Map Units for LANDFIRE: Integrating Vegetation Classification and Map Legends

1
Map Units for LANDFIREIntegrating Vegetation
Classification and Map Legends
2
LANDFIRE DELIVERABLES
VEGETATION Existing vegetation composition Existin
g vegetation structure Biophysical Settings
FIRE ECOLOGY Layers Historical fire return
interval Historical fire severity Historical fire
regime Current Succession Class Vegetation
departure Fire Regime Condition Classes
FIRE BEHAVIOR/FIRE EFFECTS Layers Fire behavior
fuel models Canopy bulk density Canopy base
height Canopy cover Canopy height FCC
Fuelbeds Fuel Loading Models

3
LANDFIRE DELIVERABLES
VEGETATION Existing vegetation composition Existin
g vegetation structure Biophysical Settings
FIRE BEHAVIOR/FIRE EFFECTS Fire behavior fuel
models Canopy bulk density Canopy base
height Canopy cover Canopy height FCC
Fuelbeds Fuel Loading Models
FIRE ECOLOGY Historical fire return
interval Historical fire severity Historical fire
regime Current Succession Class Vegetation
departure Fire Regime Condition Classes

4
LANDFIRE Map Unit Development

• All Lands Vegetative Communities
• Same level of detail
• Forestlands, Shrublands, and Grasslands
• Repeatable
• Quick and affordable
• Target Map Accuracies
• 60 to 80 percent map accuracy
• Consistent for the Map Extent (National)
• Map units mean the same thing in Florida as they
  do in Colorado

5
Map Unit Requirements

• Identifiable
• from field or plot data
• Map-able
• 30 meter resolution
• 60-80 accurate
• Scalable
• link with existing classifications
• Model-able
• provide required model inputs

6
LANDFIRE Vegetation Layers
Potential Vegetation Type (PVT) or Biophysical
Setting (BpS)
Existing Vegetation Type (Species Composition)
Existing Structural Stage
7
Existing Vegetation TypeOther Efforts
Alliances and Associations of the USNVC (Grossman
and others 1998) Sagebrush cover type map
(SAGEMAP 2002) and classification (Reid and
others 2002) developed by USGS Forest Cover
Types of the United States and Canada, Society of
American Foresters (Eyre 1980) Rangeland Cover
Types of the United States, Society for Range
Management (Shiflet 1994) GAP Cover Types for the
eleven western states complied by the BLM
8
Map Unit Requirements

• Identifiable
• from field or plot data
• through dominance of species or groups of species
  on the plots
• through individual or groups of indicator species
  on plots
• dichotomous Field Key with field and floristic
  criteria
• Sequence Table for plot data with floristic
  criteria

9
Sequence Tables

• Criteria
• Absolute cover for lifeforms
• Relative cover for floristic criteria

10
Sequence Tables

• Criteria
• Absolute cover for lifeforms
• Relative cover for floristic criteria
• Automation
• BpS_EVT_Key_Classifier
• Summaries by BpS, EVT, BpS/EVT
• Constancy/Cover by BpS, EVT, BpS/EVT

11
Map Unit Requirements

• Map-able
• Final Cleanup Sequence Tables
• Mappers incorporate QA/QC for plots during
  mapping process and update MAT (contains training
  plots)
• NatureServe runs through unclassified plot data
  and applies a qualitative classification
• LFRDB determines what plot data is releasable and
  not releasable
• Final MAT posted and available

12
Map Unit Requirements

• Scalable
• Meet different scaling needs, fine to broad, by
  linking to existing classification crosswalks
• Maintain continuity between maps of different
  scales.

13
Map Unit Requirements

• Model-able
• Provide correct combos for fuel lookup tables
  and inputs
• Anomalous combinations of vegetation composition,
  structure and site potential will not have plot
  data and thus no associated fuels inputs

14
Mapping Fuels for LANDFIREIntegrating Remote
Sensing, GIS, and Biophysical Modeling
15
Why are Fuels Important?The one factor over
which we have the most control
16
Fuels MapsThe Most Important Fire Management
LayerPotential Uses

• Predict future growth of fire
• Develop fire danger, hazard, risk layers
• Plan future fires and prioritize treatments
• Simulate fire effects-smoke, tree mortality
• Evaluate management alternatives

17
What Are Fuels? Live and dead biomass

• Biomass when burned
• Contributes to fire propagation
• Produces smoke
• Generates heat to kill flora fauna

18
Challenges in Mapping Fuels

• Canopy obstruction
• Fuel bed diversity
• Fire behavior fuel models
• Fuel complexity
• Fuel variability
• Fine resolutions

19
Fuel Bed DiversityMany categories of fuels
Dead and Live Crown foliage and branchwood
Arboreal Mosses and Lichens
Live and Dead Tree Regeneration
Live and Dead Shrub and Herbaceous
Logs
Downed Dead Woody Twig and Branchwood
Litter and Duff
Cones, Buds, Mosses, Lichens
20
Fuel Complexity

• Each fuel type important to one, but not all,
  fire applications
• Fire behavior needs description of fine fuels
• Smoke prediction requires description of all fuel
  types
• Fuel models and fuel classifications must be
  robust

21
Fuel Variability

• Fuels are continuous not discrete
• Highly variable in space and time
• Related to many factors
• Stand history
• Biophysical setting
• Community composition
• Stand Structure

22
Fuel Mapping Approaches

• Field Reconnaissance
• Remote Sensing

23
Fuel Mapping Strategy

• Correlated with many ecosystem attributes
• Governs fuel dynamics
• Classifications available

• Species Composition
• Stand Structure
• Biophysical Setting

24
Fuel Variability Example
FBFM 9 - Pine Litter
FBFM 2 - Conifer Grass
FBFM 5 - Live Shrub
25
Remote sensing
These systems characterize the physical and
chemical properties of atmospherically
transmitted radiation. The reflected radiation is
coupled with atmospheric models and fitted to
geographic location, time and date to determine
apparent surface reflectance.
This remotely sensed data can be either directly
or indirectly related to identifiable materials
such as shade, various soils, non-photosynthetic
vegetation, green biomass, live fuel moisture,
diverse vegetation species and unique land-cover
types.
26
LANDFIRE Fuels
Two major divisions of fuels are recognized by
the LANDFIRE project surface fuels and canopy
fuels.

• Surface fuels are those biomass components that
  occur on the ground (less than six feet tall) and
  are the fuels that carry a surface fire.
• Live or dead, herbaceous or shrub, downed dead
  woody, litter, and duff
• Fire Behavior Fuel models have been developed
  to predict fire behavior.
• Canopy fuels are those aerial biomass components
  higher than six feet that can carry a crown fire
  and are usually consumed in the crown fire.
• 1) Bulk density (kg m-3),
• 2) Canopy cover (),
• 3) Canopy height (m),
• 4) Canopy base height (m).

27
FARSITE Input Layers
Base Vegetation Layers
Ancillary Layers
28
METHODS

• Creating the surface fuel maps
• Create lookup tables for FBFM13
• Creating the surface fuel model maps from the
  lookup tables

29
Fire Behavior Fuel Models

• Describes expected fire behavior
• Not a description of actual fuel conditions
• Complicated procedure to construct models
• Fuel model construction subjective
• Assessment in field is subjective

30
METHODS continued

• Creating the canopy fuel maps
• Calculate all four canopy characteristics for all
  plots in the LANDFIRE reference database with
  comprehensive tree data using the FUELCALC
  program
• Classification and regression trees were used to
  link the calculated reference data to Landsat
  satellite imagery and a series of 30-meter,
  spatially-explicit gradient layers representing
  climate, fire ecology, soil, and topography.

31
METHODS continued

• Performing QA/QC procedures
• Create fuels QA/QC ruleset
• Check all layers for data gaps
• Check all layers for logic inconsistencies within
  LANDFIRE layers
• Performing accuracy assessment
• Calculate accuracy of statistical models
• Calculate classification accuracy of fuel model
  keys
• Calculate pixel accuracy of fuel maps
• Calculate mapping accuracy of fuel maps

32
LANDFIRE Fuel Layers

• Standard 13 Fire Behavior Fuel Models (FBFM13).
• Canopy bulk density (CBD)
• Canopy cover (CC)
• Canopy height (CH)
• Canopy base height (CBH)

33
Fire Behavior Fuel Model
34
Canopy Base Height
35
Canopy Height
36
Canopy Cover
37
Crown Bulk Density
38
LANDFIRE Fuel Layers

• Standard 13 Fire Behavior Fuel Models (FBFM13).
• Canopy bulk density (CBD)
• Canopy cover (CC)
• Canopy height (CH)
• Canopy base height (CBH)

39
LANDFIRE Fuel Layers

• Standard 13 Fire Behavior Fuel Models (FBFM13).
• Canopy bulk density (CBD)
• Canopy cover (CC)
• Canopy height (CH)
• Canopy base height (CBH)
• plus
• New 40 Fire Behavior Fuel Models (FBFM40)
• Fuel Characteristic Classification System (FCCS)
• Fuel Loading Models (FLM)

40
Other Analysis ToolsNew Fuel Models and Fuel
Classifications
New
Fire behavior,
Fire effects
Fuel Models
41
New set of 40 fire behavior fuel models (FBFM40)

• The new set of 40 fire behavior fuel models
  (FBFM40) are hierarchically organized by fuel
  strata and fuel loading.
• The 40 fuel models have already been implemented
  into the BEHAVE fire modeling system and the
  FARSITE fire growth model.
• Subtle modifications in fuelbeds as a result of
  fuel treatment activities should be represented
  by these 40 fuel models.

42
The FCCS Fuelbed concept was developed by the
Fire and Environmental Research Applications
(FERA) at PNW, Seattle. It includes complete
descriptions of typical fuel situations around
the nation. The FCCS summarizes fuel by component
using canopy, ground, and surface fuel
stratifications.
LANDFIRE Prototype Zone 16
Fuel Characteristic Classification System (FCCS)
National Fuelbed Map
43
Questions???