Hlp 1999 03 prsnttn ndwkshpcoarsescaledisturbancepatterns

Coarse-Scale Disturbance Patterns

on FMF Landscapes

• Types of disturbances

• Frequency of disturbances

• Seral-stage representation

• Non-forested lands

• Natural disturbance integration

Disturbance Types on the FMF

• Forest Fires -

• lightning & people caused

• Stand-replacing and stand-maintaining

• Wind

• Herbivory

• Flooding / stream meander

• Slides

• Insects & disease

• Cultural

Harvesting

Oil, gas, and mining activity

Urban development

Based on experience and

knowledge to date, the majority of

the forest on the FMF is even-

aged.

… meaning that we can use stand age

data to represent coarse-scale

disturbance activity patterns.

Why are Good Age Data Important?

• Helps to identify landscapes of differing disturbance

activity.

• Disturbance Frequency

•Fire Cycle -average time to disturb equivalent area of

a landscape.

•Fire Return Interval - average time between

disturbances at a single site.

• Seral-stage retention across a landscape (ie. how much

area in “old growth”?).

• Fuel-management and fire risk.

• Determines what a “patch” is.

Potential Inventory Age Problems:

•Representing “natural” conditions is difficult.

• Cultural activity (harvesting, oil and gas, roads,

urban, recreation, fire / pest / disease control).

•Accuracy is questionable.

• Ground sampling intensity and location

• Aging methods

• Breast height age

•Class definition issues:

• Age-class boundaries width

• Width of the “Older than” class.

• What a “patch” is.







• Aging methods






Representing “Natural” Conditions, or

Bandaloop Luck

Jasper National Park:

No harvesting, excellent fire protection, & complete

fire history age maps*

Weldwood FMA:

Most of a fire history map* from 1961, little

harvesting prior, excellent fire protection since

* A fire history map is NOT the same as inventory age.

- methodology, sampling intensity.







• Aging methods






How Good Are the Fire History Stand

Ages?

0

10

20

30

40

50

60

70

80

90

100

-5 -4 -3 -2 -1 0 1 2 3 4 5

No. of 20-year Age-Classes Younger Than Inventory Age

Perc

en

t of

plo

ts

Sub-alpine

Upper Foothills

Lower Foothills

PSP Data are Older

than Stand Origin

Age

PSP Data are

Younger than Stand

Origin Age







• Aging methods






1885

100 ha patch 1880-1899

With Coarse,

Innaccurate Maps:

1885

1895

1875

80 ha patch 1880-1889

10 ha patch 1870-1879

10 ha patch 1890-1899

With More Accurate,

High Resolution Maps:

How Do We Want to Define a Patch?

Tradeoff between effort and accuracy.

Do we need to know the exact year of

origin of every patch of every age?

Probably not.

A 10-20 year class allows for several

years worth of disturbance activity.

Ecologically, they are probably similar

10-20 years later. Much easier to define

and map.

PATCH = The contiguous, gross area

of same-aged (within 20 years) forest. (includes islands & non-forested areas).

0

5

10

15

20

25

30

35

40

45

1970 1930 1890 1850 1810

Year of origin

Perc

en

t A

rea

Subalpine (east) Age-Class Distribution

From Forest Inventory

Existing Forest

Inventory used an

“1810 and Older”

Age-class.

How is it really

distributed?

Actual Distribution of the "Older Than" Age-

Class for the Upper Foothills

0

5

10

15

20

25

30

35

40

1990 1950 1910 1870 1830 1790 1750 1710

Origin

Pe

rce

nt o

f "O

lde

r T

han"

Ag

e-

Cla

ss

The Origin Date

“1810 and Older”

Represents 61,900

ha, or 12% of the

Landscape.

Original Age-Class

0

5

10

15

20

25

30

35

40

45

1970 1930 1890 1850 1810 1770 1730 1690 1650

Year of origin

Perc

en

t A

rea

Original Data

Corrected Data

Subalpine (east) Age-Class DistributionWith and Without Age Corrections

Upper Foothills Age-Class DistributionWith and Without Age Corrections

0

5

10

15

20

25

30

35

40

1990 1950 1910 1870 1830 1790 1750 1710

Year of Origin

Pe

rce

nt

Are

a

Original Data

Corrected Data

Lower Foothills Age-Class DistributionWith and Without Age Corrections

0

5

10

15

20

25

30

35

40

45

1990 1950 1910 1870 1830 1790 1750 1710

Year of Origin

Pe

rce

nt

Are

a

Original Data

Adjusted Data

Non-forested

Year of stand origin<18501851 - 19001901 - 19501951 - 1996No data

JNP Age-Class Distirbutions

0

5

10

15

20

25

30

35

40

45

1990 1950 1910 1870 1830 1790 1750 1710 1670 1630 1590 1550

Age-Class of Origin

Pe

rce

nt

of

Are

a

Subalpine JNP

Montane

FIRE

CONTROL

1950 Age-Class Distribution of the

Foothills East

0

5

10

15

20

25

30

35

40

45

1950 1910 1870 1830 1790 1750 1710 1670 1630

Age-Class of Origin

Perc

en

t o

f A

rea

Subalpine East

Upper Foothills

Lower Foothills

Average 1950 Fire Cycles on FMF

Landscapes

Landscape From 1950

A.C. Dist’n.

Lower Foothills 76 years

Upper Foothills 82 years

Subalpine east 110 years

* Subalpine JNP 150 years

* Montane 91 years

* includes ~20-30 years of fire control

Older Forest as of 1950 on FMF

Landscapes

Landscape %>300 yrs %>200 yrs %>100 yrs

Lower Foothills 0 0 16

Upper Foothills 0 4 21

Subalpine east 2 15 45

* Subalpine JNP 7 37 59

* Montane 0 3 30

* includes 20-30 years of fire control

HOWEVER, despite high data quality, the

description of age-class “pattern” has been

limited by the sample size (1).

So far….

• High degree of confidence that most stands are

even-aged, and therefore well represented by age-

classes.

• Distinctive rates of disturbance for each natural

subregion = landscapes.

• Distinctive levels of young, mid-seral, and old

growth forest in each natural subregion.

1920

100 ha

Increasing the Sample Size - How?

1920

90 ha

1940

1920

55 ha

1940 1960

In 1960, there were 55 ha in the 1920 age-class

1920

90 ha

1940

In 1940, there were 90 ha in the 1920 age-class

1920

100 ha

Original Area Disturbed in 1920 is 100 ha

Areas in 1950 Projected Areas in 1940

1950 area = 100 ha

1930 area = 200 ha + 20% x 100 = 220 ha

1910 area = 200 ha + 20% x 100 = 220 ha

1890 area = 600 ha + 60% x 100 = 660 ha

Rollback Demonstration

assume proportional to area:

Upper Foothills “Rollback”

Period % Area in 1950 Est. Area Disturbed

1930-1949 2 2

1910-1929 8 8

1890-1919 19 22

1870-1889 34 51

1850-1869 12 36

1830-1849 10 47

1810-1819 <1 6

Average 24

24% every 20 years = 1.2% per year = 83 year fire cycle,

versus 82 years from previous estimate.

Rollback Estimates of Areas

Disturbed on FMF Landscapes

Period Lower Upper Sub east Sub JNP Montane

1930-1949 2 2 1 - -

1910-1929 11 8 15 6 8

1890-1919 11 22 23 12 17

1870-1889 53 51 27 25 50

1850-1869 55 36 4 4 20

1830-1849 67 47 27 12 24

1810-1829 6 1 4 5 6

Area Disturbed per 20 Year Period for

the Upper Foothills Landscape

0

0.2

0.4

0.6

0.8

1

2.5 12.5 22.5 32.5 42.5 52.5

Percent of Landscape burned every 20 years

Cu

mu

lati

ve

Pro

ba

bilit

y

(0.26 + 0.012 x C. Prob) x 100 = Pct. Burned

Cumulative Probabilities of Rate of Burn for

the Foothills Model Forest

0

0.2

0.4

0.6

0.8

1

2.5 12.5 22.5 32.5 42.5 52.5 62.5

Percent burned in 20 years

Cu

mu

lati

ve

Pro

ba

bilit

y

Subalpine eastUpper FoothillsLower FoothillsSubalpine JNPMontane

Area Disturbed per 20 Year Period for

the Upper Foothills Landscape

0

0.2

0.4

0.6

0.8

1

2.5 12.5 22.5 32.5 42.5 52.5

Percent burned every 20 years

Ra

nd

om

Nu

mb

er

(0.26 + 0.012 x 0.6) x 100 = 26.7% Burned

Project the Range of Disturbance

Rates Over Time, Through Space

• Equations derived from rollback estimates of

20 year disturbance rates.

• Spatially explicit disturbance model

(LANDMINE)

• Disturbance size equations.

• Random number generator.

• Initial conditions.

0

25

50

75

100

Y P M O

AC Distribution

Landscape in 1950 - Initial Condition

AC Dist’n

0

25

50

75

100

Y P M O

Landscape in 20 Years?

0

25

50

75

100

Y P M O

AC Dist’n


0

25

50

75

100

Y P M O

AC Dist’n


0

25

50

75

100

Y P M O

AC Dist’n


0

25

50

75

100

Y P M O

AC Dist’n


0

5

10

15

20

25

30

0 10 20 30 40 50 60 70 80 90

Percent Area 0-40 Years

Fre

qu

en

cy

Projected Historical Range of Area for the Young Seral-Stage of the Sub-alpine East

0

5

10

15

20

25

30

0 10 20 30 40 50 60 70 80 90


Fre

qu

en

cy

1995

1950

Projected Historical Range of Area for the Young Seral-Stage of the Sub-alpine

In which year was the

percent of young forest

more “natural”?

0

5

10

15

20

25

30

0 10 20 30 40 50 60 70 80 90


Fre

qu

en

cy

.Projected Historical Range of Area

for the Pole Seral-Stage of the Upper Foothills

0

5

10

15

20

25

30

0 10 20 30 40 50 60 70 80 90


Fre

qu

en

cy

1995 1950

Projected Historical Range of Area for the Pole Seral-Stage of the Upper Foothills

0

5

10

15

20

25

30

35

40

0 10 20 30 40 50 60 70 80 90

Percent Area >200 Years

Fre

qu

en

cy

1995

1950

Projected Historical Range of Area for the Mature Seral-Stage of the Upper Foothills


Landscapes






* Montane 0 3 30


0

5

10

15

20

25

30

35

40

0 10 20 30 40 50 60 70 80 90

Percent Area >200 Years

Fre

qu

en

cy

1995

1950

Projected Historical Range of Area for the Mature Seral-Stage of the Upper Foothills


Landscapes






* Montane 0 3 30


0

10

20

30

40

50

60

0 10 20 30 40 50 60 70 80 90


Fre

qu

en

cy

1995

1950

Projected Historical Range of Area for the Mature Seral-Stage of the Lower Foothills

Simulated Historical Range of Percent of

Young Softwood on Weldwood FMA

0

5

10

15

20

25

30

35

40

5 15 25 35 45 55 65 75 85 95


Fre

qu

en

cy

30,000 ha

240,000 ha

The Range Shrinks as the

Area Increases


Pole-aged Softwood on Weldwood FMA

0

5

10

15

20

25

30

5 15 25 35 45 55 65 75 85 95


Fre

qu

en

cy 30,000 ha

240,000 ha


Old Softwood on Weldwood FMA

0

10

20

30

40

50

60

70

80

90

100

5 15 25 35 45 55 65 75 85 95

Percent Area 200+ Years

Fre

qu

en

cy 30,000 ha

240,000 ha

0

10

20

30

40

50

60

70

80

90

100

0 0.6 1.2 1.8 2.4 3

Millions of Hectares

Per

cen

t A

rea

1

Range of Young Forest Percentages

From Simulation for the Lower Foothills

0-40 Years

0

20

40

60

80

100

0 0.6 1.2 1.8 2.4 3

Millions of Hectares

Perc

en

t A

rea

Range of Old Forest Percentages

From Simulation for the Lower Foothills

101-140 Years

In theory, a very large land area may

“stabilize” all age-class variation.

In practice, an “equilibrium point”

does not exist (where there is no

variation), but smaller areas do exhibit

higher age-class variability than larger

ones.

Age-Class Variability Patterns

This means that smaller areas (<100,000 ha)

experienced regular, extended periods of

extreme age-class variation, and probably

had no “old growth” to speak of for many

years, even decades.

Corollary of this is that “old growth”, like

any other age-class, tends to cluster over

very large areas.

Age-Class Variability Patterns

0 10 20 30 Kilometers

Age-class

<1800

1851 - 1900

1901 - 1950

1951 - 1996

Old-growth is not

evenly distributed

in space

• Age-class distributions have temporal patterns,

the range and shape of which may be unique to

natural subregions.

Is the Lower Foothills range of disturbance

indicative of spring vs. summer fire events?

• The range of temporal variability is generally

quite wide, ranging from between 1 and 2

percent of the land area disturbed to over 50%

for a given 20 year period.

• Areas of the same age-class cluster over large

areas.

Age-Class Summary

What Else at Coarse Scales?

• Fire threat

• Climate change

• Interaction of fire with other natural

disturbances

• Non-forested areas

• Fire and salvaging

Overview of the Foothills Model Forest

Area Forested Non-For Total

Lower Foothills 260,500 35,000 295,500

Upper Foothills 520,000 67,000 587,000

Sub-alpine east 242,500 43,000 285,500

Sub-alpine JNP 401,000 633,000 1,034,000

Montane 85,000 10,000 95,000

TOTAL 1,509,000 788,000 2,297,000

Of Non-Forested... ~ 620,000 ha rock and ice

~ 84,000 ha non-flammable

~ 84,000 ha flammable

Non-Forested Areas: How Much Do

We Know?

• Build up of duff, peat?

• Potential for growing trees - soil temperature,

germination sites?

• Biodiversity implications? Who uses them?

Carbon storage?

Natural Disturbances Will Happen

… so we might as well take advantage.

• Integration of temporal age-class variability.

• Create ecological benefits of natural disturbance

that cultural disturbances cannot create (given

that salvage policies allow for it).

• Disturbance of non-forested areas.

• Stand-level structure, leaving biomass on

site, nutrient release, compaction.

• Fewer wood flow surprises?????

Impact of Fire on Timber Supply

Ideal Example Only (not based on FMF)

Fire Cycle Today Harvest Level Reduction

(with fire control) / Expected Long-run Loss

1,000 years 2%

500 years 5%

250 years 10%

200 years 15%

150 years 18%

100 years 30%

•Non-forested areas are part of the landscape and

much of it evolved with fire. We do not know the

full impacts of removing fire on these areas.

• Natural disturbance will happen, and there are

some positive aspects that a salvage policy can take

advantage of.

• Disturbance of non-forested areas.

• De-stabilizing age-class pattern over time.

• Fine-scale patterns that cultural disturbance

activities cannot match.

Disturbance Design Issues at

Coarse Scales

Disturbance Design Issues at

Coarse Scales

• In this case, ecological subregions are good

means of differentiating areas with distinctive fire

cycles.

• Age-class distributions vary over space. Same-

aged areas tend to cluster.

• Age-class distributions vary tremendously over

time. Where there are “nodes” of old-growth

today, there were likely none 100 years ago.

Documents

Hlp 1999 03 prsnttn ndwkshpcoarsescaledisturbancepatterns