NEPTUNE Training Session: Part 2: Wildfire Patterns

Foothills Model Forest Natural

Disturbance Program

September, 2006

Dr. David Andison

Outline:

1. Natural patterns of wildfires.

2. NEPTUNE origins.

3. Natural vs. cultural disturbance patterns.

What is the natural pattern concept all about, and

(how) is it relevant to my world?

Give me some working examples of what a natural

pattern-inspired disturbance plan looks like.

How do patterns of past, current, and future cultural

disturbances compare to those of wildfires?

(How) Are natural disturbance dynamics critical

to other known, important ecological processes?

What are the patterns and processes of natural

disturbance?

Do I need to learn new terminology to understand

or use natural patterns?

Will our current system / budget allow natural

patterns to happen?

(How) Do natural-inspired disturbance patterns fit

with other economic and social values?

A Hierarchy of Needs

Give me some operational tools with which to help

me design landscapes with natural patterns in mind.

Theory

Language

Knowledge

Examples

Relevance

Convergence

Acceptance

Tools

What are the patterns and

processes of natural

disturbance?

… at intermediate scales?

0

10

20

30

40

50

60

70

80

<1 1

-40

40-8

0

80-2

00

2-60

0

600-

1,000

1-2,

000

2-5,

000

5-10

,000

10,0

00+

Forest Patch Size-Class (ha)

Perc

en

t A

rea

Area

NumbersMost fires are small,

… but the large ones count for most land

Disturbance Sizes & Disturbance Numbers

(In the Alberta Foothills)

Event area shapes are simple

and increase with increasing

event size.

Shape = 2.2

Shape = 2.2

Shape = 2.4

Shape = 2.8

How Many Disturbed Patches

are in an Event?

Event Size WC Alta. Sask.

<200 ha 3 1

200-1,000 ha 13 3

>1,000 ha 30 10

Disturbed Patches

• There is almost always one HUGE patch.

• Alberta average 73%

Figure 13. Size of the Largest Disturbance Patch as a

Percentage of Net Disturbed Area in Multi-Patch

Events

0

5

10

15

20

25

30

35

21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100

% of the Disturbed Area Within an Event That is Accounted for by

the Largest Disturbance Patch

Re

lati

ve

Fre

qu

en

cy

(%

)

How Much Event Area is

Matrix Remnants?

Event Area in Matrix Remnants for

WC Alberta and Saskatchewan

0

10

20

30

40

50

60

0-10 11-20 21-30 31-40 41-50 51-60

Percent of Event Area in Matrix Remnants

Re

lati

ve

Fre

qu

en

cy WC Alberta

Saskatchewan

NO RELATIONSHIP TO EVENT SIZE!

How Much Event Area is

Island Remnants?

Event Area in Island Remnants for

WC Alberta and Saskatchewan

0

10

20

30

40

50

60

0-10 11-20 21-30 31-40 41-50 51-60

Percent of Event Area in Island Remnants

Re

lati

ve

Fre

qu

en

cy WC Alberta

Saskatchewan

NO RELATIONSHIP TO EVENT SIZE!

How Much Event Area is

Total Residuals?

NO RELATIONSHIP TO EVENT SIZE!

Event Area in Residuals for WC

Alberta and Saskatchewan

0

10

20

30

40

0-10 11-20 21-30 31-40 41-50 51-60

Percent of Event Area in Residuals

Re

lati

ve

Fre

qu

en

cy WC Alberta

Saskatchewan

Island Density by Size for WC Alberta

and Saskatchewan

0

10

20

30

40

50

60

70

80

<-0.5 >0.5-2 >2-10 >10-40 >40

Island Size-Class

Re

lati

ve

Fre

qu

en

cy WC Alberta

Saskatchewan

Island Area by Size for WC Alberta

and Saskatchewan

0

10

20

30

40

50

60

70

<-0.5 >0.5-2 >2-10 >10-40 >40

Island Size-Class

Re

lati

ve

Fre

qu

en

cy WC Alberta

Saskatchewan

Breakdown of the Area of Island Remnants of

Different Sizes by Disturbance Patch Size

0

10

20

30

40

50

60

70

80

90

100

10-80 ha 81-600 ha 601-5,000 ha >5,000 ha

Disturbance Patch Size-Class

Co

ntr

ibu

tio

n t

o T

ota

l Is

lan

d R

em

na

nt

Are

a (

%)

Islands <0.2 ha

Islands 0.2 - 1 ha

Islands 1-10 ha

Islands >10 ha

Small Fires Have Higher Levels of Small Island Area

Figure 18. Number of Island Remnants per Disturbed Patch

0

50

100

150

200

250

300

350

1 10 100 1,000 10,000

Disturbed Patch Area (ha)

Nu

mb

er

of

Isla

nd

Rem

nan

ts

Island Mortality Levels for WC Alberta

and Saskatchewan

0

10

20

30

40

50

0% 1-24% 25-49% 50-74% 75-94%

Island Mortality Levels

Re

lati

ve

Fre

qu

en

cy WC Alberta

Saskatchewan

Anatomy of Residuals in

Saskatchewan

0

5

10

15

20

25

30

35

40

Re

sid

ua

l C

on

trib

uti

on

(p

ct

of

ev

en

t

are

a)

Islands (6-24%)Islands (25-49%)Islands (50-74%)

Islands (75-99%)Islands (100%)Matrix (100%)

}

Anatomy of Residuals in WC Alberta

0

5

10

15

20

25

30

35

40

Re

sid

ua

l C

on

trib

uti

on

(p

ct

of

ev

en

t

are

a)

Islands (6-24%)Islands (25-49%)Islands (50-74%)

Islands (75-99%)Islands (100%)Matrix (100%)

Islands That Also Form

Corridors – 21% (in Sask).

Corridor Matrix

Peninsular

Matrix

Matrix That Also Form

Corridors – 28% (in Sask)

Edge Island Area – 56%

(Alberta)

Detached Island

Edge Island

Average Contribution of Event Components by Vegetation Class For

Saskatchewan

0%

20%

40%

60%

80%

100%

Forest Non-Commercial Forest Non-Forest

Vegetation Class

Per

cent

of E

vent

Are

a

Matrix RemnantsIsland RemnantsDisturbed

64% of Event Area Burns

68% of Commercial Forest Burns

53% of Non-Contributing Forest Burns

43% of Non-Forest Burns

In The Event Burnt

Comm. Forest 100 ha (x 68%) 68 ha

Non-Contr. Forest 25 ha (x 53%) 13 ha

Non-Forest 5 ha (x 43%) 2 ha

Total 130 ha (64%) 83 ha

How Do Different Residuals Respond?

Quite

Responsive

Most

Responsive

Least

Responsive

Disturbance Events

Matrix

Remnants

Disturbed

Patches

Island

Remnants

Disturbed

Areas

Individual

Remnants

Matrix Corridors

Matrix Peninsulas

Island Corridors

Edge Islands

Detached Islands

Island Mortality

Who Needs a Standardized

Language or Methods??

Areas in Island Remnants by Fire

Area for 3 Alberta Studies

0

2

4

6

8

10

12

Study A Study B Study C

Per

cen

t Are

a in

Isla

nd

Rem

nan

ts<500 ha

500-1,000 ha

>1,000 ha

Delong & Tanner

recent aerial photos

of older fires in sub-

boreal BC and a

narrow def’n of

“island”.

FMF ND Program

aerial photos taken

right after the fire of

only older fires, a

minimum resolution of

0.01 ha, and a broad

def’n of “residual”.

Eberhart and Woodard

aerial photos taken

right after recent fires

in Alberta, at a

minimum resolution of

2 ha, and inexact def’n

of “residual”

Name That Polygon.

Island Density by Size for WC Alberta

and Saskatchewan

0

10

20

30

40

50

60

70

80

<-0.5 >0.5-2 >2-10 >10-40 >40

Island Size-Class

Re

lati

ve

Fre

qu

en

cy WC Alberta

Saskatchewan

Island Area by Size for WC Alberta

and Saskatchewan

0

10

20

30

40

50

60

70

<-0.5 >0.5-2 >2-10 >10-40 >40

Island Size-Class

Re

lati

ve

Fre

qu

en

cy WC Alberta

Saskatchewan

Breakdown of the Area of Island Remnants of

Different Sizes by Disturbance Patch Size

0

10

20

30

40

50

60

70

80

90

100

10-80 ha 81-600 ha 601-5,000 ha >5,000 ha

Disturbance Patch Size-Class

Co

ntr

ibu

tio

n t

o T

ota

l Is

lan

d R

em

na

nt

Are

a (

%)

Islands <0.2 ha

Islands 0.2 - 1 ha

Islands 1-10 ha

Islands >10 ha

Small Fires Have Higher Levels of Small Island Area

Island Mortality Levels for WC Alberta and

Saskatchewan

0

10

20

30

40

50

0% 1-24% 25-49% 50-74% 75-94%

Island Mortality Levels

Re

lati

ve

Fre

qu

en

cy WC Alberta

Saskatchewan

… and large events have more island area with

high levels of mortality.

Areas in Island Remnants by Fire

Area for 3 Alberta Studies

0

2

4

6

8

10

12

Study A Study B Study C

Per

cen

t Are

a in

Isla

nd

Rem

nan

ts

<500 ha

500-1,000 ha

>1,000 ha

Small Islands in out in

“Natural” Fires yes no yes

Partial Burns no yes yes

Matrix Remnants no yes yes

All 3 are “right”, but applied 3 different

methods, and 3 different sets of definitions.

The Desire for a Tool for

Disturbance Event Planning

• 10 years of research!

• Comfortable with the concept

• The confidence that it is accepted, and here to

stay.

• Desire to understand how current disturbances

differ from historical.

• The spatial language is more involved and less

directly applicable than originally imagined.

• Eager get more involved with the nuts and

bolts.

What is the natural pattern concept all about, and

(how) is it relevant to my world?

Give me some working examples of what a natural

pattern-inspired disturbance plan looks like.

How do patterns of past, current, and future cultural

disturbances compare to those of wildfires?

(How) Are natural disturbance dynamics critical

to other known, important ecological processes?

What are the patterns and processes of natural

disturbance?

Do I need to learn new terminology to understand

or use natural patterns?

Will our current system / budget allow natural

patterns to happen?

(How) Do natural-inspired disturbance patterns fit

with other economic and social values?

A Hierarchy of Needs

Give me some operational tools with which to help

me design landscapes with natural patterns in mind.

Theory

Language

Knowledge

Examples

Relevance

Convergence

Acceptance

Tools

Tool Design is Everything

Recall: - If we build it, will they come?

- Is the past a model for the future?

- It is a new science:

- Red flag check

- Temporal high grade check

- Spatial high grade check

- Cherry-picking check

NEPTUNE ArcGIS Tool that automates the conversion of

shapefiles of disturbances into the new spatial

language, and compares patterns to NRV for:

- Event size

- Event shape

- No. of disturbed patches

- Size of largest disturbed patch

- Disturbed patch shape

- Pct area in matrix remnant

- Pct area in island remnant – by event

- Pct area in island remnant – by dist. patch

- Pct area in total residuals

- Island sizes

Cherry-

picking

Mistik Full FMA (1988-2005)

Comparison with Historical Natural Disturbance Patterns

Percent Area of Event as Residual Remnants (Matrix + Island)

(Event Count: 597 events analyzed)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0-10 >10-20 >20-30 >30-40 >40-50 >50-60 >60-70 >70-80 >80-90 >90

Percent Area as Residual (Matrix + Island)

Rel

ativ

e F

req

uen

cy

Current History

Spatial

High-grade

Red Flag

Temporal

High-grade

NEPTUNE

• Output saved, stored, and compared as

shapefiles and Excel spreadsheets.

• Does not make decisions or

suggestions.

• The only output format is frequency

distributions.

• Currently calibrated for WC Alberta and

western Saskatchewan.

• Partner list; ASRD, HWP, ANC, Mistik

Management, + ? more?

Temporal High-

grading

NEPTUNE is Just a Model (So Be Careful)

• It will almost always give you an “answer”.

• It must simplify the messy world.

• There is no right or best solution.

• It is impact / value-neutral.

“DECISION-SUPPORT”

NRV Does Not Translate Perfectly

to CRV (So Be Even More Careful)

• Linear disturbances

• Timelines

• Permanency

• Overlapping disturbance events

• Conventions vs. spatial language

• Resolution

NEPTUNE Can Help

Address These Gaps:

• What does a natural pattern inspired plan look like?

• (To what degree) Will the current “system” allow it

to happen?

• Are there economic or social barriers /

convergences?

• How does NRV compare to current practices

(CRV)?

• Help me design a more natural disturbance plan.

Potential NEPTUNE Uses

• Learning.

• Communicating.

• Operational harvest planning.

• Prescribed burn planning.

• Harvest / PB monitoring.

• Strategic planning – “state of the forest”

• Assistance for regulators.

• Administration / accounting.

• Hypothesis generation for issue-based questions.

• ???