some other
barriers, which was folks were scared, didn't
know what it was. And me working at the technology
company, they didn't want to listen to my opinion
because they're like, you're just trying to sell
me something. And there was a real dearth of information
within the wetland community and environmental
community writ large around when could emerging
or established or innovative technologies be useful
in the environmental work that I'm doing? And
I didn't see that community really anywhere. And
then what would happen when we would have these
conversations is maybe one district was trying
it, one state was doing it, but there wasn't cross
sharing and there wasn't consistency in terms
of when and how this information could be applied.
That also led to sometimes there being regulatory
barriers or perceived regulatory barriers which
just shut down the conversation entirely. At that
point, I thought, okay, where could better policy
or process or folks on the other side of the room
that had a technology background be helpful in making use of innovative tools for environmental decision making? So
I landed at Epic, where I now lead our technology
team. And that's what we think about all day.
It's just what are the tools, people, processes,
regulations, changes that are necessary to complement
environmental decision making or complement our
use of technology environmental decision making.
Learned a bit, still trying to figure it out.
But I see a lot of value in technology and complementing
our restoration work, and hopefully we can figure
out how. Some of the things that I would look
forward to in the next couple of years is up to
date information about where wetlands are that
should be consistently available for all the US
acoustic monitoring to hear the sights and sounds of restoration, easy. And a to track what is there
if we can't necessarily see it, and just a publicly
available information, much more easy. So we don't
have pockets of information that live within one
consulting firm or one agency or one company.
LIV: Right.
JESSIE: We should be able to have all that information
publicly and easily accessible, and we have the
technology to do that. That's pretty easy lift.
So some are hard, some are easy, but I think within
the next five years we'll see a lot of progress
on the technology front.
CAMERON: Everybody to the Ecobite podcast where we delve a little
deeper into the presentation we just heard from
Jesse Mar. I'm here with my co host, Liv.
BECCA: I'm Liv Haney, product manager for Egobot.
CAMERON: And we also have another member of the Epic team, Becca Matson. Go ahead and introduce yourself, Becca.
LIV: Hi there. I'm the director of the Restoration
Economy Center and I'm a colleague of Jesse Moore.
And we have some crossover sometimes that relates
to technology and wetlands.
CAMERON: Lots of technology, lots of wetlands.
Let's get into it. Liv, you have a question? Go
for it.
BECCA: Yes. Jesse, thanks for presenting. I thought
you brought up a lot of interesting and important
topics regarding wetland, the future of wetland
restoration and conservation. One thing you mentioned
on sort of at the end is data management and data
moving from agencies and stuck in private sectors
and sort of being held captive by certain elements
of the environmental world. Can you explain more?
How does data currently flow to regulatory agencies?
How does it move through the pipeline of development
and permitting in the country?
JESSIE: Yeah, that's a great question. I think
Becca and I will probably tag team this one. And
maybe one thing to say is I don't think it's always
intentional. It's just the nature of how our data infrastructure has been set up. And these
systems were created prior to the Internet in
a lot of cases and prior to us having really strong
data standards. So you see kind of a hodgepodge
depending on where you are in the country, who
you're working for and how you're working. But
generally what you can imagine is that we have
a baseline information that the National Wetland
Inventory provides of where wetlands are. We can
start to see that anyone has access to that information.
And then if I'm a consultant thinking about, well,
if I'm trying to do some kind of development project,
and if that development project impacts a wetland,
then I'm required to mitigate whatever impact
that I might have. Sometimes when I'm doing that
planning process, I may or may not have access
to up to date information about where wetlands are from, the prior issue of
that data being 50 years out of date. And this
is when Beck and I actually first got to know
each other several years ago, thinking about that
planning process. So then I'm doing a project.
I have some information. I haven't yet been to
the field. And then I send a consultant to go
actually survey the landscape and say, all right,
what's out there? What is my house or road or
transmission line going to impact? And then you
can get much higher quality information about
where wetlands are, where any type of stream or
other ecosystem feature might be. And that detailed
information is maybe that then is like sent back
to the company that wants to do the initial impact.
Maybe part of that shared with the Army Corps,
which Becca can speak to, and maybe this will pass to Becca, is like, what
then happens and where does that data go?
LIV: Yeah, that's right. So I've been finishing
up some research that's based on the Corps of
Engineers orum data set. So that's where they
collect all the information about permits and
also wetland and stream mitigation banks. And
we've been hearing a lot that they are taking
a long time, and you just haven't known why or
if that's true or not. So that information comes
into this big, huge data set. And then that data
set over there sometimes doesn't speak to another
data set over here. So there's the data set of
all the permits and there's a data set of wetland
and stream restoration banks. They don't necessarily
talk to each other very well. We found some inconsistencies
there. And then overall, the point I'd like to
make is that we're not seeing sort of basic project
management technology being integrated into the flow of permitting at the corporate engineers. And I'm certain that it's
probably the same at other agencies as well. So
if there's a timeline, if there's a process for
paperwork going through and the data going through,
it keeps getting hung up or stuck on a desk. And
it turns out that it's not the agency staff who
are moving the process forward. It's usually the
applicants who are calling and saying, where is
this permit? What's going on? And they're pushing
the process instead of the agency pulling the
process. And I'll just give my little fact that
our data analysis showed that it's supposed to
take 225 days to get a wetland or stream restoration
project approved, and it's taking on average over
1000 days. So it's not meeting its targets, and
everybody knows it. Now we have the data to prove
it.
CAMERON: I wanted to go a little deeper into Edna.
Is this the new big dating app? Is this the replacement
for Ancestry.com? What does it mean to me if you
could.
LIV: Sure, yeah. I definitely have a lot of knowledge
of environmental DNA because I took on a project
to do an online training about edna and so I had
all these professors talk about it but they talked
at such a high level that it was impossible to
understand, so I had to translate everything. So it means environmental DNA and that's little bits and pieces
of your body, of an animal's body that are shed
into the environment. So it's hair, saliva, fur
and doodoo and stuff like that and it gets into
the environment and actually has a long track
record in academic literature of using edna, particularly
for aquatic species to find out if they're present
or not. So some of the uses of this might be if
you're trying to track how far along an invasive
fish species is coming up the river or something,
you can kind of like put little stations and sample that e DNA. And the great thing is that the
species doesn't have to be present and you don't
have to harm the species to find out where it
is. You just scoop up a cup of water basically,
and in that cup of water is the scales and the guts and the blah,
blah, blah. Right, and then you sort of process
it using some of the same technology that we've
used to detect whether or not we have COVID. Right,
so some of that technology is used in that edna
process to say, ding, yes we have the species,
no we don't have the species. And it's even going
further than that and into something called metabarcoding,
community metabarcoding. And that's where as long
as you have sort of the map of all the different
animals and their genes and their DNA, their code
so to speak, you can sort of take a sample, scoop
up that water and not just find one species but
find a whole community of species. So it's really
exciting technology. Definitely, we've seen it
being used by federal agencies, not like this
is the only thing that we're going to use now,
but it's like another tool in the toolbox, so
to speak.
BECCA: Becca, who's primarily doing this sampling
of Edna, who's doing the sampling and analysis
of that?
LIV: I think it's mostly consultants and universities
at this point. Probably it's being driven by universities
and academics and then they graduate and they
sort of bring that technology to the field and
they've got these little backpack samplers. So
it's not like you have to just collect it and
then send it off to the lab. You can now analyze
it on the spot as well. And my older colleague
at my last job, he was making a Star Trek reference
where they had this little analyzer and they'd
go onto a new planet and they'd be like, blip
blip blip, what's living here? And we can find
out just from that. So I think that's the direction
we're heading in.
CAMERON: It also tells you if there's ghosts around.
LIV: Cold spots. Right.
CAMERON: So when we talk about permits piling
up, it takes the person that submitted the permit
to actually get it to move. And then we look at
Florida, who took over their 404 regulation, and
it went so well in New Jersey, it went so well in Michigan. And we're like, okay, cool. States
can figure out their wetlands. And Florida is
one giant wetland. I would know. I come from there.
Also afraid of snakes. Jesse so it's like, who
needs to motivate this change? Because it's like
the states seem to have it under control in some ways. Florida kind of I don't know if
they're still figuring out how to do it, if they're
developing an internal technology that seems like
it could help them move things along. But up to
this point, it seems like it hasn't gone too well.
So does it need to be federal, or do you think
it needs to be driven by states?
LIV: I think you might be waiting a while if it's
federal. Right?
CAMERON: That's true. Too well, with the IIja passed, a lot of money is
going into the permitting process, as we talked about in our last EcoBike session. So if that even factors
in, but I'd love to hear your perspective on that.
JESSIE: I think there's many pieces of this puzzle.
I think the one thing that makes sense from a
federal level and I'll just beat my tech drum,
is the digital infrastructure to be able to see
all of this information in one place and the project
management type programs that don't make sense
to replicate differently for every single state.
Because a lot of these consulting firms, one,
work across different states, so it's hard if
you have to submit something different in every
place. Two, it becomes hard to aggregate that
information. And three, we know that wetlands
don't have political boundaries. They often might
be at the intersection of two different states.
And when you think about projects like the Chesapeake
Bay that require multistate approach or even the
Gulf of Mexico, right, you have many states that
feed into this overall wetland restoration goal,
and that type of information should be made easily
accessible and of the same format across all different
states. Then each state can put their own flavor
on it. But what is the baseline of information
and quality of digital infrastructure that needs
to be there? So then all states can thrive, whether
they have the capacity to do it themselves or
not.
BECCA: That's awesome. Jesse that's helpful perspective
to sort of put everything into. I have a follow up, I guess, two parter questions. So the first part is I
also worked previously for a large scale environmental
consulting firm. What I noticed while I was doing
field surveys and just the people that I was working
with and who was training me in wetland sciences
is that I found that it was definitely an older
crowd. There wasn't a lot of young up and coming
wetland scientists that were just like taking
the reins over. So how does pushing innovation
work in a field that's still sort of stuck in
the past in a lot of ways? And how can we make
that more accessible to as many people as possible
and really make it easy for them to follow and
understand? The other part of the question sort
of is permitting reform is the only thing I've
seen really regarding environmental news apart
from environmental disasters and emergencies.
But in terms of what's up and coming with environmental
permitting and EPA reform and the IJA is all about
permitting reform and the understanding that in
order to have meet clean energy goals or meet
just energy infrastructure goals, period, we have
to have permitting reform to be able to speed
through these processes. So how do we get the
innovation into the hands of the scientists and
those that are doing the analysis and doing the
surveys? And then how do we get that survey data
pushed through permitting so that we can actually
build things?
JESSIE: This is a phenomenal question. And one
of the things that we think a lot about on the tech side
and kind of what brought me here was it's not
that we don't have innovative technology. I feel
like we do have a lot of the solutions to some
of the things that we're focused on. And it's
really a cultural shift. And when we break down
the cultural shift and our technology program
thinks about like, people process and sometimes
it's data the people is what are the skill sets
that you want to hire for in these agencies? So
who's on the other end of that? And it's hard.
As I was saying, from a technology perspective,
if the person that you're talking to is like,
I've never heard of satellite imagery because
it's a newer thing, or if you're hiring junior
level GIS staff, of which I was one, and not the
senior level positions that can think about innovation
and data strategies more holistically and set
the tone for the agency. So we think about the
leadership positions, we think about then the
staff, the general staff, but then also a third
critical point that we think about are the points
of entry for technology or data innovation. So
where partnerships offices that maybe are interagency
by nature, because my fun stat that Becca had
hers is there's 25 federal entities that manage
water data. We know that environmental agencies
have complementary and overlapping data needs
most of the time. So how do we have one partnership's office that is able to properly
vet and evaluate this innovation and make sense
for these 15 programs and that can kind of then
help us be a bit more strategic and thoughtful
about where and how to use innovation? And then
I think that the second part of that question
is right. What I notice is there's actually not
a ton of regulatory or policy change that needs to happen as much as it's just like the education and those
other aspects related to the leadership and teams
that could be helpful in sparking innovation or
using it where necessary.
LIV: And then I'll take on the sort of the streamlining permitting part of things. So that's an
area that our organization is definitely really
interested in because we think restoration should
be permitted quickly. It's environmentally beneficial.
Okay, so one of the things that we've been hearing
as we're starting to have informational, interviews
with people who are doing restoration projects,
is we keep hearing that one of the major barriers
or the things that slow down permitting has been
just the lack of resources at the US. Army Corps
of Engineers and other agencies and the turnover
and the training that they need. So I have a lot
of empathy for thinking about those staff, thinking
what they have to go through. They're probably
like the bad guy on everybody's radar. Nobody's
like yay permitting person. So I think we had
to come at it with a lot of empathy and think
about what they're going through. So that's one
of the things that keeps coming up is have more
resources and maybe some of that IIja money is
being funneled there, and that would be great.
The other thing I think is making things more
accountable, having leadership that say this is a priority, and then also introducing technology to
help out with project management. And one of the
really good example that I've seen that's come
out recently is a project that's called Peep in
Virginia, and it stands for Permitting and Enhancement
and Evaluation Platform. And what it looks like
to me is a pizza tracker, like the Domino's Pizza
Tracker. You order your pizza and it's like, okay,
very familiar. We have pizza night every Friday
at our family.
CAMERON: Now that's technology I can wrap my head
around.
JESSIE: Exactly.
LIV: I have a blog, it's called if you can track
a pizza, you can track a permit. The pizza goes
in the oven, or the permit comes on the desk.
It gets assigned to a staff member. Right. And
then it's going in the oven, and then it's ready
for you. Right? So that's kind of what this Virginia
Peep program does. And it has this really wonderful public transparency. It says the permit is here, it's with our agency,
or now it's gone out for review with the state or the Federal Fish and Wise Wildlife Agency, and now it's over
there. And in addition to that, it also has automated
emails that say this is the deadline. Hey, you're
a week out from the deadline. Have you done it
here's where you can get to the information. It
might even have sort of a cloud storage system
so that everybody's going to the same place. So
I think it's just a brilliant and it's a no brainer
kind of technology that I hope would be saving
time with the agencies as well. You're not taking
that time answering requests of where's my permit?
Where's my permit? Or even getting blamed for
delaying a permit when it's not even the agency's
fault. Maybe it is on the desk of the applicant,
but if you're not tracking that, you don't know
where it's at. So that's one of the things that
I think is really great about streamlining permitting
processes, is creating that accountability well,
through transparency, creating accountability,
but also bringing in that technology to kind of
streamline that process.
CAMERON: I guess I just want to figure out. So
you're talking about data sets like the NWI being
30 plus years old, other data sets being 50 plus
years old. What was the situation at that point
in time that allowed for such a robust data set
to be created? Like every wetland in our country,
also, I assume, 30 to 50 years ago, we did not
have the technologies that we have now. So it's
like, what was the situation that allowed it to
happen then? And now that we're in our situation,
where we have technology that could do it in a
10th of the time, let's say, and we're just not
it's just kind of blowing my mind. Like, what
caused it to happen then and why is it not happening
now?
JESSIE: Yeah, I think Becca and I can also probably
tag team this question, but.
CAMERON: You guys are like pro wrestlers. Get
them in the headlock tap.
JESSIE: But maybe just to explain this a little
bit further. Right, so we have the National Wetlands
Inventory, but the timeliness of the information
for each state or region varies. So in some parts
of the state, the information is much older than
in other parts of the state or other parts of
the US. That have been updating that information
over time. Where and how that information has
been updated really varies. I would say region
to region. And I couldn't give you at least maybe
Becca knows why certain areas are better than
others. But for example, we saw a proposal from
Minnesota where they were going to take $10 million
in ten years just to update their state's map.
You're like, that is too much money and way too
much time. Like, we need much higher quality information.
Now. What I've heard about how this information
was created in the past is literally like people looking out the side of planes and trying to map where they
were. They take aerial imagery and are manually
drawing it, depending on the types of information.
And I think the one thing that Becca and I have
also been thinking a lot about is, like, there
are dozens of wetland classifications, and the
National Wetlands Inventory wants you to have
all those classifications where sometimes, depending
on where you are in the decision making process,
you just need to know likelihood of wetland or
not. And then maybe I can avoid that entirely.
And then when I need to do field work, I can do
the much more detailed things. So I think, what
is the world in which we just have a baseline
likelihood or not? And then higher quality information supplemented by field visits and beck I don't know if you want
to add anything to that.
LIV: So you mentioned that the state of Minnesota
was it was going to spend 10 million yeah, michigan.
JESSIE: I can fact check myself, but it's okay.
LIV: Yeah. There you go. So at my last job, I was collaborating with the Chesapeake Conservancy and to use artificial intelligence, deep
learning, actually, to predict and identify where
wetlands were based on satellite data and the
different bands of light that you can see, and
you can't see along with LiDAR data, which is
sort of like point cloud data that creates, like
a 3d terrain and then some other data as well.
Imagery, I think, too. Anyway, so they took all
that data, they applied deep learning and got
a 94% accuracy rate, which is pretty amazing,
and they'd like to expand that model to the entire
Chesapeake Bay. And I want to say the price tag
on that is order of magnitude. It's around 500,000
for five states, a huge area in the MidAtlantic
region. So I think there could be tremendous cost
savings in using deep learning and newer technologies
to kind of predict where things are now that needs
training data. Right. So good data in, good data
out. Garbage data in, garbage data out. And at
the last company that I worked for, there were
a lot of electric power companies who were collecting
wetland data. And we thought, awesome, we have
this wonderful data set. We'll use this to be
a training data set for the model. And then we
found out that the data was arbitrarily cut off
by the edge of the right of way. Transmission
lines have those sort of open areas that are about
100ft or so. So we couldn't train the machine
with this, or else it would think that all wetlands
had a straight edge to it. Right. So you need
good training data in order to do this. But every
single project that needs a wetland delineation
is collecting that data. Right. And once it becomes
a finalized permit, that should be public information,
but in essence, it's not because it's probably
saved on a PDF file, and then you'd have to do
a Freedom of Information Act request to get it.
So I think a wonderful idea if we could get there
would be for all those thousands, millions of
dollars that are spent on field data and creating
that and collecting that, that goes to a federal
agency, if those could be in the public domain
in the future. So that's my wish list.
BECCA: Jesse I definitely did a lot of the likely
not likely wetland presence surveys with my experience
in consulting and. They were really helpful for
planning purposes. And a lot of times we would
use just publicly available data to do this sort
of analysis of a probability of wetland presence.
And that helped engineers get a better design
so that we would have a smaller area to actually
go survey. In my experience at university, I worked
on a couple of citizen science projects using
crowdsourced or community sourced information.
I actually used a lot of cocoa rods data, which
is community collaborative rain hail and snow
network. So I was wondering if thinking about maybe hobbyists cocoa. Raza is really popular with farmers and
people that are out there actively taking this rain data, precipitation information already. And now
they can just contribute that to other people in their
area. I was wondering if you thought that there
would be any space for that for wetlands in the
future, potentially, whether it's just not like
a likely or not likely presence, but thinking
of people like fishermen that are going out and
trudging through areas where they're going to
then go fish like they're already walking through
the area. Can we just get them to take a quick
map and upload something? Do you see any future
for that?
JESSIE: I mean, I would say yes, 100%. And I think
that gets to another aspect that we think about
on the tech team is how do you get government
agencies to be more amendable to using third party
information or community science data? And so
I would say all of that kind of goes hand in hand,
but that we have a completely underutilized resource
of people that are doing this already. And if
we have the technology to easily capture that
information on your phone and upload it to a central
repository, which we do, then the next step is
then how do you operationalize that into decision
making and, or get comfortable with that even
existing. And whether it's used for official or
unofficial purposes, we should at least have that
information in a larger repository. So, yes, and
let's figure out how.
LIV: Yeah, and I did a previous research project
on citizen science species data because a lot
of folks were coming at it from a place of concern
like what if somebody says there's a yada yada
species over here, endangered species here, just
to hold a project? And it's not true. I think
the best and most widely used citizen science
projects and platforms have some kind of check
in place to make sure that it's like, oh, you
see a zebra? I don't think so. Or you see a certain
bird at this time of year that's out of its range.
It's not normal. Are you sure? And we'll have
a real person go and check that record and ensure
that that's correct or not. So what we're seeing
in the research that I did previously was that
citizen science apps like Ebird and Inaturalist, they can be considered research grade
in some situations. And they have been used in
models for bald and golden eagles with the Fish and Wildlife Service. And I see them cited in listing decisions
for endangered species and things like that. So
I think they're getting more widely adopted. And if you can address what
is that concern and is there any safeguard for
those concerns, I think it'll just keep on moving
forward.
CAMERON: Well, great. You guys cited so many things,
and they're going straight back to your website
because you have so much amazing content there.
We'll have a bunch of links in the description
of this podcast, but please check out the Environmental
Policy Innovation Center. They are doing great
work there in writing amazing articles, and Jesse
and Becca are a part of that. And thank you so
much for joining us today and having a discussion for Ego Bytes.
LIV: Yeah, thanks for having us.
CAMERON: Of course. All right, y'all. We'll catch
you next time.
BECCA: Thanks, guys.