Landsat in Action - Monitoring Canada's Forest with Mike Wulder

Video Transcript
Download Video
Right-click and save to download

Detailed Description

Mike Wulder with Canada's Forest Service talks about the value Landsat images have for mapping and monitoring Canada's forested areas.


Date Taken:

Length: 00:02:25

Location Taken: Sioux Falls, SD, US

Video Credits

Producer: Steve Young


Mike Wulder, senior research
scientist with the Canadian

forest service.

I think the first time
we mapped the entire

country, we used maybe

area of Canada. Now we're
using say 10,000 per year.

The key thing to mention
though, I think, is the

is having the time series.
So having the year on year

information for us has
been just critical.

Whenever you do a landcover
map you're always somewhat

crestfallen because the first
question that somebody

asks is, that's a nice map,
but how has it changed?

So before when we had
these singular coverages

it was a lot harder. You
had to do all these inference

and sort of speak to the
change in a less rigorous

manner. But now we're
actually able to map out

the change, we know what's
happened where and when

and even now we even
know the disturbance type.

It's been an opportunity
for us to get just that much

more of a systematic
understanding of the 

dynamics of the forested
land base over Canada.

Anything that's historic,
it has to be Landsat.

The temporal depth, Landsat
is really the only game in town,

as far as having the initiation
in '72, with the MSS

then the '82 or '84 depending
on how you want ot look at

the TM era beginning, that is
really where we're using

for a lot of our work as the
baseline. So that means that

Landsat's really the main
source of data.

The calibrated reflectance,
for us, means that when we

build models, we can build
it on one year and we can

apply it on another year.
So we don't always have

to be thinking, aw, to bad
we didn't do it for this

particular year. We have to
go back and redo all our

models for this other time
period. but we've been able

to now, with the calibrated
data and the reflectances,

when we do our modeling
say for, again, volume

biomass heights, we can build
those using either plot data

or lidar data. Something
that gives us information

about the structure,
anything 3D, we can then

apply those models through
the whole time series

and then once we put them
together temporally, 

we can ensure that there's
some integrity to the flow

of the values through time.
So that's again with Landsat

we can do and with other
sensors we really don't

have that opportunity.