The Physics and Ecology of Carbon Offsets
Dennis Baldocchi
Professor of Biometeorology
Ecosystem Sciences Division/ESPM
University of California, Berkeley
UC Merced Feb 24, 2010
If Papal Indulgences can save us from burning in Hell:
Can Carbon Indulgences save us from Global Warming?
Alexander VI
Sixtus IV
Innocent VIII
Julius II
Leo X
Does Planting a Tree Really Offset your Carbon Footprint?
Haiti vs Dominican Republic
Working Hypotheses
• H1: Forests Can Mitigate Global Warming
– Forests are effective and long-term Carbon Sinks
– Land-use change (via afforestation/reforestation) can help offset
greenhouse gas emissions and mitigate global warming
– Forests transpire water effectively, producing clouds, rainfall and
a lower planetary albedo
• H2: Forests Contribute to Global Warming
– Forests are optically dark and absorb more energy than short
vegetation
– Forests convect more sensible heat into the atmosphere than
grasslands, warming the atmosphere
• Landuse change (more forests) can help promote global warming
– Forests use more water than other vegetation, and water vapor
is a greenhouse gas and scarce resource in semi-arid regions,
eg California
Issues of Concern and Take-Home Message
•
Trees are Inefficient Solar Collectors
–
–
–
•
Ecological Scaling Laws Must be Obeyed when Planting Trees and Cultivating
Plantations
–
–
•
You need more than 500 mm of rain per year to grow Trees
Best and Moistest Land is Already Vegetated
New Forested Lands needs to take up More Carbon than current land use
It’s a matter of scale, A lot of ‘trees’ (more than the US land area) is needed to be planted to
offset our profligate carbon emissions
There are Energetic and Environmental Costs to soil, water, air by land use change
–
–
–
–
–
•
Self-Thinning Occurs with Time
Mass scales with the -4/3 power of tree density
There Must be Available Land and Water
–
–
–
–
•
Much vegetation operates less than ½ of the year and is a solar collector with less than 2%
efficiency
The Ability of Forests to sequester Carbon declines with stand age
Solar panels work 365 days per year and have an efficiency of 20%+
Forests are Darker than Grasslands, so they Absorb More Energy
Changes in Surface Roughness and Conductance and PBL Feedbacks on Energy Exchange
and Evaporative cooling may Dampen Albedo Effects
Forest Albedo changes with stand age
Forests Emit volatile organic carbon compounds, ozone precursors
Forests reduce Watershed Runoff and Soil Erosion
Societal/Ethical Costs and Issues
–
–
Land for Food vs for Carbon and Energy
Energy is needed to produce, transport and transform biomass into energy
Can we offset Carbon Growth by Planting Trees?
All Seedlings Do Not Grow to Maturity
Pine Seedlings in Finland
Old Growth Forest in Finland
http://www.helsinki.fi/~korpela/forestphotos.html
ESPM 111 Ecosystem Ecology
Yoda’s Self Thinning Law
• Planting trees may be a
‘feel-good’ solution, but it
is not enough
#N ~ Mass -3/4
Mass ~ #N -4/3
– self thinning will occur
• Energetics of Solar
Capture Drives the
Metabolism of the System
Enquist et al. 1998
Kleiber’s Law
Metabolic rate (B) of an
organism scales to the 3/4
power of its mass (M)
The Metabolic Energy needed to Sustain an
organism INCREASES with Mass, to the ¾
power
B M
3/ 4
ESPM 111 Ecosystem Ecology
Energetics at Landscape scale is Scale Invariant
• Energetics/Metabolism of the
System <B> is weighted by the
sum of the product of the
Energetics of class, i, times the
number of individuals in this
class, N
• Energy/Metabolism, B, scales
with Mass, M, to the ¾ power,
Kleiber’s Law
• Number of Individuals scales
with Mass to the -3/4 power,
modified Yoda’s Law
• Energy/Metabolism of the
System is scale invariant with
Mass, exponent equals zero
 B   Bi Ni
B M
3/ 4
Mass
Number 
a
BT  N i Bi
M
3/4
i
M
3/4
i
3/4
~M
0
Energy Drives Metabolism:
How Much Energy is Available and Where
FLUXNET database
6000
-2
-1
Rg (MJ m y )
8000
4000
2000
0
0
10
20
30
40
50
Latitude
60
70
80
90
Potential and Real Rates of Gross Carbon Uptake by Vegetation:
Most Locations Never Reach Upper Potential
GPP at 2% efficiency and 365
day Growing Season
tropics
GPP at 2% efficiency and
182.5 day Growing Season
FLUXNET 2007 Database
Probability Distribution of Published GPP Measurements, Integrated Annually
FLUXNET Database
0.05
0.04
mean= 1033+/- 631 gC m-2 y-1
n=253
pdf
0.03
0.02
0.01
0.00
0
1000
2000
3000
GPP (gC m-2 y-1)
Global GPP = 1033 * 110 1012 m2 = 113.6 PgC/y
4000
Ecosystem Respiration (FR) Scales Tightly with Ecosystem Photosynthesis (FA),
But Is with Offset by Disturbance
4000
Undisturbed
Disturbed by Logging, Fire, Drainage, Mowing
3500
FR (gC m-2 y-1)
3000
2500
2000
1500
1000
500
0
0
500
1000
1500
2000
2500
FA (gC m-2 y-1)
Baldocchi, Austral J Botany, 2008
3000
3500
4000
Probability Distribution of Published NEE Measurements, Integrated Annually
FLUXNET Database
0.10
0.08
mean= -225 +/- 227 gC m-2 y-1
n=254
pdf
0.06
0.04
0.02
0.00
-1600 -1400 -1200 -1000 -800
-600
-400
-200
NEE (gC m-2 y-1)
0
200
400
600
It’s a matter of scale
A lot of ‘trees’ need to be planted to offset our profligate carbon use
• US accounts for about 25% of Global C emissions
– 0.25*8.0 1015 gC = 2.0 1015 gC
• Per Capita Emissions, US
– 2.0 1015 gC/300 106 = 6.66 106 gC/person
• Ecosystem Service, net C uptake, above current rates
– ~200 gC m-2
• Land Area Needed to uptake C emissions, per Person
– 3.33 104 m2/person = 3.33 ha/person
• US Land Area
– 9.8 108 ha
– 10.0 108 ha needed by US population to offset its C emissions
Naturally!
1. Carbon is Lost with Disturbance
2. Net Carbon Uptake Decline with Age
Sink
Interannual
Variability
400
-2
-1
Carbon Flux (g C m y )
600
200
0
Gap
-200
-400
Source
-600
OA
SOBS
OJP
NOBS
HJP94
F89
F98
HBS00
WP39
DF49
HDF88
HDF00
HJP02
EOBS
OMW
HJP75
BF67
-800
0
20
40
60
80
100
120
140
160
180
Stand Age (years)
Brian Amiro 2007 AgForMet
All Land is Not Available or Arable:
You need Water to Grow Trees!
Scheffer al 2005
Carbon sequestration by plantations can dry out streams
[Jackson, et al., 2005, Science].
Working Hypotheses
• H1: Forests have a Negative Feedback on Global
Warming
– Forests are effective and long-term Carbon Sinks
– Landuse change (more forests) can help offset greenhouse gas
emissions and mitigate global warming
• H2: Forests have a Positive Feedback on Global
Warming
– Forests are optically dark and Absorb more Energy
– Forests have a relatively large Bowen ratio (H/LE) and convect
more sensible heat into the atmosphere
– Landuse change (more forests) can help promote global
warming
Land Use Affects Mass and Energy Exchange
Forests are Darker and Possess Lower Albedos than Crops/Grasslands
Crop
Deciduous
Forest
Conifer Forest
Evergreen Broadleaved Forest
O’Halloran et al. NCEAS Workshop
Its not Only Carbon Exchange:
Albedo Changes too with planting Forests
Summertime Albedo
0.20
AK
MB
SK
SOA
0.15
0.10
0.05
0.00
0
25
50
75
100
125
150
Time since disturbance (years)
Amiro et al 2006 AgForMet
175
Should we cut down Dark Forests to Mitigate Global Warming?:
UpScaling Albedo Differences Globally, part 1
•
•
•
•
•
Average Solar Radiation: ~95 to 190 W m-2
Land area: ~30% of Earth’s Surface
Tropical, Temperate and Boreal Forests: 40% of land
Forest albedo (10 to 15%) to Grassland albedo (20%)
Area-weighted change in Net Solar Radiation: 0.8 W m-2
– Smaller than the 4 W m-2 forcing by 2x CO2
– Ignores role of forests on planetary albedo, as conduits of water
vapor that form clouds and reflect light
– Discounts Ecological Services provided by Forests
Other Complications associated with
Reliance on Forest/Carbon Sequestration
•
•
•
•
Fire
Nutrient Requirements
Ecosystem Sustainability
Deleterious effects of Ozone, Droughts
and Heat Stress
• Length of Growing Season
• Ecosystem Services
– Habitat, Soil Erosion Prevention, Biodiversity
Should we cut down dark forests to Mitigate Global Warming?:
UpScaling Albedo Differences Globally, part 2
km2
MJ m-2 y-1 albedo albedo
area
rad
change
wt value
tropical
1.75E+07 6.00E+09
0.05
0.15 5.25E+15
temperate 1.00E+07 5.00E+09
0.05
0.15 2.50E+15
boreal
1.30E+07 4.00E+09
0.1
0.1 5.20E+15
Earth
5.10E+08
ave
sum
1.30E+16
time/land
0.805 W m-2
Case Study on Energetics of Land
Use Change:
Comparative study of an Annual
Grassland and Oak Savanna
Case Study:
Savanna Woodland adjacent to Grassland
Solar Radiation
Net Radiation, Grassland
Net Radiation, Savanna
2006
30
-2
-1
Energy Flux Density (MJ m d )
35
25
20
15
10
5
0
0
50
100
150
200
250
300
350
Day
1. Savanna absorbs much more Radiation (3.18 GJ m-2 y-1) than
the Grassland (2.28 GJ m-2 y-1) ; DRn: 28.4 W m-2
0.30
Grassland
Savanna
2006
0.25
PAR Albedo
0.20
0.15
0.10
0.05
0.00
0
50
100
150
200
250
Day
2. Grassland has much great albedo than savanna;
300
350
Landscape Differences
On Short Time Scales, Grass ET > Forest ET
Monthly Averages
1.0
LE/LEeq
0.8
0.6
0.4
0.2
Savanna Woodland
Annual Grassland
0.0
0
2
4
6
8
Gs (mm s-1)
Ryu, Baldocchi, Ma and Hehn, JGR-Atmos, 2008
10
12
14
16
Role of Land Use on ET:
On Annual Time Scale, Forest ET > Grass ET
California Savanna
440
420
-1
Evaporation (mm y )
400
380
360
340
320
300
Oak Woodland
Annual Grassland
280
260
240
02_03
03_04
04_05
05_06
Hydrological Year
Ryu, Baldocchi, Ma and Hehn, JGR-Atmos, 2008
06_07
4a. U* of tall, rough Savanna > short,
smooth Grassland
2002
4b. Savanna injects more Sensible
Heat into the atmosphere because
it has more Available Energy and it
is Aerodynamically Rougher
0.4
0.3
0.2
0.1
14
0.0
0.4
0.6
0.8
1.0
-1
0.2
u*, oak woodland, daily average
Grassland
Savanna
2006
12
-2
0.0
Sensible Heat Flux Density (MJ m d )
u*, grassland, daily average
0.5
10
8
6
4
2
0
0
50
100
150
200
Day
250
300
350
2006, Ione, CA
40
Potential Temperature, Oak Savanna
30
Air Temperature (C)
315
Grassland
Savanna
2006
20
10
0
310
305
300
295
290
285
b[0] -2.67
b[1] 1.012
r ² 0.953
280
275
0
50
100
150
200
Day
250
300
350
275
280
285
290
295
300
305
Potential Temperature, Grassland
5. Mean Potential Temperature differences are relatively small (0.84 C;
grass: 290.72 vs savanna: 291.56 K); despite large differences in Energy
Fluxes--albeit the Darker vegetation is Warmer
Compare to Greenhouse Sensitivity ~2-4 K/(4 W m-2)
310
315
Conceptual Diagram of PBL Interactions
3000
Shortwave
Energy
2500
Longwave
Energy
pbl ht (m)
2000
1500
1000
PBL Height
500
Time 3
0
6
8
10
12
14
16
18
Time (hrs)
Time 2
3000
Vapor Pressure
Temperature
Time 1
2500
ea (Pa)
2000
Latent Heat
Sensible Heat
1500
1000
500
0
6
8
10
12
14
Time (hrs)
H and LE: Analytical/Quadratic version of Penman-Monteith Equation
16
18
ET-PBL Oak-Grass Savanna Land Use
•The Energetics of
afforestation/deforestation
is complicated
T surface
310
300
290
albedo = 0.3; Rc=2560 s/m
albedo = 0.15; Rc = 320 s/m
•Forests have a low
albedo, are darker and
absorb more energy
280
6
8
10
12
14
16
18
20
Time
800
•But, Ironically the darker
forest maybe cooler (Tsfc)
than a bright grassland due
to evaporative cooling
Rnet (W m-2)
600
400
200
albedo = 0.3; Rc=2560 s/m
albedo = 0.15; Rc = 320 s/m
0
6
8
10
12
Time
14
16
18
20
ET-PBL Oak-Grass Savanna Land Use
250
200
LE (W m-2)
•Forests Transpire
effectively, causing
evaporative cooling, which
in humid regions may form
clouds and increase
planetary albedo
•Due to differences in
Available energy,
differences in H are smaller
than LE
150
100
albedo = 0.3; Rc=2560 s/m
albedo = 0.15; Rc = 320 s/m
50
0
6
8
10
12
14
16
18
20
16
18
20
Time
500
H (W m-2)
400
300
200
albedo = 0.3; Rc=2560 s/m
albedo = 0.15; Rc = 320 s/m
100
0
Axel Kleidon
6
8
10
12
Time
14
Theoretical Difference in Air Temperature: Grass vs Savanna
ET-PBL Model
304
albedo = 0.15; Rc=320 s/m
albedo = 0.30; Rc = 2560 s/m
Air Temperature, K
302
300
298
296
294
292
290
4
6
8
10
12
14
Time
Summer Conditions
16
18
20
Temperature Difference Only Considering Albedo
ET-PBL Model
296
Air Temperature, K
294
albedo = 0.25; Rc = 160 s/m
albedo = 0.15; Rc = 160 s/m
292
290
288
286
4
6
8
10
12
Time
Spring Conditions
14
16
18
20
Tsfc can vary by 10 C
by changing albedo and Rs
304
303
315
302
301
305
300
300
299
298
295
Tair can vary by 3 C by
changing albedo and Rs
297
290
0.5
296
0.4
800
600
0.3
400
0.2
Albedo
295
294
200
0.1
0
Rs
292.5
293
292
292
291.5
291
Tair
Tsfc
310
291
290
289
290.5
288
0.5
290
0.4
800
600
0.3
400
0.2
Albedo
200
0.1
0
Rs
289.5
Tair can vary by 3 C by
changing Ra and Rs
292
293
291.5
292
291
290
290.5
289
288
80
Tsfc can vary by 10 C
by changing Ra and Rs
290
60
800
600
40
Ra
289.5
400
20
200
0
0
Rs
306
315
304
310
302
305
Tsfc
Tair
291
300
300
295
298
290
80
296
60
800
600
40
400
20
Ra
200
0
0
Rs
294
Future Carbon Emissions: Kaya Identity
C Emissions = Population * (GDP/Population) *
(Energy/GDP) * (C Emissions/Energy)
•Population
•Population expected to grow to ~9-10 billion by
2050
•Per capita GDP, a measure of the standard of living
•Rapid economic growth in India and China
•Energy intensity, the amount of energy consumed
per unit of GDP.
•Can decrease with efficient technology
•Carbon intensity, the mass of carbon emitted per
unit of energy consumed.
•Can decrease with alternative energy
CO2 in 50 years, at Steady-State
• 8 GtC/yr, Anthropogenic Emissions
– 45% retention
• 8 * 50 * 0.45 = 180 GtC, integrated Flux
• Each 2.19 GtC emitted causes a 1 ppm increase in
Atmospheric CO2
• 833 (@380 ppm) + 180 = 1013 GtC, atmospheric
burden
• 462 ppm with BAU in 50 years
– 1.65 times pre-industrial level of 280 ppm
• BAU C emissions will be ~ 16 to 20 GtC/yr in 2050
• To stay under 462 ppm the world can only emit 400
GtC of carbon, gross, into the atmosphere!
How Does Energy Availability Compare with Energy Use?
• US Energy Use: 105 EJ/year
– 1018J per EJ
– US Population: 300 106
– 3.5 1011 J/capita/year
• US Land Area: 9.8 106 km2=9.8 1012 m2 = 9.8 108 ha
• Energy Use per unit area: 1.07 107 J m-2
• Potential, Incident Solar Energy: 6.47 109 J m-2
– Ione, CA
• A solar system (solar panels, biomass) must be at least 0.1%
efficient, working year round, over the entire surface area of the US
to capture the energy we use to offset fossil fuel consumption
• Assuming 20% efficient solar system
– 8.11 1010 m2 of Land Area Needed (8.11 105 km2, the size of South
Carolina)
Concluding Issues to Consider
• Vegetation operates less than ½ of the year and is a solar collector
with less than 2% efficiency
– Solar panels work 365 days per year and have an efficiency of 20%+
• Ecological Scaling Laws are associated with Planting Trees
– Mass scales with the -4/3 power of tree density
• Available Land and Water
– Best Land is Vegetated and New Land needs to take up More Carbon
than current land
– You need more than 500 mm of rain per year to grow Trees
• The ability of Forests to sequester Carbon declines with stand age
• There are Energetics and Environmental Costs to soil, water, air and
land use change
– Changes in Albedo and surface energy fluxes
– Emission of volatile organic carbon compounds, ozone precursors
– Changes in Watershed Runoff and Soil Erosion
• Societal/Ethical Costs and Issues
– Food for Carbon and Energy
– Energy is needed to produce, transport and transform biomass into
energy
– Forests play positive roles for habitat, biodiversity, carbon storehouses
and resources