It was a good morning everyone.

You are going to be on the architect for
the first four.

We are going for
by the Georgia Tech librarian are and

are you a librarian who
are part of this series

that specially critics have put
who work here in our library and

he says most are right that there

are and I also think people who work
at the library videotape this so

if you're in pain or
there's someone you would love the pun

can play the good part which
is your life as a Tory and

you can search either for architectural
planning or Drug Bill Bratton.

Who happens to be our day we do this hour
the last Thursday of every month in which

we are going to buy the best for you and

then we're lucky to have Dr Bill Drummond
what I'm gonna I will speak on her.

Good morning.

Yes it's the Labor Party all Mormons or

all started running this circuit for
the reason

I'm going to soak it right there are there
for you right here for the fund raiser.

Would you recommend to put
it on at the moment so

I would take the first plan and run and

when I was told there was
a laying down market for

you and he also told me that

you know you can be a lot more and
I asked for.

Well you know was there a bachelor.

Are you ready for me and welcoming.

DOCTOR.

It's the research talking about today.

I've been working on for a couple years.

It's getting to the point of before
long seeing the light of day so

I would quaintly appreciate feedback and
thoughts from both from students and

faculty members in fact from
what I say students will

have a significant role in the future
because I'll be talking about a time

period over which many of you students
will have your working life times.

And so those of us on the faculty and

practice will have to
hand off the torch to you

in the next generation to see through
a fair amount of what I'm talking about.

Now.

This research actually started.

With the prior piece of work
that appeared in the most recent

Actually I think the second most
recent issue of The Journal

of the American Planning Association.

And in that piece of research.

I asked have climate action plan or

made a difference to greenhouse
gas emissions and I ran some

statistical analyses and discovered
that yes climate action plan or.

Have by producing climate action
plans of maybe reduced emissions

by half a metric ton per person.

That won't mean much to you at this
point but I'm going to try and

fill in what half a metric ton is.

But as a sideline to that research.

Control denser and

more sprawling urban areas and discovered
a pretty significant effect from

urban form on emissions and so
this is a follow up to that art of.

Well one focusing on the question of
the relationship of urban form and

greenhouse gas emissions have the two
central research questions here.

First how much higher are residential
carbon dioxide emissions

in sprawling areas compared areas
with more compact growth patterns.

So we're asking how much higher.

We will find out that they are higher and

everybody presume sprawl
area emissions are higher.

So we're going to ask how much higher.

And then number two.

We're going to ask how much could changes
an urban form reduce those emissions.

If we develop more compact
Lee If we develop smarter.

So I'm going to try and address those.

And here is the conceptual framework
Here's how things fit together.

It's not all that complicated
Up here we have planning tools

policies the interventions that
planners architects Lynskey

architects know will help produce
more compact development patterns.

Here is urban form in particular
are areas sprawling or

compact then the question.

Urban form has effects on emissions and
in particular looking at

building emissions from residences
transportation emissions from trips

at people's houses and then land cover
emissions from trees or lack of trees.

They took a beautiful old tree
down out here when they were

putting up Klaus building.

And then given the missions
the effect of the missions on

climate change global warming and
then the negative effects.

This research focuses on these two levels.

We're just looking at urban form.

And how it affects building emissions
transportation emissions and

land cover emissions at residences and
other folks do research

in these other areas but it's these middle
two layers that I'm focusing on here.

To zoom in a little bit on each area.

So if we have urban form and we're looking
at residential building emissions urban

form can affect a through number of
rooms in the house the size of the rooms

in the type of house attached
detached multifamily how

many stories in the house and
all of these have affects Sun building.

Heat and energy emissions.

We're looking at emissions from
electricity from lights in appliances.

Plus heating heating of homes
with natural gas fuel oil and

what have to use some control variables
and I'll explain basically why and

what those are and there's also
a semi wild card here in terms

of the urban heat island effect and
I'm looking out there for why I'm stunned.

My colleague the expert in urban
heat islands I don't see him.

But I'll mention him in a little
bit because urban heat islands have

pretty complicated interactive effect
with building heat and energy emissions

the hotter Atlanta is the more electricity
we use for air conditioning in the summer.

Then for transportation emissions.

We have compact or
sprawling urban form and

then urban form affects public
transit ridership we reduce or

increase our driving depending if we're
living in a compact area urban area in.

Compact areas are more walkable and
more by cable and

so people like Michael Elliot may.

They move into a place explicitly.

Well it was one of the considerations so
he could walk or

bike to campus and so the walkability or

bike ability of areas is
also a result of urban form.

All being lead to different levels
of transportation emissions

from driving from taking public transit.

Actually I didn't count any emissions for
walking or biking.

Although maybe you exhale a little more
strongly than you would if you walked and

so the result is transportation
the missions and again some control

variables will use but
third category is basically trees.

That trees store large amounts of carbon.

They pull it in each year they
store if we take them down.

We release carbon to the atmosphere.

In particular looking at
the difference in tree cover

from sprawling to suburban areas and
what that difference is

here actually sprawling areas have
an advantage they have more trains and

so sprawling areas actually take
up more carbon than urban areas.

And again.

Relation to the urban heat island
effect because trees have an effect on

temperature which can then have
an effect on electricity or

heat emissions all that we're going
to leave that for future work.

OK question number one
how do we measure sprawl

because this will be statistical analysis.

I use a sprawl index that sexually
calculated from six different

variables including gross population
density percent population and

low density areas percent high density
percent living in urbanized areas

as defined by the Census average block
size and percentage of small blocks.

I wish Doug Allen were here because
we use both block size in percentage

of small bill blocks are an indicator
are one of the contributors to sprawl.

Those are combined through
a statistical approach

using individual data at the block level.

And so
I actually got all the census data for

every single block and
then processed it for the entire country.

I actually got a nasty email.

Well wasn't that nasty from Robert and

Perry asking why I was using five gigs
of space on one of the computers.

So this is block level data that
we process with these six factors.

Statistically combine it
into one sprawl index.

Then we set five hundred and
is the mean value.

Normalize it to a standard
deviation of twenty five.

So higher values are sprawly are areas.

It's probably are is a word and
lower value so more compact.

And in fact here's the distribution.

Of local areas and

I'm only looking at metropolitan areas
because when you get out into rural areas.

It's hard to say what sprawl is we
can calculate the sprawl index but

when you have mixed agricultural
areas in small towns.

It really is not an area that we
typically think of as containing sprawl.

So these are metropolitan areas.

These are local areas and
then this is the distribution.

You can see is actually left skewed.

There is a long tail toward
the more compact side.

And says these are places like
New York and downtown Chicago and

Boston and
then these are areas like Forsyth County

don't want to insult anybody but
these are our spoke earlier areas here.

So it's not a symmetric distribution.

In the sprawl index is abstract but
we can relate it to

measures that we're more familiar with
like gross population density and so

even though there are six factors in
the sprawl index we can map sprawl

to gross population density and back and
I will toward the end of the talk.

I used the census and essay boundaries.

So basically twenty eight
counties in the Atlanta area.

I'll show you some maps of these if self

a majority of people in a local area
live in an and S. sank clue to that.

NET to give you some context.

This is from the US two thousand and
eight inventory of greenhouse gases

and this is in millions of metric tons.

Our total emissions were about seven
thousand newly and metric tons so

seven billion metric tons in the U.S.
land use up take from trays

actually took away nine hundred
forty million metric tons.

So our net emissions were
about six thousand residential

heating three forty three million
metric tons eight forty two for

residential electricity and
then gasoline for cars and trucks.

A thousand million metric tons.

These are huge numbers and so for

most of the analysis I'll talk
about metric tons per person.

If we translate these into
metric tons per person or U.S.

emissions are about twenty three
metric tons per person per year.

Now.

Of the.

If we take away land use we go down to
about twenty metric tons per person.

Resident.

Heating up one point one
metric tons per person.

Residential electricity two
point eight metric tons and

then transportation three
point five metric tons.

So altogether.

Each of us in the US
averages about seven and

a half minute trip tonnes per
person of residential emissions.

So if you keep in mind that seven
point four seven point five number.

You'll have a context.

Now to give you a little
additional context.

We don't need to have
been highlighting those.

What is a metric ton.

If we take a metric ton of carbon
dioxide extract the carbon powder it.

We get something like
flour powder graphite

which is carbon is something like flour.

So a metric ton of carbon dioxide
which in itself is invisible

odorless weightless in the atmosphere.

But if we take a metric ton of
carbon dioxide extract the carbon

we would get eighty five
five lb bags of flour.

And so
one metric ton is a pretty substantial

amount of powder and so
if we talk about a metric ton.

This is the amount of flour that

corresponds to the carbon
in one metric ton.

I'm teaching a climate change planning
class in the students just calculated

from the Georgia Tech inventory
each square foot of campus

releases ten pounds of carbon a year.

So you could imagine every square foot but
we put five pound bags of flour.

Those are our campus emissions
from buildings each year.

So this is not inconsiderable
Now when you drive one mile.

You put about a panel and the equivalent
of a pound of flour into the atmosphere.

To do.
Dr one mile.

So this is invisible odorless and
weightless.

But if we could actually see it and
feel it.

These amounts are very substantial.

So eighty seven five pound black bags for

one metric ton each of us
averages twenty per year.

So what would that be sixteen hundred
five pound if I got my math right.

So over a thousand of these a year
each of us emits into the atmosphere.

Now if we think about in practical
terms what does this translate

into if we all set out to reduce
our emissions by one metric ton

a year we would have to drive
two thousand fewer miles.

If we want to reduce our
emissions by one metric ton.

If we are flying we'd have to reduce our
travel by thirty seven hundred miles.

So actually one trip to the coast

is get you in the range of
a metric ton of emissions.

If we turn on our household electricity.

And then substitute pedal
power to run our T.V.'s

we would have to go without household
electricity for a month and

a half to reduce our
emissions by one metric ton.

Well on the positive side if we planted
twenty five tree seedlings allowed

them to grow for ten years those trees
would absorb one metric ton of carbon.

Generally speaking mature trees or
about three hundred to an acre.

So this is maybe a tenth of
an acre planted with trees

would reduce us by one metric ton.

So this is meant to give you
an idea an intuitive idea.

These amounts that we're talking about

because when they're these abstract
numbers millions of metric tons.

It's a little hard to connect those
great amounts with our everyday.

Actions like driving taking plane
trips turning on the electricity for

our three hundred watt projector there.

So this is meant to give you an idea.

Now how do we calculate local emissions

is anybody here received
the American Community Survey and

returned it to the Census Bureau
not a single if you've gotten one.

The American Community Survey is the
replacement for the Census long form and

it will be the primary source
of social economic data for

the country in the future.

Because the Census long form used to give
us all this all information is gone.

Starting with the two thousand and
ten census.

So the substitute is the American
Community Survey you guys answered

question eleven A on
the American Community Survey last month.

What was the cost of electricity for
this residence.

And then also similar questions for
natural gas for heating oil.

There's also a question about what
your commuting time to work is.

So this is the source of the data
that I'm using for the calculations.

The A.C.S. surveys one percent
of the US population each year.

It's similar to the Census long form.

All right.

Pete since we have it no more.

And a selection of these answers.

Maybe your guys are actually released the
entire questionnaire is released without

personal identifying information but
the entire questionnaire is released for

this sample that they call poems
the public use micro sample and

so every year the A.C.S.
gives us data on a million

households including their electricity
usage through this problem sample and

natural gas and commuting time so
they ask about electricity natural gas.

If you were well in the last month or
so we have this huge source of data.

A survey of everyone in the U.S.

scientifically conducted
by the Census Bureau.

They also ask how do you commute.

Do you drive.

Do you bike take transit and
your travel time.

So this is fantastic but there's a catch.

Because certain people are worried
about privacy concerns.

They only release it at a geography
level called the perma and

the Puma doesn't correspond to
anything that we're already

familiar with it's not
counties it's not cities.

The puma short for
public use my course ample area for you.

G.I.'s people.

I think Puma is related to Tiger.

It's one of the funniest G.I.'s jokes ever

the lowest level of geography for for

the Puma is one hundred thousand people so
the Census Bureau goes around and

designates areas with a minimum population
of one hundred thousand then they

release a sample of questionnaires for
each of those Plumas.

Maximum population four hundred thousand.

And so if we look at the Pumas for

our area one county can be a plume up so
up here I think that's Cherokee County.

Is its own plume up here in Dekalb County

with six hundred people more than
six hundred thousand people.

We have six individual
Pumas indicate that county.

Were or if your county doesn't
have enough population for

a prima like Athens Clark County.

It can be combined with other
counties together into a poem up.

Now the Census Bureau
also actually sometimes.

Plumas even crossed county lines and
so because they

are dedicated to protecting Lesley and
Steve's privacy and so

they are ensure that these
are large aggregations for

which they release these
individual questionnaires.

So how do we calculate emissions
then given this detailed data.

What I do I see some electricity
expenditures for each Puma and each state.

Then calculate each whom a share
of State Electricity expenditures

Pumas luckily are nested
within states then the E.P.A.

has already calculated a missions for

electricity at the state level and
natural gas and fuel oil.

So I multiply each boom a share
times the state level of emissions.

Then I repeat for natural gas and
heating oil then for transportation

they unfortunately don't ask how far you
drive but how long you drive to work.

I used commuting time as a proxy for
the distance and

sat then use commuting time for
each transportation mode.

And then the overall we're
cutting off partly here Leslie.

What we come up with.

Is when we add together these
a mission is seven point four metric

tons per person of residential
based emissions but

once we go through this process
with the prune in the A.C.S.

we can then distribute those emissions
within these individual small

areas with a minimum of one
hundred thousand in population.

New trees are a problem because
the Census Bureau doesn't ask you

anything about how many
trees there are on your lot.

But fortunately the remote
sensing geeks have come.

Hi old.

Coverage of the in a cloud free coverage
of the entire United States classifying

every grid cell which is thirty
meters by thirty meters so

roughly the size maybe it's not a football
field or so and so they've classified

the Lin cover all never a grid cell
including lots of categories of forest.

So I used this total United States.

Land cover map identified all the cells
with Foster tree cover calculated those

within each Puma although I'll mention
I only looked at the urbanized areas.

So I basically discarded the undeveloped
areas because it probably isn't

fair to count the trees of areas that
are sitting out there waiting for

development I only did
the developed areas.

Then multiplied each Puma share and
the E.P.A. has already calculated

Trini uptake at the state level and
so I've broken up those E.P.A.

numbers and then distributed them
to each individual put them up.

OK The fun part.

Some maps.

Here is the sprawl index for Atlanta and

what I've done is broken up each
of these Pumas into ten categories

the most compact is the most
compact ten percent.

In the Country of the PUMA So
the most compact ten percent Pumas

all the dense urban areas
Atlanta does not have a single

puma in the most dense category
of the US that there are.

Elsewhere.

So there's none in the most
dense the red are the ten

percent of most sprawling pumas and
we've got a lot of those and

these boundaries are the M.S.A.
boundaries George so

this gives you an idea of the geographic
area that we're looking at.

So the red areas are the most sprawl.

The blue are the most compact and

Atlanta does not have a single area
in the ten percent most compact.

Now if we look at building
emissions per person and

using a same distribution in Kwan tiles.

The lowest ten percent are in bloom.

The highest ten percent are in red.

Actually our building emissions
are not bad in the Atlanta metro area

because it's not all
that cold in the winter.

For the most part.

So our building emissions
are not that bad.

You can see we don't have any
areas in red or the orange and

in general they're lower in
the in areas but then they go

up to the outer areas the highest category
is actually that inner circle there.

Scrolled a little too much on here
you can see these yellow runs

are actually the highest category of
building emissions are not the ones

all the way out at the periphery.

They're actually a little higher than
these ones at the periphery so in Atlanta.

The highest building emissions
tend to be out Northern Gwinnett

northern Cobb I won't go into
the reasons for that but.

If we look at transportation
a missions per person.

For the Plumas we actually have a run in

the very low category here and
that actually is southwest Atlanta.

And that's probably partly because
of the available of mass transit.

But probably What's the other reason
that Southwest Atlanta is very low in

transportation emissions lack
of cars related to income and

so emissions is not just
a factor of where you live but

income and we'll talk about
other factors as well too.

And then as we go out
the very highest emissions.

In these counties these Pumas out here

we're an urban developing for
a niche kind of north going net.

Down to Fayette and places like that.

Land cover up to date from three S.
where it.

These are the areas where the most uptake
these are the least not too surprisingly

our urban developed areas
don't have as many trainees.

And so for these areas there's less carbon
pulled out of the atmosphere by trees.

So here's where suburbs
actually have an advantage.

And if we put this together we get
total emissions per person and

these areas actually around.

I twenty eight and south are the lowest
buildings transportation and trays and

we see these areas kind of at
a ring out in the suburban

areas are the highest So the general
shape of the nations is like a donut.

That's low in the urban center.

It increases going out into the suburbs
and then actually should get out to

the periphery things start getting more
world and they start decreasing again.

So if we were looking at this and

three dimensions it would be
something like a donut shape.

Now to compare a couple other
areas of course I've done this for

the whole country.

But here is Baltimore Washington.

And Washington has whom is in
the most dense category bump them or

has put them is in
the most dense category.

And so those two cities.

In their densest areas
are considerably denser than

Baltimore County up here
northern Baltimore County.

Comes out pretty high
in the sprawl index and

then if you go from D.C.
we get areas that are higher.

On the spiral index.

This is total emissions per person for
the same area.

Again using the power seen near low

downtown Baltimore and Washington and

then going up with their suburbs are not
as high as Atlanta suburbs here.

Total nation by the way these
are the same scale as well do so.

And here is the Boston area and

the Boston area has roughly
the same population as Atlanta.

About four million or so and
here the core of downtown

Boston was eight in the densest category.

Whereas Atlanta has not a single one.

And you can see if you go up far enough
you start getting into Raleigh here areas

but within the black area of the N.S.A.
itself.

There's not a single.

That's in the highest.

Category in Boston.

So even though the Boston N.S.A.
has roughly the same population.

It's a much more compact urban core and

if we look at total emissions for
Boston we can see a very low

in the still low one coming
out here that starts getting

a little mixed as you come out and
not really to start getting out of state.

Yet in some of these areas
that are the orange or

red with a higher level of measure.

So this is you a feel for how the missions
were all looking at these layoffs.

We haven't you would have said is that
there clearly is a relationship between

urban form and emissions.

But there are some complications
that if we just look at those maps

the skeptic to come back and say Well I
think there's a different explanation.

It's not urban or
misleading the lower emissions.

So the skeptic might say these
deaths places tend to be

in New England the middle Atlantic and
Midwest those areas.

The sprawl areas tend
to be in the sound and

so if we're looking at
electricity consumption for

air conditioning the northeast
in the Midwest which uses less

electricity or air conditioning
on their emissions are lower.

Anyway.
And

that's where these compact areas
are concentrated whereas sprawl areas

tend to be on include
Texas here Arizona Nevada.

These sprawling areas are in areas that.

Use more electricity so

what we think is a difference in urban
form is really a difference in Klein.

So as a researcher as researchers we
have to control our climate we can't

just look directly at the relationship
that we have to control over climate in

a similar way different areas of
the country different levels of carbon

in their electricity in new ways and
they tend to have more nuclear.

And that hydro electric and.

Telling the architect here.

They tend to have hydro
in the Pacific Northwest.

And so here in the Senate.

We tend to have a lot of coal.

So our electricity is naturally
much dirtier in carbon the earth.

Than electricity in places
with nuclear hydro.

And it might happen that places are all
more also get dirtier light for

us so we also have to control for
the carbon content of electricity so

that we can use these different events so.

Last one.

I'll mention even though we have
to control for a bunch of others.

Isn't because.

It may happen.

That poor people tend to
live in the urban core and

rich people tend to live in the suburbs.

If we don't control over income.

We might think that
the lower emissions for

urban areas are actually do the more
poor people living there so

we have also happened for over in brown so
on giving you some examples where actually

maybe a dozen other variables for these
different analyses we have control for.

But I'm explaining why.

Now one last note before we
actually get to the results of

the straightforward way to do this would
just be to take the sprawl and that and

use it to predict emissions
with the control variables.

Well for reasons that I'll
explain it a little that.

I've taken a slightly more
complicated approach and

rode all the nineteen
thirty of us into ten rooms

of one hundred ninety
three in each group and

predict the emissions for
each of these ten groups separately.

If I did it with traditional
linear regression it would force

a certain form here and
that allows each group to vary.

So I can look and see if the relationship
between the sprawling nice and

urban form is linear not linear.

So I'm breaking up all of whom listen to
these ten groups the same coloring I used

on the map already ten different
groups of the most compact the most

broad technical name for
these were the best aisles because we're

breaking up the groups and
it turns all right results.

OK Those of you first
class credit you will need

to understand this table completely for
Harley's class.

No actually I'm just
doing this to show you.

I've actually run their regressions and

I've gotten lots of
statistical results here.

I'm not going to go into
all the old variables but

there's a whole bunch of control
variables that are that.

And these are the ones that really
count here and to show it more clearly.

Once we've controlled for
in carbon content of electricity climate

all these other variables in we've
asked What is the effect of bourbon for

this is what the graph looks like
where this is the most compact

and this is the most thorough so
compact sprawl.

What we see is there's a piece
of building a nation and

we're just looking at building first that
the sprawl area has about half of that for

a ton higher and you will emissions
per person than the most compact area.

So urban for
at the maximum produces an effect

of half a minute return per person
on residential building nations.

Now there's something unusual.

Once you start getting out from this
high area out into the bridge building

nations start building then I won't
get into the reasons for that but

I suspect it's House side probable.

But there is also something
unusual because we can see from

this high level of sprawl as you move
into the urban core emissions go

down the here and
then they start going up again.

Once we get into the central
part of the urban core and

then anybody make a guess
on why building emissions

may start rising again as we get
into the central part of cities.

OK lack of trees were.

Just look at buildings here.

Urban Heat Island Effect and my hypothesis
is that we made the same of the urban

heat island effect here where once we
start getting into the core of the land.

It's hotter than out in
the suburbs of the world areas.

So we burn more electricity all ourselves
in the summer so emissions start rising.

Actually I tried this out on one stone
our urban heat island effect and

first he said Yeah it looks
like urban heat island.

I should have stopped there.

But then he said what high rise buildings
use more energy for elevators and

there is wind to start as ratios and
then he started complicating it.

So I said I think I'll stick
with urban heat island for

now but there's more to the story
than that and I can't prove that

this is the urban heat island effect but
I suspect that it might be.

You know you know in the low low.

You know and
remember we're just looking at

residences here but
not openable windows and things like that.

So you know it's probably
more complicated but

I would still like to think that's
partly the urban heat island effect and

we can see if we tried to
fit a lot into this that.

It would be tough to fit a line into
this that and that's why I'm not using

traditional linear regression
because linear regression.

We're trying more or less to life
to this data and there are more

complicated things going on here that
you can represent with a straight line.

So that's why I divided
up into these groups so

that I could look at each group individual
and luckily I've done plenty of them.

So I can do that now friends for

patient of nation and
again I'm not going to go into a lot.

I will say I should have
said the building model

actually explained eighty
seven percent of the variation

in whom a building a nation
that's a pretty high are square.

You know we're still thirteen
percent unexplained.

But this whole set of variables explained
eighty seven percent of the variation.

You know a lot of that is viewed in a lot
of due to lack of carbon content of

electricity but this is a very good
model terms of overall explanatory

transportation model explains about

half of the variation in transportation
of nations for a moment and

again I'm not going to go into all that
intro variables jump of the story.

Now if we go from contact
because we're all

this is what transportation
emissions of life and we can see

beyond the very first death style
a line what does that super bad way.

Although at the very end and

the safe range transportation
emissions start on off a little bit.

Probably because at the very edge
people start working locally.

Instead of commuting into the N.S.A..

But out here are where the highest
emissions are we can also see something

of a threshold that public trains to
support a bowler's or densities and

so there's actually a big jump here
between the most compact ten percent of

sprawl areas and then what I call
the very compact not quite as

compact with still pretty compact this
difference is a metric ton per person.

So that's a big difference from the very
most dense places with pretty dense

places and it's the most dense
places that land has enough

by the way of these ones where
transportation emissions are very low.

No just the scale here we're talking in
the range of one to two metric tons per

person compared to the most dense
areas of France nation emissions and

so the difference and for explication of
nation building emissions was to happen

that for some of the biggest difference
here we're talking two metric tons.

So the effect of urban
form of transportation

was significantly larger in
the building that then finally for

land cover are squared
was twenty four six.

So I was explaining about half
the variation in three carbon

up today and I won't go into
all that intro variables again.

But here's what the results look like

where really just good here
higher levels are better.

The further you go out in the core and
that's pretty close to linear that

the more carbon uptake we see tree but

also notice the scale here we're talking
point two metric tons per person.

And so suburbs are better in
terms of trees but the effect is.

That of buildings.

Maybe ten percent of
the effect on transportation.

So the effect is not nearly as big.

As.

More so.

One reason or.

Right.
And

in fact this is a static
cross-sectional model.

We're not looking at last
three carbon element or

disturbing of soil organic carbon
her look at me at one point.

Of the car and up a tree and
it was a different and

more complicated model to
model the process a version of

the natural environment to build cars and
the car losses that come from the.

ORES.

Very well and the complication is
I mean our old air was far and

all the trees and

Western nurse were a lot more efficient
at it as second growth came back and

so we see of our tree is
really growth forest.

North Georgia agriculture.

That's really why we are in
a metropolitan area lives of three.

So it turns out to be complicated.

So I kind of just taking the step you stay
but we would like to have it be a more

dynamic model where we actually penalize
development but we have very urban areas.

Suburban areas but
that's not what brought you here but

you are right this is
looking at one point.

I think.

Be able to give.

Better from nation on that than I could.

I couldn't tell her almost
all of it is through trees.

Round vegetation actually trees and
soil are major so

hard that we use our educational
ground as so but it's a pretty small.

Fact you know there's five leave
that to rise and we're not

looking at here is things like right for
fitting buildings for energy efficiency.

We're just trying to isolate
the effect of one one car.

So if we put these together now and
what I'm doing

here is actually putting these across
the sprawl in that at the bottom.

This is the overall result of the beds and
then when we get out here to the bridge.

The difference is about for
tons per person.

Overall the clean densest and

the least bit as if we add
the three effects together

a negative effect or
come here to the densest place for cover.

Here is the building of back and then the
red effect is the transportation of that.

So you can see it's dominated
by transportation emissions

buildings at or subtract sun and
then land cover.

Has some of that and so this is all
a comparison of the densest ten

percent of the areas of which we don't.

Now a slightly simpler way maybe
a fair way to think of this.

Let's just divide all of these areas
into compact areas and smaller.

Arbitrarily say that compact
fifty percent contact

the sprawl fifty percent or sprawl.

If we compare the emissions that way

without looking at
the strains of suburban.

And we just compare sprawl versus
compact the difference in building

emissions first for all in your areas
is only three metric tons per person.

For transportation forty eight metric
tons per person minus point zero six for

land and
about one point one metric tons per person

this number is probably a fair
number keep in mind one metric ton

one point two metric tons because
it's not looking he exploring

a compact versus broad but
is looking at everywhere together and

we designate areas either come back there
are areas or sprawl areas and we asked.

What is the difference of emissions
from fifty percent of compact areas and

we see here.

How much higher are sprawling Nations.

With compact emissions that we just
find small and compact that well.

So at the age range of metric tons

of middle level one point
two metric tons per person.

OK that's the first
question of then a shot.

This will take nearly as long.

If we develop more compact really how
much would emissions have been reduced.

And so

I'm going to make a couple assumptions
here again this is a spat a analysis.

I'm not looking in the future.

But let's assume that ten percent
most contact areas stay that way.

That's our maximum density.

We'll take the other vessels and
or push or shove it.

Pending wish for you say.

This.

Program will push him into the next
quantum how and so the lowest ten percent

then become the next low as ten percent
will move every one of quantum tile and

say and the first one tile
the urban one will keep the same.

Then what will do is work out delayed
the overall percentage increase and

density and
percentage were duction in a nation.

So by pushing our suburban areas up so

that every area is somewhat more dense and
compact.

Then we see an action and so
the result is something like this and

I'm giving this to you in a more
understandable persons per square mile.

And here is per cent reduction
in emissions that if

we push your show or
nudge our areas that higher density.

Alter Net result is basically for
a ten per cent increase in

density we get a two point two
reduction in emissions holding constant

everything else and so forth.

Contest among us we can think of this
as the emissions elasticity of sprawl.

So if we increase the S.P. by ten percent.

We reduce emissions by two point
two percent holding constant

all these other variables.

We increase tensity by twenty five percent
and missions reduce by five point five.

If we double density by
one hundred percent.

If our metropolitan areas were
developed twice as densely

emissions would be reduced
by twenty percent.

And here is a question for thought and

discussion of our students are going
to be practicing for the next.

Forty fifty years or so.

In station.

Life.

Over that period.

What do you think is the feasible
increase in men for college and

density assuming not being wildly
optimistic but perhaps being optimistic.

How much do you think over the next fifty

years forty fifty years our
students will be working.

How much do you think it is we may
be able to reduce our deficit.

How many people would say ten percent or
less.

Now let's let the skeptics here and
then ask you by percents ten percent or

less you think is these
of all yesterday and

you think we might through reform
be able to reduce emissions.

One of jeepers Well now they think we
might be able to increase density by ten

twenty five percent.

Everybody has to hear is everybody.

OK I see some faculty wait over here.

Ten twenty five.

On this twenty five would be able through
urban form to reduce emissions by five

percent and
I think it may be feasible we're getting

pretty sensitive by fifty percent
of that problem and there.

OK we're getting into the moderately
up in this that area.

You guys think we could reduce
emissions by eleven percent.

How many think we might be
able to get in the range of

percent to a doubling of urban density.

Interesting.

I don't know if I could have predicted
it would be in one category here.

So severe up in this thing we
may be able to reduce emissions.

You know when they're ready just ten to
twenty percent by a shifting urban for

the higher density.

So that's what this research shows now.

So to answer questions.

What's the current cost of
suburban sprawl is substantial.

And our average is a lot
of one point two that.

Close out of our seven point or
so sixteen percent of residential

admissions using our lower
average number of one point two

we could describe the differences in
urban core of that seventy five percent

transportation thirty percent
buildings minus five percent to land

is roughly the breakdown there so
the next question.

We actually live kind of discussed is this
reduction is substantial enough to be

worth pursuing is a new planning
architecture landscape architecture goal.

I would answer that.

Yes then how much can we reduce
emissions by fighting sprawl.

Not enough because the experts say we need
to reduce emissions by eighty percent

eighty five percent stabilizes climate so
urban form won't solve the problem.

But we can make a contribution.

If we can double density or
increase it by fifty percent.

So.

The question that I've asked have
been how feasible is a doubling over

the next fifty years you can see there is
a broad range of opinion and it's not that

students are often mistaken faculty or
not but looking out of their hands.

There is a range among students and
battle.

So why don't policy recommendations.

To reduce the nation's we should
I'm saying this is a work in nation

martial all of our transportation lane and
building tools that we know

to fight sprawl that bring
down the nation's along.

Many of the reasons that we have for
However and

this is directed in
particular the students.

So badly need a new set of innovative and

creative approaches to
both reshaping urban.

More and reducing inventions and
it's really your working life

time that you're going to come up with and
then put these into practice.

So this is a particular challenge
to the students will be doing.

That as a Monday but
still important first step.

We should make require.

Carbon accounting and estimation of
carbon tax as a required element

of every transportation plan and
every way and

this is started happening in states
like California and Florida.

That would be undone large.

But this should be a worldwide element
of every transportation plan and

every laid plan.

If we can do this and

if you guys can be you students can be
successful over your life and careers.

We will end up living in golden
cities that are powered by elegantly

designed by architects alone and
near as long coastline.

And we will have taken a big
step more toward sustainable.

OK I don't know how
much time we have left.

Do we have any left questions or
this question.

Michael.

We're actually two thousand
by two thousand and

eight it was four separate years.

Well I thought.

Well you can think of one
hundred thousand people so

draw them so you have a minimum of one
hundred thousand people problems they

don't really correspond to what we think
of as neighborhoods or you know anything

like that although that tries to draw
those of a relatively low margin.

You know they try and get them so

that they are similar within
themselves but different for.

But if you get a hundred thousand people.

So the past now only has six hours for
me so that.

They don't count down but not always so

a lot of the challenge
here I calculate all my.

Facts all the holes I didn't go into but
luckily there are lots of data for

all age of the House people
in the house all that stuff

is yours.

It's tough projections about
one hundred million people

over the next four years
that the US population and.

Only Nations

one hundred places those
places of work say.

So all of these calculations
are her capital ones.

And so on per capita we have a population.

If there are attacks among us.

I hope you're working on some hassle.

For instance.

Errors and stuff like that.

So there's no question.

It's been.

Well partly as well.

The.

Carbon or.

Snippers.

Mixed metaphors there.

And that living here or
will they be living out in charity.

And there was a certain
extent as a planner and

but it's also architects well but

it's also personal preference
mortgage tax deduction and.

Other factors like that people
have played with that and

there is a National Research Council
recent study on France for

Keisha that just tried to look at
different scenarios for different places.

Those were the speculative what I try
to do here is take on what we know now.

And as that as a basis for
what would happen.

So people of tried to do that but
first yeah answer the question where

were those hundred thousand and
what were those houses.

There's

a little heart because we're at one
hundred thousand people here is a low

sweet and my suspicion is this is as low
as you'll ever see in a national study.

Unless we start systematically
gathering information

required by some piece of
legislation like Waxman Markey or

other pieces of legislation
now seen actively.

There's a door there.

So.

I can cut this down individual levels and
there are calculators out there.

I should say California is
really in terms of electricity.

Their personal At first the emissions
have been level since ninety seventy or

so when the whole rest of the day then
increasing pretty substantially and

so in a lot of ways we do have
a lot to learn from California

electricity I don't know I don't
know transportation of nations.

I doubt they have.

Status no electricity or

arrived yet.

Well.

There actually is moderately
there's controversy or has or

plans of the level of
the nation's reduction

by shifting over for
now University of Utah.

That's pretty optimistic about
reducing this transportation and

overseas study actually Mike Myers was
one of the first was much more pessimist.

This actually was slightly somewhat
on the up in the stick side for

reduction but generally but this stuff
actually looks at transportation buildings

transportation people are just
looking at transportation.

So I would say that these
numbers reasonable and

maybe more positive than I would expect
about the effect of urban or nation.

I think overall we expected
it to be lower but

I think the question is what is the carbon
footprint of dollars versus further.

What is controlling for all these other
factors like the carbon content of life.

Presently.

And and

those differences are even I suspect
if you don't control for those things.

What is there a difference is
more modest let me say this.

If we add together that one point
three pairs of the seven or.

That on average urban
dwellers have fifteen percent

less carbon emissions because urban areas.

Right.
But the question is are we going to her or

something like that or

an extreme deference was to metric tonnes
of I'd say more typical is one that.

It depends when you say carbon footprint.

Are you just measuring emissions.

And urban dwellers may be significantly.

Lower but they also because
they tend to be less suburban.

It's a little more complicated
if you're rich or.

You have more choices about where
you can live closer to work as well.

So it's in a little bit more.

So I would say I think

typical carbon footprint just add up
the emissions without trolling for

these other factors can be a little
misleading and overstate their

reasons for.

Yeah yeah.

There's so I'm wondering

whether or not.

Yeah I actually had enough time shell of
a similar analyses controlling for N.S.A..

So we just pluses and
minuses with the N.S.A.

this and within and between and
there's a difference.

And then say it's probably less than.

You face the facts and say.

One way to think of this is
just saying let's just compare

the essay say take an average
with an M.S.A. a little higher.

And so that in effect is moving back and

forth and
say what I looked at was that difference.

Plus say comparing plans of the Boston.

You know are the same.

Within and
the same numbers are similar but so.

The difference is the difference.

Boston.

And one for us are differences between so
with that last boss.

So here is roughly.

Well here is another irony of where

should people be living in the West
to have the lowest overall emissions.

Apart from why.

California climate is moderate less
heating much cooling coastal California.

Their electricity emissions are lower
not so great but not so bad.

From a sustainability point of view the
encouraging people to live in California

from the missions more I see
Steve saying over there.

There is a very good one.

All right with what you already was.

But there are missions are still lower yet

the personal effects were trying
to save the planet yours.

So you know we ought to be living
in places where lower emissions.

You know from it's not worse.

The only separation but so

if we did people between metropolitan
areas like from Atlanta to Boston or

Atlanta or California we get it
pretty significant reductions.

And

well you can think of two
sets of policy redirect where

people live by sense prices and then where
they live within the metropolitan area.

And so I take your point that it may be

overestimating because we're
also say that moving from

suburban Boston to urban land for

days whereas the policy from

the metro area are much more
local you know really at

the regional level regional
press Haitian if we HAD A or C.

is here we have regional
authority legislative redirecting

where people are I mean they
are City Little center.

And Mission places trying to
redirect people but how successful.

That's been so
thanks that's of the point in fact I now

have another chart overall and
then they're followed as well.

Wife.

You know let's say detached
versus attached housing.

That actually consider that a verb for and

not control for it in here because I
want that variation between types of

housing being rolled into
the urban form variables.

I have done some have
to now sees to try and

separate out the effect type
of housing size house size.

They haven't worked as well
as I had hoped at this point

they have this distinctly significant and

I'm still messing with to try and
pull out.

Those different half ways and
see how each one.

I would suspect that house.

Size is probably the most
important variable that how urban

form of vaccinations then maybe
house tie and maybe room size

would use the visit.

The word is what is

give us just a bit example.

Well it seems like the big difference
in all this from Boston to Atlanta.

The movement that was there and
I was one of those.

OK So the question of scale then there is.

In the size of the metropolitan area.

I think there certainly are especially
in terms of public transit.

It's pretty clear the threshold for.

Once you or if you put in place system.

This time of the practices.

There are interactions
are more complicated

here but I think there
certainly are threshold that.

Explicitly say look at land prices for
if we

think of land prices as reflection or
less risk.

Yes.

Or the role

so you know this is

the goal

of mine.

The problem is pretty abstract.

What is five four.

I think the other
complication with the waist

down really those variables
are all scale and so

you calculate you know block
size you need enough of an area

to get a logs population the trouble
is if you start getting too small.

Like in a census track area and
thousand people or so.

You might get a purely residential track.

With all residences and
then next track has a lot of commercial.

And when you calculate the sprawl
index population density.

Once you start excluding nonresidential
land uses it starts getting problematic.

So the question would be.

Can you logically calculate the sprawl
index for census tracks as census or

development as development
doesn't start losing

people before it's been a lot
better because it's been a life.

This is the first time that it's really
been the life sucking only area.

And my feeling is it probably works so
people are going

to have a mixture of land use but once
you start getting smaller and smaller.

You start dealing with population density.

Rose population.

Yes So
you start comparing apples to oranges.

You know residential mostly residential or
sort of those.

So that's an interesting question and
I really haven't seen people

wrestle with the question At what
level of this we're all disappear.

What level you stop measuring and
you have to resort to

the index did have things like him highly
dense areas and heavily very low density.

So those sort of but
I don't know nothing particularly comes

to take one more Eleanor.

I haven't yet but
what I should be able to do.

I've thought about is
this aggregate these up.

Well what we're doing is
essentially comparing

Atlanta to Wasilla as metropolitan
areas calculated sprawl and

that sort of follows in areas
where the land always comes out.

Usually in the top three but almost always
in the top five other areas bounce around

but we always come out as one
of the sprawl is Metropolitan So

that's on the agenda and
it should be given troll variables and

other things of the US a level that's
kind of a different problem and some ways

say fine about that for

for a while I think you know if
there are those broad measure I did

it myself but there's a series of
more using so they see it out there.

In the one and then my.

That's too much work.

So I'd be interested in there are measures
of that kind of thing might be a or.