Landscape-scale fuel treatments, modeled
fire behavior, and actual fire effects in the
northern Sierra Nevada
Brandon Collins
Anu Kramer
Scott Stephens
Bill Elliot
Sue Miller
PHOTO: Z. Balsen
PHOTO: Mt. Hough staff, Plumas NF (2005)
Meadow Valley project
PLUMAS NATIONAL FOREST
Management Direction: HFQLG Pilot Project
Strategy: Defensible fuel profile zones (DFPZs)
Study area: 18,600 ha
Landscape proportion treated: 20 %
Implementation: completed (pending scheduled underburns)
Last Chance & Sugar Pine projects
(Sierra Nevada Adaptive Management Project)
SIERRA &TAHOE NFS
Collins et al. 2010 J. For.
Management direction: 2004
SNFPA
Strategy: Strategically placed
area treatments (SPLATs)
Study area: 3500 - 4500 ha
Landscape proportion treated:
20 - 25%
Implementation: EAs revised
Fuel treatment examples (from Blodgett, but applicable to Meadow Valley):
PRE
POST
Base landscape data:
IKONOS imagery
VESTRA vegetation type map
Vegetation inputs derived from:
Imagery classification (field plot validated)
548 field plots (46 post-treatment)
Analysis unit:
4 m and 10 m pixel
Stand (0.1 – 130 ha, average = 3 ha)
Fuel model assignments:
Vegetation type (fixed)
Calibrated with 2008 Rich fire
CART analysis on plot fuel loads
Treatments:
Pixel adjustments (plot-based)
Post-treatment field plots (TSSM)
Wildfire Simulations
• Fire behavior model: Randig
o
uses Minimum Travel Time algorithm incorporated in FlamMap
o
allows for multiple wind directions
o
generates burn probability based on many simulated fires (n = 5000)
• Weather:
97th percentile fuel
moistures
o
19 mph winds, primarily
SW, but also NE, S
o
• Analysis:
calculate flame length
for crown fire initiation
per stand (n = 3300)
o
partition burn probability
output based on critical
flame length
o
Not treated
2010
2020
Treated
Not treated
2030
2040
Treated
Conditional burn probability (> critical flame length)
0.12
0.10
0.08
0.06
0.04
0.02
0.00
2010
No Treatment
Treated
2020
2030
Simulated calendar year
2040
Fire size in one burn period (ha)
5000
4000
3000
2000
1000
0
2010
2020
2030
No treatment
2040
2010
2020 2030
Treated
2040
(with maintenance treatments)
Fire
Name
Cause
Year
Final
Size (ha)
Stream
lightning 2001
1472
Boulder
Complex
lightning 2006
1388
Antelope
Complex
lightning 2007
9389
Moonlight accidental 2007
26,390
•Antelope Lake analysis
watershed: 18,426 ha
• Proportion of watershed burned
between 2001 and 2007: 56.4 %
Proportion of burned
area within watershed
(2001-2007)
Unchanged
8%
Low
13%
Moderate
26%
High severity
53%
Proportion of
watershed burned
RdNBR
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
Fire severity class (RdNBR-based)
High
Moderate
Low
Unchanged
Stream
(2001) Boulder (2006) Antelope (2007) Moonlight (2007)
50
40
30
Fire severity class
(RdNBR-based)
Unchanged Low
Moderate
High
Mean
Patch size (ha)
20
10
0
1100
1000
900
Area-weighted mean
2407 ha
400
300
200
100
0
Stream
(2001)
Boulder
(2006)
Antelope
(2007)
Moonlight
(2007)
Runoff and erosion in
Antelope Lake fires
GEOWEPP EROSION
MODELING
Sub-watersheds within each
fire perimeter (n = 3)
Soil burn severity
• runoff dominated by base
flow
> WEPP model augmented to
incorporate base flow
• erosion primarily in
channels
• little evidence for
differences in erosion among
simulated management
scenarios
> erosion driven by weather
rather than vegetation cover
Summary
•Fuel treatment tradeoffs/effects are
well documented at STAND-SCALE
•Current management challenge is to
‘scale-up’ to LANDSCAPES
> Area restrictions constrain ‘on
the ground’ implementation,
prohibiting theoretical treatment
designs
(recommend term ‘COORDINATED
LANDSCAPE FUEL TREATMENTS’)
•Noticeable reductions in crown fire
potential & modeled fire sizes, despite
deviation from planned network
>20 % of landscape treated
> Allocate proportions within
“problem fire” area
> Recommend planning for greater
proportions
Summary con’t
• Uncharacteristically large standreplacing patches pose a significant
management challenge
> sensitive species habitat
> reforestation
> recreation
> watershed impacts?
•When fully implemented coordinated
landscape fuel treatments introduce
discontinuity to homogenized
landscapes
Acknowledgements:
•Field work:
Kurt Menning
Bridget Tracy
Nick Delaney
•Data Analysis:
Kurt Menning
Collin Dillingham
Jay Miller
•Funding:
USDA FOREST SERVICE REGION 5:
Plumas/Lassen Administration Study
(http://www.fs.fed.us/psw/programs/snrc/)
Storrie Fire Restoration