hourly biomass burning emissions inventory derived … · summary • modis vegetation...
TRANSCRIPT
HOURLY BIOMASS BURNING EMISSIONS INVENTORY DERIVED FROM GOES DATA
Xiaoyang Zhang, Shobha Kondragunta, and M. K. Rama Varma Raja
17th EI conference, June 2-5, 2008
Outline1. Fuel loading dataset derived from MODIS
(Moderate Resolution Imaging Spectroradiometer) data2. Burned area simulated from GOES
(Geostationary Operational Environmental Satellites) fire data 3. Biomass burning emissions across the
Continuous USA4. Summary
Modeling Emissions from Biomass Burning
GOES fire size
Fuel type AVHRR moisture condition
MODIS vegetation properties
Emissions
Fuel loading Fraction of fuel consumption
Emission FactorBurned area
Model of Biomass Burning Emissions
E---biomass burning emissions (kg)A---burned area (km2)M--biomass density/fuel loading (kg.km-2)C--fraction of combustionF--fraction of emissioni and j define the fire (pixel) locations l is the fuel typek is the time period
(1)∑∑∑∑= = = =
=K
k
L
l
J
j
I
iijklijklijkijkl FCMAE
1 1 1 1
GOES fire productMODIS vegetation propertiesAVHRR moisture conditionAVHRR moisture condition
MODIS Vegetation Property-based Fuel System (MVPFS)
Forest foliage
Forest branch
GrassShrub
Litter
Coarse woody detritus
Vegetation Cover Fraction (MOD44B)
LAI (MOD15A2)
Land Cover Type (MOD12Q1)
tons/ha
Comparison of Fuel Loadings from NFDRS, FCCS and MVPFS
McNally fire area (35º49’-36º18’N, 118º10’-118º37’W).
TM TM NFDRS
FCCS MVPFS
Fire in July 2002
Simulation of Burned Area From GOES Fire Size
Fire occurrences detected from GOES satellite in 2002
•Spatial resolution: 4km•Temporal resolution: 30min• Instantaneous fire sizes in subpixels detected from 3.9 µm and 10.7 µm infrared bands
Subpixel fire size in GOES Fire Product
0%
20%
40%
60%
80%
100%
2002 2003 2004Fire-size processed Fire-pixel saturated Fire-pixel cloudyHigh probability fire Medium probability fire Low probability fire
With fire size
Without fire size
Low probability fire
Fourier-Fitted Diurnal GOES Fire Sizes---Representative diurnal curves
00.05
0.10.15
0.20.25
0.30.35
0.4
0 5 10 15 20Solar local time (hour)
Fire
siz
e (k
m2)
ForestSavannaShrubgrasscrop
Fitting Diurnal Fire Size for Individual GOES Fire Pixels
0
0.2
0.4
0.6
0 5 10 15 20
Solar local time (hour)
Fire
siz
e (k
m2)
Representive diurnal curvecurveFir size detected fromGOES Simulated diurnal curve
Burned Areas Converted from GOES Fire Size
FA α=A -- the area burned within a specified time period (km2) F -- the subpixel fire size (km2)α -- a constant coefficient
Training and Testing Burned Area Detected from TM Imagery
483km2
311km2
Difference: (483-311)/311=55%TM imagery and Fire Perimeter obtained from National Park Service-USGS National Burn Severity Mapping Project
Comparison of Simulated GEOS Burned Areas with TM Burn Scars
0
200
400
600
800
1000
1200
1400
1600
0 200 400 600 800 1000 1200 1400 1600
TM-based burn scars (km2)
GO
ES b
urne
d ar
eas
(km
2) GOES_area=23.16+0.95TM_scarR2=0.96n=20
Comparison of Simulated GOES Burned Area with National Inventory
0200400600800
100012001400
0 200 400 600 800 1000 1200 1400Burned area from Natioanl Inventory (NEI) data (km2)
Sim
ulat
ed a
rea
Simulated Burnt Area from GOES Fire Size vs Burned Area from NEI in 2002
1:1
Burned Area Simulated from GOES Fire Size(2006)
GOES Pixel-based Burned Areas(2000-2006)
GOES Emission Evaluation: CMAQ-Based CO at 500 mb and Surface Layer
Provided by George Pouliot
GOES Emission Evaluation: Comparison of AIRS CO with CMAQ (Atmospheric Infrared Sounder)CO Generated with and without Fire Emissions
Annual PM2.5 Emissions
Biomass Burning Emissions (CH4, CO and NOX) in 2005
Annual Emissions for Different Species (106kg)
3257.32081.31684.73109.53226.7CO
309.6199.9160.8296.4307.8PM2.5
228.0145.7117.9217.65225.9TNMHC
168.9111.688.6163.2170.3NOX
129.682.767.0123.7128.3CH4
52.034.427.350.352.5SO2
32.620.816.831.132.3NH3
9.96.65.29.610.0N2O
20062005200420032002
San Diego Fire Emissions in October 2007
Florida Fire Emissions in May 2007
Diurnal Variations in Biomass Burning Emissions for Different Years
Diurnal Pattern in Emissions in Different Ecosystems
PM2.5
0
5000
10000
15000
20000
0 2 4 6 8 10 12 14 16 18 20 22 24Local solar time (hour)
PM2.
5 (M
g)
ForestSavannaShrubGrassCrop
aNOX
0
2000
4000
6000
8000
10000
0 2 4 6 8 10 12 14 16 18 20 22 24Local solar time (hour)
NO
X (M
g)
ForestSavannaShrubGrassCrop
b
CO
0
50000
100000
150000
200000
250000
0 2 4 6 8 10 12 14 16 18 20 22 24Local solar time (hour)
CO
(Mg)
-
ForestSavannaShrubGrassCrop
c
Seasonal Shift of Diurnal Patterns in Burning Emissions.
CO (Mg)
0
20000
40000
60000
80000
100000
120000
0 2 4 6 8 10 12 14 16 18 20 22 24Local solar time (hour)
CO
(Mg)
-
JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember
Seasonal Variations in Monthly Emissions.
Monthly variation
0
10000
20000
30000
40000
50000
60000
70000
80000
0 2 4 6 8 10 12Month
Em
issio
n (M
g)
0
100000
200000
300000
400000
500000
600000
700000
CO
em
issio
n (M
g)
N2ONH3SO2CH4NOXTNMHCPM2.5CO
Variations in Biomass Burning Emissions (CO) for Different States
Month with Severe Biomass Burning Emissions from 2002-2006
Correlation of Times Series of PM2.5 Emissions with Precipitation
in California
0
5000
10000
15000
20000
25000
30000
January
April
July
October
January
April
July
October
January
April
July
October
January
April
July
October
January
April
July
October
PM2.
5 (M
g)
020406080100120140160180200
Mon
thly
rai
nfal
l (m
m)
PM2.5 (Mg)Monthly rainfall (mm)
2002 20052003 2004 2006
Controls of Drought on Biomass Burning Emissions
California (2002-2006)
0
5000
10000
15000
20000
25000
30000
0 50 100 150 200 250 300Monthly rainfall (mm)
PM2.
5 (M
g)
Y=12167.29X-0.9449
R= -0.71n=60
Florida (2002-2006)
0
2000
4000
6000
8000
10000
0 50 100 150 200 250 300Monthly precipitation (mm)
PM2.
5 (M
g)
Y=37171.17X-0.9094
R= -0.58n=60
Summary• MODIS Vegetation Property-based Fuel System
(MVPFS) provides realistic fuel loading data and can be updated easily.
• Burned area at high temporal and spatial resolution can be simulated from GOES active fire observations with a reliable quality.
• Hourly biomass burning emissions (PM2.5, CH4, CO, N2O, NH3, NOX, SO2, and TNMHC) inventory can be created in near real time from satellite-based fuel data and burned area.
• Biomass burning emissions inventory developed in this study reveals distinct variations in diurnal, seasonal, and interannual patterns and spatial shifts in states and ecosystems.