Title: Mapping Wildfire Disturbances in Southern
1 John Rogan jrogan_at_clarku.edu Geography
Department Clark University
Mapping Wildfire Disturbances in Southern
California Using Machine Learning Algorithms
2Research Context
- Land cover/use change Mapping and Monitoring
- Growing interest in Rapid Response Information
Systems - Empirical information about wildfire cause and
behavior can inform wildfire risk analysis - Paucity of historic burn severity information
3Wildfire Monitoring Programs
4Monitoring Fire Effects
- The physical environment and its response to fire
- Factors affecting fire behavior
- Ecosystem/watershed damage assessment
- Evaluating success of a management ignited fire
- Appraising the potential for future treatments
5Existing Monitoring Efforts (BAER) Watershed Scale
Source California Department of Forestry and
Fire Protection
6Existing Monitoring Efforts Ecoregion Scale
Source California Department of Forestry and
Fire Protection
7Existing Monitoring Efforts National Scale
A Detected from satelliteB Conventional
methods
Source Canadian Center for Remote Sensing
8Existing Monitoring Efforts Global Scale
Source University of Maryland Global Fire
Product
9Research Objectives
- Test a new methodology to map fire severity in
San Diego County (1985-2000) - Employ machine learning to map severity, while
integrating environmental variables with spectral
variables (categorical and continuous) - Examine the contribution of ancillary variables
to burn map accuracy
10Fire/Burn Severity
- Severity - A descriptive term that integrates
various phenomenological characteristics of a
fire-altered landscape - Physical and biological manifestation of
combustion on vegetation and soil - Direct Effects Influenced by
- Fuel consumption - Topography
- Crown scorch - Disturbance history
- Soil heating
- Bole Charring
11Previous Research
- Focus on mapping burn scars (coarse resolution)
- Recent emphasis on burn severity/mortality/damage
levels (fine-medium resolution) for impact
assessment - Retrospective burn area mapping at medium
resolution (e.g., Hudak and Brockett 2004IJRS) - BUT, challenges remain
12Scene Model
13Post-Fire IKONOS-2 Image
14California Wildfire Threat
Source California Department of Forestry and
Fire Protection
15San Diego County
16Study Area Significance
- Impacts of natural disturbance processes are
increasing in severity - Public lands began burning more frequently than
private lands in the mid-1970s. This trend is
increasing - Population increase and
- peri-urban spread into
- fire-prone areas (WUI)
17Landsat TM and ETM Data
18Environmental Variables
19Ground Reference Data
Composite Burn Index Key and Benson (2000)
- SITE-Dominance
- Grassland (6)
- Chaparral (10)
- Conifer-Hardwood (5)
- Mixed (12)
20Wildfire Effects (After Key and Benson)
21Methods (Data Processing Flow)
22Haze Correction (Pre-)
Band 1
Band 2
(a)
(b)
Band 3
Bands 4,3,2
23Haze Correction (Post-)
(a)
(b)
24Spectral Mixture Analysis
- Decomposition of mixed
- pixel spectral response
- Production of fractional
- representation of sub-
- pixel proportions
- Biophysically-meaningful
- estimates of land cover
- components
25Classification Tree Analysis
- A type of MLA used to predict membership of cases
of a categorical dependent variable from their
measurements on one or more predictor variables - Hierarchical, non-linear recursive partitioning
- Structurally explicit
26Desired Map Accuracy
Source Rogan and Franklin (2001)
27Case Study (Pre-Fire)
28Case Study (Post-Fire)
29Case Study Results - SMA
Shade
GV
BV
Soil
Vegetation Map
RMS
Fire perimeter
30Case Study Results - Variable Selection
31Case Study Results Burn Severity
ACCURACY (kappa)
CLASS
No burn Vegetation 90
No burn Water 100
Severe burn 87
82.5
Moderate burn 60
Light burn 74
No burn Bare Soil 84
32County-Wide Results
- Variable Selection by Site(s)
- Grass Burn, Soil
- CHP Burn, GV, Soil, Veg, Slope
- CON/HDW Burn, GV, Soil, Veg, Slope, Shade
- Mixed Burn, GV, Slope, Veg, Shade, Slope,
Aspect - Mean Burn Map Accuracy by Site(s)
- Grass 87 (SD 11)
- CHP 81 (SD 10)
- CON/HDW 84 (SD 7)
- Mixed 70 (SD 16)
33County-Wide Results
- Time since fire (TSF)
- Most problematic for grasslands, where TSF gt 3
months - Least problematic for CHP, CON/HDW
- Map accuracy
- Lowest for complex classes (e.g., mixed)
- Highest for simple classes (e.g., grassland)
- Variable Selection
- Many for complex classes (e.g., mixed)
- Few for simple classes (e.g., grassland)
34Implications
- Map accuracy
- Range 70-80, depending on landscape type and
TSF - Subtle burn classes are least accurate
- Variable Selection
- Varied by landscape type (all used for complex
areas) - Implication for fire risk mapping?
- The larger the fire, the greater the potential
for confusion caused by landscape heterogeneity
35Wildland Fire Mapping Triangle
Burn Severity Map
Predictive Vegetation Modeling
Image Processing and Enhancement
Machine Learning
.search for standard methods for mapping fuels
and fire regimes at high (spatial) resolutions
over broad areas. Rollins et al. (2004, p.
86)
36Acknowledgements
- NASA Land Cover Land Use Change Program
- US Forest Service and CDF
- SDSU Janet Franklin and Doug Stow
- UCSB Dar Roberts and Alexandria Digital Library
- U of Arizona Steve Yool