Natural Disturbance

as a Template for

Ecosystem-based

Management

Species and Spaces

FEESA

July 13, 1997

2

Outline

Relevance of natural disturbance to

ecosystem-based management (“Why?”)

Landscape Patterns

Island Remnants / Meso - scale patterns

Stand Structure

Summary

3

Why interest in natural disturbance?

Management for all species individually is

impractical

All species respond to patterns created by

natural disturbance (directly or indirectly) Wildfires have occurred for thousands of years

Some management practices already

resemble some natural disturbances Forest harvesting & wildfire

Increased similarity between managed and

natural disturbances would better conserve

populations

4

Operating Within the

Range of Natural Variability

“The unmanaged forest... is subject to a variety

of natural processes and disturbance such as

succession, nutrient cycling, fire, flood,

blowdown, insect attack and disease.

These processes vary in duration, frequency of

occurrences, size of affected area and severity.

For each type of process, there will be a

particular range of variability.

Alberta Forest Conservation Strategy, draft July 8, 1996

5

Operating Within the

Range of Natural Variability...

The result is a mosaic of natural communities

representing the range of natural variability in

forest ecosystems.

Each of these communities plays an integral role

in maintaining the diversity and function of the

forest ecosystem.

An understanding of “natural variability” is

fundamental to the implementation of ecological

management.

Alberta Forest Conservation Strategy, draft July 8, 1996

6

Operating Within the

Range of Natural Variability...

Human activities are replacing natural

disturbances.

This trend has the potential to simplify the

diversity of the forest and, over the next several

decades, may push it beyond the historic range

of natural variability.

Ecological management proposes that we use

human activities to maintain that range of natural

variability.

Alberta Forest Conservation Strategy, draft July 8, 1996

7

Achieving Ecological Management

through Adaptive Management

“The key idea behind ecological

management is that we can substitute

planned, human activities for random and

naturally occurring biological processes

within the range of variability of those

processes.”

Requires the following 4 new

assumptions:

Alberta Forest Conservation Strategy, draft July 8, 1996

8

Assumption 1

Human activities can approximate natural

processes, at least to some degree.

From a forestry perspective, this may be

accomplished through: the use of cutblocks of different sizes and

shapes,

by variation in rotation age, and

by varying the percentage of removal within

harvested areas.

Alberta Forest Conservation Strategy, draft July 8, 1996

9

Assumption 2

“That our treatment of the forest will fall

within the range of natural variability with

respect to area, shape, age class and

severity of disturbance (within socially

acceptable limits). Natural processes will

still inevitably occur.”

Alberta Forest Conservation Strategy, draft July 8, 1996

10

Assumption 3

That human activities carried out in a

manner that bears some resemblance to

natural ecological events, will result in the

conservation of biodiversity, maintenance

of ecosystem structure and function, and

generate a sustainable flow of goods and

services provided by the forest.

Alberta Forest Conservation Strategy, draft July 8, 1996

11

Assumption 4

That by a process of “adaptive

management”, forest managers will be

able to determine what practices work

and to detect problems, and correct them,

before irreversible damage occurs to

either individual species or the ecosystem

as a whole.

Alberta Forest Conservation Strategy, draft July 8, 1996

Scale is important

100 km

10 km

0.001 km

Mosaic of many wildfires

Structure within a wildfire

Structure

within a

stand

13

Landscape Disturbance Regimes

Describe natural disturbance regime across the

Foothills Model Forest, including frequency, size, and

spatial arrangement of stand-replacing wildfires Weldwood FMA area

Other Provincial lands

Jasper National Park

Assist ecological basis for management planning Harvest scheduling in working forests

Weldwood’s Hinton FMA area

Fire management in protected areas

Jasper National Park, Willmore Wilderness Park

Goal is to conserve biological diversity by maintaining

future seral stage representation and spatial

arrangement within the range of natural variability

Natural Subregion

Alpine

Subalpine

Montane

Upper Foothills

Lower Foothills

0 10 20 30 40 50 Kilometers

Jasper

National Park

Foothills East

15

Inventory: Stand Origin Map

1:50,000 time-since-fire mapping coarse resolution, smallest patches not mapped

air photo interpretation of historical fire

boundaries

ground sampling for evidence to date fire

events

adequate for past 150 years, less

confidence in age of older stands

0 10 20 30 Kilometers

Age-class

<1800

1851 - 1900

1901 - 1950

1951 - 1996

17

Analysis

Stratify by ecological units patch age

patch size

patch spatial arrangement

Relationships between stand origin and topography (slope, aspect)

watercourses

Understand process behind the observed

patterns

18 Age-class midpoint

Patch Age (as of 1950)

Lower Foothills

(300,000 hectares)

0

10

20

30

40

Upper Foothills

(629,000 hectares)

0

10

20

30

40

Subalpine

(109,000 hectares)

0

10

20

30

40

30 70 110 150 190 230 270 310 350

Are

a (

%)

Sizes of patches created by wildfire

10,000 hectares

20

Patch size

Lower Foothills

0

20

40

60

Upper Foothills

0

20

40

60

Subalpine

0

20

40

60

<40 40 80 200 600 2000 5000 >10,000

-80 -200 -600 -2000 -5000 -10,000

Patch size (ha)

Are

a (%

)

(20-yr classes)

21

Changes in area & patch size

Total Area

0

5

10

15

20

1950 1995

pe

rce

nt

of

are

a

Young forest Old forest

Mean Patch Size

0

200

400

600

800

1950 1995m

ean

patc

h a

rea (

ha)

Young forest Old forest

22

Island remnants

Wildfires create burned matrix in which

unburned residuals are embedded

These residuals are variable in size

These residuals are not arranged

randomly aspect, slope, watercourses

Post-burn structure in the burned matrix

is also variable, due to variable fire intensity

variable forest structure prior to wildfire

23

Island remnants

24

Island Remnants

25

Stand Structure

Estimate the range of variability in

deadwood structures of stands recently

disturbed by wildfire and by logging in the

Rocky Mountain Foothills of Alberta

Work in progress, results are preliminary

26

Deadwood in forest ecosystems

Disturbance can quickly generate large

inputs of deadwood wildfire, wind, insect outbreaks

Exists in 3 “pools” (Harmon et al. 1986): present before disturbance

created by disturbance

added after disturbance

Standing vs down “populations” of

deadwood

Deadwood

28

Deadwood in Rocky Mountain Foothills

Stand-replacing wildfire has been a dominant

process shaping the distribution and

structure of forest communities

Most of the landscape exists as patches

created by very large wildfires (over 100 km2)

Tree mortality following wildfire is variable,

but frequently very high

Tree boles usually remain, forming a

“cohort” of standing deadwood, which

eventually enters down deadwood

population

29

Deadwood and forest harvesting

500 km2 harvested in Alberta, 1994/95

Clearcutting is dominant harvest method

Historical trend increased utilization e.g., less than 5% merchantable wood

volume remaining

Recent trend decreased utilization (?)

30

What we don’t know

Range of variability in deadwood structures post-wildfire communities vs post-harvest

communities

Fate of the deadwood cohort after wildfire vertical arrangement (standing vs down)

colonization by lichen, fungi, plants

decay, assimilation into organic layer

Time to “convergence” after fire, logging Lee, Crites et al. (Boreal Mixedwood)

31

What we could do if we knew more

Identify extent to which the range of

“natural” variability in deadwood

structures is beyond the “managed” range

Have historical “catalogue” for

comparison with future managed forests

Design studies that assess the

effectiveness of management options

intended to expand the “managed” range

biodiversity

productivity

##

#

#

#

#

#

#

#

#

#

#

##

#

#

S

Alberta

Yukon

0 500 1000 1500 Kilometers

33

Study areas

1 Mountain wildfire 34 yr.

3 Foothills wildfires 35 - 40 yr.

3 Foothills harvests 23 - 27 yr.

34

30 plots per site

(15 in Mountain

burn)

(almost) randomly

located

Sample plot layout

35

30 m

450 m2

Sampling method

down deadwood

standing dead

standing live

Chaba

Number of snags

0 8 16 24 32 40

# p

lots

0

10

20

30

Burned

0

10

20

30

Number of snags

0 8 16 24 32 40

# p

lots

0

10

20

30

0

10

20

30

Number of snags

0 8 16 24 32 40

# p

lots

0

10

20

30

0

10

20

30

Number of snags

0 8 16 24 32 40

# p

lots

0

10

20

30

Lynx

Gregg Smith

Logged

Burned Burned

Burned

Logged

Burned

Logged

All snags

Burned sites 0 ->500 snags /ha,

usually <100

snags /ha

Logged sites

snags very rare,

usually absent

Chaba

Number of snags

0 4 8 12 16 20

# p

lots

0

10

20

30

Burned

0

10

20

30

Number of snags

0 4 8 12 16 20

# p

lots

0

10

20

30

0

10

20

30

Number of snags

0 4 8 12 16 20

# p

lots

0

10

20

30

0

10

20

30

Number of snags

0 4 8 12 16 20

# p

lots

0

10

20

30

Lynx

Gregg Smith

Logged

Burned Burned

Burned

Logged

Burned

Logged

Only snags >

largest live tree

Burned sites usually <50 snags

/ha, occasionally

higher

Logged sites virtually none

Gregg

0

10

20Burned

0

10

20

0 40 80 120

160

Number of pieces

# p

lots

Logged

0

10

20

0 40 80 120

160

Number of pieces

# p

lots

Logged

Chaba

0

10

20

# p

lots Burned

Lynx

0

10

20 Burned

Smith

0

10

20 Burned

0

10

20

# p

lots

Logged

0

10

200

10

20

0

10

20

0

10

20

0

10

20

0102001020

Number of pieces

of down deadwood

Burned sites highly variable

Logged sites less variable,

usually fewer

Chaba

0

5

10

15

# p

lots

Burned

Gregg

0

5

10

15 Burned

Lynx

0

5

10

15 Burned

Smith

0

5

10

15 Burned

0

5

10

15

0 4 8 12 16 20 24

Mean piece diameter (cm)

# p

lots

Logged

0

5

10

15

# p

lots

Logged

0

5

10

15

0 4 8 12 16 20 24

Mean piece diameter (cm)

# p

lots

Logged

Size of down

deadwood

> 7 cm diameter

Burned sites average 12 cm

diam.

Logged sites similar

Chaba

0

10

20

30

# p

lots

Burned

Gregg

0

10

20

30

Burned

Lynx

0

10

20

30Burned

Smith

0

10

20

30Burned

0

10

20

30

0 8 16 24 32 40 48

Mean piece height (cm)

# p

lots

Logged

0

10

20

30# p

lots

Logged

0

10

20

30

0 8 16 24 32 40 48

Mean piece height (cm)

# p

lots

Logged

Height of down

deadwood

Burned sites average up to 20

cm off ground

Logged sites usually on ground

Chaba

0

5

10

# p

lots

Burned

Gregg

0

5

10

Burned

Lynx

0

5

10Burned

Smith

0

5

10Burned

0

5

10

0 20 40 60 80

Plant cover (%)

# p

lots

Logged

0

5

10# p

lots

Logged

0

5

10

0 20 40 60 80

Plant cover (%)

# p

lots

Logged

Attached plants on

down deadwood

Burned sites variable

Logged sites variable

Chaba

Volume (m3/ha)

0 80 160 240 320 400

# p

lots

0

5

10

15

Burned

0

5

10

15

Volume (m3/ha)

0 80 160 240 320 400

# p

lots

0

5

10

15

0

5

10

15

Volume (m3/ha)

0 80 160 240 320 400

# p

lots

0

5

10

15

0

5

10

15

Volume (m3/ha)

0 80 160 240 320 400

# p

lots

0

5

10

15

Lynx

Gregg Smith

Logged

Burned Burned

Burned

Logged

Burned

Logged

Total volume of

down deadwood

Burned sites variable, up to

> 300 m3 / ha

Logged sites less variable,

average 1/3 of

burned sites

43

Summary

Natural disturbances are highly variable

and not well characterized

The range of variability created by natural

disturbances can be estimated at a range of

scales

Management options can be developed that

perpetuate at least part of the range of

variability caused by natural disturbances

Monitoring and research are needed to

estimate the effectiveness of alternative

management options

44

Natural Disturbance Program

Program Team Dan Farr, Foothills Model Forest

Don Harrison, Alberta Environmental Protection

Hugh Lougheed, Weldwood of Canada

Alan Westhaver, Jasper National Park

Project Collaborators Dave Andison, Bandaloop

Rick Bonar, Weldwood

George Mercer, Jasper NP

Luigi Morgantini, Weyerhaeuser

Marie-Pierre Rogeau, Banff

Chris Spytz, Weldwood