Decarbonizing Aviation

UCSD Deep Decarbonization InitiativeUniversity of California San Diego

Wednesday November 6, 2019

Andreas W. SchäferAir Transportation Systems Laboratory (www.ATSLab.org)

University College London(a.schafer@ucl.ac.uk)

• Strong demand growth• Dependency on high-energy content fuels• High capital intensity• Long lifetime of equipage• Comparatively low profitability• Cost pressure has resulted in significant fuel efficiency improvements• Non-CO2 climate warming ≧ CO2 warming • No silver bullet

Key Sector Characteristics

Spot the Difference: Five Decades of B737

By Ju

lien.

scav

ini-

Own

wor

k, C

C BY

-SA

3.0,

ht

tps:/

/com

mon

s.wik

imed

ia.o

rg/w

/inde

x.ph

p?cu

rid=1

7180

722

ClassicNext GenerationMAX

Mor

rell

P., D

ray

L.M

., 20

09. E

nviro

nmen

tal a

spec

ts o

f fle

ettu

rnov

er, r

etire

men

t and

life

cycle

, Fin

al R

epor

t, OM

EGA.

Long Technology Lifetime: Narrowbodies

Half of all aircraft introduced today will still operate in 2050

Driver of Fuel Efficiency: Low Airline Profit Margins

https://www.iata.org/publications/economics/Reports/chart-of-the-week/Chart-of-the-week-7-June-19.pdf

US Energy Information Administration. U.S. Kerosene-Type Jet Fuel Retail Sales by Refiners, Average Retail Prices of Electricity

CO2 is not the only Warming Agent

Le

e D

.S.,

Fa

he

y D

.W.,

Fo

rste

r P.M

., N

ew

ton

P.J

., W

it R

.C.N

., L

im L

.L.,

Ow

en

B.,

Sa

use

nR

., 2

00

9.

Av

iati

on

an

d G

lob

al

Cli

ma

te C

ha

ng

e i

n t

he

21

st

Ce

ntu

ry,

Atm

osp

he

ric E

nv

iro

nm

en

t 4

3(2

2–

23

):3

52

0-3

53

7.

Dray

L.M

., AT

SLab

; dat

a: S

abre

, 201

7; IC

AO, 2

018

Demand Growth 2000-2016

Dray

L.M

., AT

SLab

; dat

a: S

abre

, 201

7; IC

AO, 2

018

Projected Demand at around 2030

Dray

L.M

., AT

SLab

; dat

a: S

abre

, 201

7; IC

AO, 2

018

Projected Demand at around 2050

CO2 Emissions Identity

CO2 = CO2

EE

RTK RTK

Air Transport Demand

EnergyIntensity

FuelComposition

CO2 Emissions

+5.4%/yr-3.2%/yr0+2.2%/yr1980-2015:+4.2%/yr-1.6%/yr0+2.6%/yr2015-2050:

CO2 Emissions Identity

CO2 = CO2

EE

RTK RTK

Air Transport Demand

EnergyIntensity

FuelComposition

CO2 Emissions

Energy intensity across modes: square cube law

Aircraft(US: 1991-2010)

Trucks(8 countries:1940-2010)

Railways(6 countries:1980-2010)

Gucwa M., Schäfer A., 2013. The impact of scale on energy intensity in freighttransportation, Transportation Research Part D: Transport and Environment, August.

Freight Transportation

Water Vessels

Schäfer A., Yeh S., A Holistic Perspective on Passenger Travel Energy and Greenhouse Gas-Intensities, under review.

Water Vessels

(3 sources: current

technology)

Passenger Transportation

Declining Energy Intensity

Le

e J

.J.,

Lu

ka

ch

ko

S.P

., W

ait

zI.

A.,

Sch

äfe

rA

., 2

00

1,

“H

isto

ric

al

an

d F

utu

re

Tre

nd

s

in A

ircra

ft P

erfo

rm

an

ce

, C

ost,

an

d E

mis

sio

ns”,

An

nu

al

Re

vie

w o

f E

ne

rg

y a

nd

th

e

En

vir

on

me

nt 2

00

1,

26

: 1

67

-2

00

.

Scha

fer,

Evan

s, Re

ynol

ds, D

ray,

2015

. Co

sts o

f miti

gatin

g CO

2Em

issio

ns

from

Pas

seng

er A

ircra

ft, N

atur

e Cl

imat

e Ch

ange

, Dec

embe

r 201

5.

Marginal Abatement Costs: US Narrowbody Aircraft Fleet

First-order Fleet Impact

Schäfer A.W., Evans A.D., Reynolds T.G., Dray L., 2016. Costs of mitigating CO2 emissions from passenger aircraft, Nature Climate Change 6:412-418.

LocalEnvironment

Impacts

Local/NationalEconomicImpacts

GlobalEnvironment

Impacts

AircraftTechnology & Cost

AircraftMovement

Airline & Airport Activity

Air Transport Demand

Global Climate

Air Quality & Noise

Regional Economics

Modelling Aviation Systems

www.ATSlab.org

Modelling Aviation Systems

www.ATSlab.org

Future Aviation in a Non-Disruptive World: Key Model Inputs

(a) Population

0.6

0.8

1.0

1.2

1.4

Popu

latio

n,ra

tio w

ith 2

015

SSP1SSP2SSP4

(b) GDP per capita

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Cons

tant−p

rice

MER

GDP

per c

apita

, rat

io w

ith 2

015

DECC LowDECC MidDECC High

(c) Oil price

0

50

100

150

200

250

300

Oil p

rice,

year

201

5 do

llars

/bbl

2000

2010

2020

2030

2040

2050

2060

Year

Level 1Level 2Level 3

(d) Carbon price

0

50

100

150

200

Carb

on p

rice,

yea

r20

15 d

olla

rs/tC

O2

2000

2010

2020

2030

2040

2050

2060

Year Dray

L., S

chäf

erA.

W.,

Al Z

ayat

K., 2

018.

“Th

e Gl

obal

Pot

entia

l for

CO 2

Emiss

ions

Red

uctio

n fro

m Je

t Eng

ine

Pass

enge

r Airc

raft

”, Tr

ansp

. Res

. Re

c., 2

672(

23),

40-5

1.§

Future Aviation in a Non-Disruptive World: Technology Dynamics

20

40

60

80

100

120

Flee

t, th

ousa

nd

SSP1

Pessimistic Central Optimistic

20

40

60

80

100

120

Flee

t, th

ousa

nd

SSP2

20

40

60

80

100

120

Flee

t, th

ousa

nd

2000

2010

2020

2030

2040

2050

2060

Year

SSP4

2000

2010

2020

2030

2040

2050

2060

Year

2000

2010

2020

2030

2040

2050

2060

Year

Current (Jet)NEO (Jet)NextGen (Jet)

FutureGen (Jet)NextGen (OR)BWB

TurbopropAdv. Turboprop

20

40

60

80

100

120

Flee

t, th

ousa

nd

SSP1

Pessimistic Central Optimistic

20

40

60

80

100

120

Flee

t, th

ousa

nd

SSP2

20

40

60

80

100

120

Flee

t, th

ousa

nd

2000

2010

2020

2030

2040

2050

2060

Year

SSP4

2000

2010

2020

2030

2040

2050

2060

Year

2000

2010

2020

2030

2040

2050

2060

Year

Current (Jet)NEO (Jet)NextGen (Jet)

FutureGen (Jet)NextGen (OR)BWB

TurbopropAdv. Turboprop

Technology Assumptions

Dray

L., S

chäf

erA.

W.,

Al Z

ayat

K., 2

018.

“Th

e Gl

obal

Pot

entia

l for

CO 2

Emiss

ions

Red

uctio

n fro

m Je

t Eng

ine

Pass

enge

r Airc

raft

”, Tr

ansp

. Res

. Re

c., 2

672(

23),

40-5

1.§

Future Aviation in a Non-Disruptive World: Key Model Outputs

Without biomass-to-liquids With biomass-to-liquids

Dray L., Schäfer A.W., Al Zayat K., 2018. “The Global Potential for CO2 Emissions Reduction from Jet Engine Passenger Aircraft”, Transp. Res. Rec., 2672(23), 40-51.

Industry Emissions Reduction Roadmap

Font

aP.,

Pus

hing

the

Tech

nolo

gy E

nvel

ope,

ICAO

Env

ironm

enta

l Rep

ort 2

010.

CO2 Emissions Identity

CO2 = CO2

EE

RTK RTK

Air Transport Demand

EnergyIntensity

FuelComposition

CO2 Emissions

Schäfer A., Sweeney J., Draft Paper, 2016

Fuel Shares: Yard, Passenger, and Freight Sector CO2 Intensity: Freight Railroads

Disruptive Technology Change in Rail Transportation

Kram

mer

P., 2

017.

Sim

ulat

ing

the

Impl

icat

ions

of E

miss

ions

Tra

ding

Sch

emes

for

Inte

rnat

iona

l Tra

nspo

rtat

ion”

, PhD

The

sis. U

CL E

nerg

y In

stitu

te, U

CL.

Force of Change: Economies of Scale

0.0001

0.001

0.01

0.1

1

10

100

0 1 10 100 1,000 10,000 100,0001,000,000

Ope

ratin

g Co

sts,

US$

(201

4)/R

TK

Scale (= Load Factor x Capacity), RTK/VKT = (RTK/RTKavail · RTKavail/VKT)

US for-hire trucks

US freight aircraft(1994-2013)

US, Canada railroads(1995/6-2009/10)

Ocean: container, oil, LPG, bulk (2010-14)

.1

Winners LosersFuel Shares: Yard, Passenger, and Freight Sector

Disruptive Technology Change in Rail Transportation

Disruptive Technology Change in Air Transportation

0

10

20

30

40

50

60

70

80

90

100

1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050

Flee

t by

Engi

ne Ty

pe, %

Piston engine A/C

Jet engine A/C

Turboprop A/C

Force of Change: Increase in Productivity

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

0 100,000 200,000 300,000 400,000 500,000 600,000

Ope

ratin

g Co

sts,

Cen

ts(1

960)

/Sea

t-mile

Aircraft Productivity, Seat-miles/day

Jet-engine aircraft

4-engine piston aircraft2-engine piston aircraft

Turboprop aircraft

B-707 SeriesDC-8 Series

B-720

CV-880DC-7B/C

DC-4

DC-7

L-1049G

L-1049H

DC-6B

L-049

L-749

DC-6

V-700/800 L-188

DC-3

CV-240

CV-340CV-440

M-4040.0

0.5

1.0

1.5

0 20,000 40,000 60,000

Crew

Cos

t, Ce

nts

(196

0)/S

eat-m

i

Productivity, Seat-mi/h

Force of Change: Increasing Environmental & Societal Pressure?

• LH2: 120 MJ/kg but 8.5 MJ/L, capital-intensive infrastructure required from well-to-wake, potentially zero-carbon but increase H2O emissions• LNG: 50 MJ/kg vs. 23.4 MJ/L, capital-intensive infrastructure required

from well-to-wake, little to no benefit over jet fuel on lifecycle basis• Drop-in synthetic fuels (from biomass, H2 and CO2, etc.): 42.8 MJ/kg

vs. 34.7 MJ/L, upstream infrastructure change, potentially zero-carbon• Electricity: currently 200 Wh/kg (0.7 MJ/kg) but at least 4X required,

capital-intensive infrastructure required from well-to-wake, potentially zero-carbon and no non-CO2 warming

Alternative Aviation Fuels

0

50

100

150

200

250

0.0 0.5 1.0 1.5 2.0 2.5 3.0

Net L

ifecy

cle C

O 2Em

issio

ns, g

CO2/

RPK

Primary Energy, MJ/RPK

Alternative Aviation Fuels

Coal

Crude Oil

Natural Gas

Renewables

All aircraft are assumed to have an energy intensity of 1 MJ/RPK

Ref.

Synfuel

LH2

LNG

Synfuel

LH2

LH2Synfuel (BTL) LH2 (Coal+CCS)

Solarfuels (PTL)

Impact of Biomass-based Synthetic Fuels on CO2 Intensity

Gu

cwa

M.,

Sch

äfe

rA

., 2

01

3.

Th

e i

mp

act

of

sca

le o

n e

ne

rgy

inte

nsi

ty i

n f

reig

ht

tra

nsp

ort

ati

on

, Tr

an

spo

rta

tio

n R

ese

arc

h P

art

D:

Tra

nsp

ort

an

d E

nvi

ron

me

nt,

Au

gu

st.

1

10

100

1,000

10,000

0 1 10 100 1,000 10,000 100,0001,000,000

Life

cycl

e C

arb

on

In

ten

sity

, g

CO

2/R

TK

Scale (= Load Factor x Capacity), RTK/VKT

Aircraft(US: 1991-2010)

Trucks(8 countries:1940-2010)

Railways(6 countries:1980-2010)

Water Vessels(3 sources: current

technology)

.1

BTL (80% CO2 intensity reduction)

• In the absence of disruptive technology, sector-wide net-zero CO2emissions can only be achieved through atmospheric CO2 removal, e.g., direct air capture (DAC)• Costs of DAC: $94-232/t(CO2)

(https://www.sciencemag.org/news/2018/06/cost-plunges-capturing-carbon-dioxide-air)• Very high costs in 2050: $150/t(CO2) x 1.5 bln t(CO2) = $225 bln or

$18 per PAX >> $6.12 net profit today• Detailed, airline-based modelling required to better understand

implications.

Removing CO2 from the Air

All-Electric Aircraft can eliminate CO2 and non-CO2 Warming

Le

e D

.S.,

Fa

he

y D

.W.,

Fo

rste

r P.M

., N

ew

ton

P.J

., W

it R

.C.N

., L

im L

.L.,

Ow

en

B.,

Sa

use

nR

., 2

00

9.

Av

iati

on

an

d G

lob

al

Cli

ma

te C

ha

ng

e i

n t

he

21

st

Ce

ntu

ry,

Atm

osp

he

ric E

nv

iro

nm

en

t 4

3(2

2–

23

):3

52

0-3

53

7.

Electric Aircraft Architectures (leading to Distributed Propulsion)

Natio

nal A

cade

mie

s of S

cienc

es, E

ngin

eerin

g, a

nd M

edici

ne. 2

016.

Com

mer

cial

Airc

raft

Prop

ulsio

n an

d En

ergy

Syst

ems R

esea

rch:

Red

ucin

g Gl

obal

Car

bon

Emiss

ions

. Was

hing

ton,

DC:

The

Nat

iona

l Aca

dem

ies P

ress

. doi

:10.

1722

6/23

490.

Jet engine aircraft: ! = #$%$&' ()

*+ ln

./0/1/23.4/023

All-electric aircraft: ! = #$%$&' 5)

*+

.6211789

.:/8;8241

Breguet Range Equation

Propulsive PowerFuel energy / time

Propulsive PowerBattery Power

H: Fuel energy contentE: Battery specific energyL/D: Light-to-drag ratiom: Mass

(Breguet) Range Equation: All-Electric Aircraft

Hepp

erle

M.,

2012

. Ele

ctric

Flig

ht –

Pote

ntia

l and

Lim

itatio

ns. A

VT-2

09 W

orks

hop

on

Ener

gy E

fficie

nt Te

chno

logi

es a

nd C

once

pts O

pera

tion,

Lisb

on, P

ortu

gal,

pp. 1

–30.

Modelling All-Electric Aircraft Performance

Gnad

tA.R

., Sp

eth

R.L.

, Sab

nisJ

.S.,

Barr

ett S

.R.H

., 20

19. T

echn

ical a

nd e

nviro

nmen

tal

asse

ssm

ent o

f all-

elec

tric

180-

pass

enge

r com

mer

cial a

ircra

ft, P

rogr

ess i

n Ae

rosp

ace

Scie

nces

, 105

:1-3

0.

Battery Specific Energy has increased by � 3%/yr(A doubling every 20-25 years)

Koh H., Magee C.L., 2008. A functional approach for studying technological progress: Extension to energy technology, Technological Forecasting and Social Change 75(6):735-758

Crabtree G., Kócs E., Trahey L., 2015. The energy-storage frontier: Lithium-ion batteries and beyond. MRS Bulletin 40, 1067-1076.

All-Electric Aircraft Market Size by Distance

Sch

äfe

rA

.W.,

Ba

rre

tt S

.R.H

., D

oy

me

K.,

Dra

y L

.M.D

., G

na

dt

A.R

., S

elf

R.,

O’S

ull

iva

n

A.,

Sy

no

din

os

A.P

., T

ori

jaA

.J.,

20

19

. “

Te

ch

no

log

ica

l, E

co

no

mic

an

d E

nv

iro

nm

en

tal

Pro

sp

ects

of

All

-Ele

ctr

ic A

ircra

ft”,

Na

ture

En

erg

y 4

:16

0-1

66

.

Key Air Transportation Characteristics

SchäferA.W.,BarrettS.R.H.,DoymeK.,DrayL.M.D.,GnadtA.R.,SelfR.,O’Sullivan

A.,SynodinosA.P.,TorijaA.J.,2019.“Technological,EconomicandEnvironmental

ProspectsofAll-ElectricAircraft”,Na

ture

Energy

4:160-166.

Direct Operating Cost (DOC)

• Electrification affects 75% of DOC (capital costs, maintenance, energy, en-route / airport charges)• Capital costs: lower-cost propulsors, absence of fuel system and APU

versus higher-cost, first set of batteries• Maintenance costs: potentially lower engine maintenance costs versus

higher airframe maintenance and battery replacement costs• Cost-effectiveness depends mainly on battery performance and costs,

jet fuel and electricity price• 2015 jet fuel prices ($1.8/Gal) and advanced batteries (800 Wh/kg, $100/kWh)

would require electricity prices of max $0.05/kWh• A carbon tax of $100/tCO2 would allow electricity prices of max $0.09/kWh

Break-Even Electricity Price

SchäferA.W.,BarrettS.R.H.,DoymeK.,DrayL.M.D.,GnadtA.R.,SelfR.,O’Sullivan

A.,SynodinosA.P.,TorijaA.J.,2019.“Technological,EconomicandEnvironmental

ProspectsofAll-ElectricAircraft”,Na

ture

Energy

4:160-166.

Costs of Renewable Power are declining

Global weighted average CSP, solar PV, onshore and offshore wind project LCOE data to 2017 and auction price data to 2020, 2010-2020

IREN

A, 2

018.

Ren

ewab

le P

ower

Gen

erat

ion

Cost

s in

2017

, Int

erna

tiona

l Re

new

able

Ene

rgy

Agen

cy, A

bu D

habi

.

Aircraft Warming Intensity

180-Seat NarrowbodyAircraft, Range = 400 nmi

SchäferA.W.,BarrettS.R.H.,DoymeK.,DrayL.M.D.,GnadtA.R.,SelfR.,O’Sullivan

A.,SynodinosA.P.,TorijaA.J.,2019.“Technological,EconomicandEnvironmental

ProspectsofAll-ElectricAircraft”,

Natu

reEn

ergy

4:160-166.

180 Wh/RPK

Carbon Intensity of Electricity needs to decline strongly

Historical Trend

Future Projections for limiting ∆T to 2o

(EMF27 model comparison, 450 ppm scenario)

Da

ta s

ou

rce

: IP

CC

AR

5 S

ce

na

rio

s D

ata

ba

se

(h

ttp

://w

ww

.iia

sa

.ac.a

t/

we

b/h

om

e/re

se

arc

h/re

se

arc

hP

rog

ram

s/E

ne

rgy

/IP

CC

_A

R5

_D

ata

ba

se

.htm

l)

IEA

, 2

01

8.

Wo

rld

Su

mm

ary

En

erg

y B

ala

nce

s.

• Using all-electric aircraft for flight segments up to 400-600 nmi (741-1,111 km) within the 2015 flight network would result in extra electricity demand of• 110-340 TWh (0.6-1.7%) globally• 23-83 TWh (0.6-2.2%) US• 11-33 TWh (1.3-3.7%) UK

• Around 15% of all flights in early morning (overnight charging) àremaining 85% determine investments into new power generation capacity (assuming 35% capacity factor)• 31-120 GW globally• 6.6-27 GW (US)• 1.2-3.6 GW (UK)

Electric Power Implications

• Noise contours of All Electric Aircraft• Using MIT aircraft specs and flight profile

• Parametric study to evaluate noise vs.• Number of propulsors

• Battery specific energy

• Battery charging strategies

• Mission length

• Conclusions• Noise benefits could be substantial on short

missions

• Noise highly dependent on all operational constraints and procedures, i.e., flight profiles and recharging strategies

• Take-off noise lower than conventional aircraft, due to lower fan pressure ratios and absence of combustion noise

• Approach noise likely higher than conventional aircraft, due to higher aircraft weight

All-Electric Aircraft Noise Analysis

Sch

äfe

rA

.W.,

Ba

rre

ttS

.R.H

.,D

oy

me

K.,

Dra

yL.

M.D

.,G

na

dt

A.R

.,S

elf

R.,

O’S

ull

iva

n

A.,

Sy

no

din

os

A.P

.,To

rija

A.J

.,2

01

9.

“Te

chn

olo

gic

al,

Eco

no

mic

an

dE

nv

iro

nm

en

tal

Pro

spe

cts

of

All

-Ele

ctri

cA

ircr

aft

”,Na

ture

Energy

4:1

60

-16

6.

• To become a feasible alternative, all-electric aircraft require• significantly higher specific energy and power batteries• significantly higher specific power aircraft motors and power electronics• lower battery costs and enabling economic conditions

• Enabling technologies and factors• Electric air taxis• Turbo and hybrid-electric aircraft• New business models?

• Mutually reinforcing factors with time scales measured in decades• Rising battery performance and declining costs, electricity grid decarbonization,

strong decline in renewable power generation costs• R&D on all-electric aircraft design and key components needs to start now

Conclusions: Electric Aircraft Study

Electric Aircraft Research Team• Prof. Steven R.H. Barrett (MIT)• Dr. Lynnette Dray (UCL)• Dr. Khan Doyme (UCL)• Roger Gardner (U. Southampton)• Mr. Albert Gnadt (MIT)• Dr. Chez Hall (U. Cambridge)• Mr. Weibo Li (UCL)• Mr. Marius Macys (UCL)• Dr. Antonio Martinez (U. Southampton)• Prof. Andreas W. Schäfer (UCL)• Prof. Rod Self (U. Southampton)• Ms. Vanessa Schröder (ETH Zurich)• Dr. Aidan O’Sullivan (UCL)• Mr. Bojun Wang (UCL)• Mr. Kinan Al’Zayat (UCL)

Next Steps: Aviation Integrated Model with Airline Competition

• Each airline sequentially maximizes profit (P) within its network (objective fct.)• Three decision variables: airfare (F), flight frequency (FF), type of aircraft (a)

P = ∑ "#$ %& ' "#$ ()*& − ∑ ∑,-./0&123,5,0 ' %%5,0 − ∑ ∑,-.678,5,0 ' ()*5,0

• Set of around 10 linearized constraints• Iteration until equilibrium• IBM CPLEX linear programming solver

9:9#$ ;:<=> ;:<=>?:)/. ?:)/.

Revenues Flight-related costs PAX-related costs

Aviation Integrated Model: Sample Results – Australia

0

100

200

300

400

500

0 100 200 300 400 500

Pred

icte

d Se

gmen

t Airf

ares

, US$

Observed Segment Airfares, US$

Observed NetworkPAX flows

Generated NetworkPAX flows

R2 = 0.79N = 256

SYD

Itinerary Airfares

10

100

1,000

10,000

10 100 1,000 10,000

Pred

icte

d Se

gmen

t PAX

, 100

0

Observed Segment PAX, 1000

Segment PAX

R2 = 0.77N = 256

Doym

eK.

, Dra

y L.

M.,

ATSL

ab