fire masterclass - home | landscapes and policy hub...0.05 x0.05 grid cells? 2.how to best use...

Slide #1 Fire Masterclass 11 October 2013

Fire Masterclass

Challenges of predicting future fire regimes

top-down versus bottom-up approach? Matthias Boer

Hawkesbury Institute for the Environment

University of Western Sydney


What and Why I model…

What: ‘mechanics’ of ecological systems we live in and depend on

Why: interpret state of the system & make predictions about response

• process-pattern relationships (e.g., in ecohydrology, fire ecology)

• pattern formation: emergence & self-organization, top-down control

• effects of spatiotemporal heterogeneity, scaling issues

• minute to decadal & patch to continental scales

• focus of climatology-ecology link: identify key biophysical constraints of

climate on ecological systems behaviour


Case study: Interdependence of climate, fuels & fire


Archibald et al. (2013) PNAS 110(16) Frequent, grass-fueled, fire regimes Infrequent, litter-fueled, fire regimes


Fire = f(plant growth, desiccation)

P Ep Et R

Et = actual evapotranspiration ∝ productivity

Ep-Et = D, evapotranspiration deficit ∝ drying

P Ep Et

Climatic water balance

R

Et = actual evapotranspiration ∝ productivity

Ep-Et = D, evapotranspiration deficit ∝ drying

SWITCH 1 SWITCH 2


Climate-fuel-fire model Fire activity index (1997-2010)

Mean annual Et (mm/yr)

Mean annual D (mm/yr)

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.2

0.4

0.6

0.8

1.0

AET (mm/yr)

PET-

AET

(mm

/yr)

0.1

0.1 0.2 0.3

0.4

0.5

0.6

0.7

0.8

0.99 quantile surface of fire activity index, R2: 0.80, p<0.001

Et-limited fire activity (fuel production)

D-limited fire activity (fuel dryness)

-45

-40

-35

-30

-25

-20

-15

-10

0.0

0.2

0.4

0.6

0.8

-45

-40

-35

-30

-25

-20

-15

-10

200400600800100012001400

120 130 140 150

-45

-40

-35

-30

-25

-20

-15

-10

500

1000

1500

Et (mm/yr)

D (m

m/y

r)


c

0 20 40 60 80 100

020

4060

8010

0

Tree Canopy Cover (%)

Gra

ss c

ompo

nent

of t

otal

fuel

load

(%)

tau=0.9

tau=0.5

1 2

020

4060

80

Fire regime domain

MO

DIS

tree

cov

er (%

)b a

1 2

020

4060

8010

0

Fire regime domain

Perc

enta

ge g

rass

fuel

D-limited Et-limited D-limited Et-limited

Grass fuel % Tree cover % Grass fuel %


0 20 40 60 80

0.0

0.2

0.4

0.6

0.8

1.0

MODIS treecover (%)

Prob

abilit

y

Domain 1Domain 2c D-limited Et-limited

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.2

0.4

0.6

0.8

1.0

AET (mm/yr)

PET-

AET

(mm

/yr)

0

10

20

30

40

50

60

70

80

90

a

Et (mm/yr)

D (m

m/y

r)

0.99 quantile of MODIS tree cover % R2:0.87, p <0.000

0 20 40 60 80

0.0

0.2

0.4

0.6

0.8

1.0

MODIS tree cover (%)

Fire

act

ivity

inde

x

0 20 40 60 80

0.0

0.2

0.4

0.6

0.8

1.0

treecov

Effe

ct

Effect of treecov

MODIS tree cover (%)

Fire

act

ivity

inde

x

b


120 130 140 150

-40

-35

-30

-25

-20

-15

-10

x

y

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.2

0.4

0.6

0.8

1.0

AET (mm/yr)

PET-

AET

(mm

/yr)

0.1

0.1 0.2

0.3

0.4

0.5

0.6

0.7

0.8

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

ADL

ASP

BNECBR

DRW

HBA

MEL

PER

SYD

0.1

0.1 0.2

0.3

0.4

0.5

0.6

0.7

0.8

Et (mm/yr)

D (m

m/y

r)

D-limited fire

Et-limited fire


What does my modelling mean?

• Modelling framework captures both climate-fire & climate-fuel

relationships…

• …therefore should hold (better than existing models?) for future

climate conditions and be applicable to predicting change

• Some potential effects of rising atmospheric [CO2] (e.g., via WUE)

can be accommodated in same water balance framework


Modelling Caveats

• Coarse spatial scale

• Based on relatively short observation period

• (over)simplifies diversity of fire regimes

• Characterizes climate envelops of fire regimes – what’s the

relevance of the change trajectory through climate space?


My 3 biggest modelling challenges are?

1.How to conceptualise ecosystem structure and function using 0.05°x0.05° grid cells?

2.How to best use uncertainty (or data distributions) in modelling

3.How to best combine top-down approaches (e.g. climate constraints on fuels and fire) and bottom-up approaches (e.g. vegetation modelling)

4.Determine spatiotemporal variation in the relative strengths of climate-fuel-fire interactions

fire masterclass - home | landscapes and policy hub...0.05 x0.05 grid cells? 2.how to best use...

Documents