Dennis Baldocchi University of California, Berkeley Brazil Flux Eddy Covariance Meeting

Dennis BaldocchiUniversity of California, Berkeley

Brazil Flux Eddy Covariance MeetingSanta Maria Rio Grande do Sul, Brazil

November, 2011

Micrometeorological Methods Used to Measure Greenhouse Gas Fluxes

Challenges in Measuring Greenhouse Gas Fluxes

• Measuring/Interpreting greenhouse gas flux in a quasi-continuous manner for Days, Years and Decades

• Measuring/Interpreting fluxes over Patchy, Microbially-mediated Sources (e.g. CH4, N2O)

• Measuring/Interpreting fluxes of Temporally Intermittent Sources (CH4, N2O, O3, C5H8)

• Measuring/Interpreting fluxes over Complex Terrain and or Calm Winds• Developing New Sensors for Routine Application of Eddy Covariance, or

Micrometeorological Theory, for trace gas Flux measurements and their isotopes (CH4, N2O,13CO2, C18O2)

• Measuring fluxes of greenhouse gases in Remote Areas without ac line power

ESPM 228 Adv Topics Micromet & Biomet

Eddy Covariance

• Direct Measure of the Trace Gas Flux Density between the atmosphere and biosphere, mole m-2 s-1

• Introduces No Sampling artifacts, like chambers• Quasi-continuous• Integrative of a Broad Area, 100s m2

• In situ

ESPM 228 Adv Topics Micromet & Biomet

~ ' 'a aF ws w s

c c c

a a a

m psm P

Eddy Covariance, Flux Density: mol m-2 s-1 or J m-2 s-1

Flux Methods Appropriate for Slower Sensors, e.g. FTIR

• Relaxed Eddy Accumulation

• Modified Gradient Approach

• Integrated Profile

• Disjunct Sampling

)(' dnupw cccwF '

Fxu dzc background

z

z10

( )

F F csc sz

z

~

0 1 2 3 4 5 6 7 8 9

x 104

-5

0

5

W (

m s

-1)

24 hours

4.3 4.35 4.4 4.45 4.5 4.55 4.6 4.65 4.7

x 104

-4

-2

0

2

4

W (

m s

-1)

hour

4.32 4.321 4.322 4.323 4.324 4.325 4.326 4.327 4.328 4.329 4.33

x 104

-2

-1

0

1

2

seconds

W (

m s

-1)

100 seconds

Vaira Ranch, d110, 2008

ESPM 228 Adv Topics Micromet & Biomet

Vaira Ranch, d110, 2008

0 1 2 3 4 5 6 7 8 9

x 104

14

16

18

20

CO

2 (mm

ol m

-3)

24 hours

4.3 4.35 4.4 4.45 4.5 4.55 4.6 4.65 4.7

x 104

15

15.5

16

CO

2 (m

mol

m-3

)

hour

4.32 4.321 4.322 4.323 4.324 4.325 4.326 4.327 4.328 4.329 4.33

x 104

15

15.5

16

seconds

CO

2 (m

mol

m-3

)

100 seconds

0 1 2 3 4 5 6 7 8 9

x 105

-4

-2

0

2

4

W m

/s

0 1 2 3 4 5 6 7 8 9

x 105

1.5

2

2.5

3

3.5

4

4.5

5

seconds

CH

4 ppm

D164, 2008

24 Hour Time Series of 10 Hz Data, Vertical Velocity (w) and Methane (CH4) Concentration

Sherman Island, CA: data of Detto and Baldocchi

ESPM 228 Adv Topics Micromet & Biomet

0

)('' dSww ww

0

)('' dScwF wc

Co-Spectrum

Power and Co- Spectrum defines the Range of Frequencies to be Sampled

Power Spectrum

Frequency

Spec

tral

Den

sity

Fourier Transform of Signal from Temporal to Frequency Space Illustrates the Role of

Filtering Functions and Spectra0 1 2 3 4 5 6 7 8 9

x 104

14

16

18

20

CO

2 (mm

ol m

-3)

24 hours

4.3 4.35 4.4 4.45 4.5 4.55 4.6 4.65 4.7

x 104

15

15.5

16

CO

2 (m

mol

m-3

)

hour

4.32 4.321 4.322 4.323 4.324 4.325 4.326 4.327 4.328 4.329 4.33

x 104

15

15.5

16

seconds

CO

2 (m

mol

m-3

)

100 seconds

ESPM 228 Adv Topics Micromet & Biomet

Signal Attenuation:The Role of Filtering Functions and Spectra

• High and Low-pass filtering via Mean Removal– Sampling Rate (1-10Hz) and Averaging Duration (30-60 min)

• Digital sampling and Aliasing• Sensor response time• Sensor Attenuation of signal

– Tubing length and Volumetric Flow Rate– Sensor Line or Volume averaging

• Sensor separation– Lag and Lead times between w and c

Detto et al 2012 AgForestMet

Comparing Co-spectra of open-path CO2 & H2O sensor and closed-path CH4 sensor

Co-Spectra are More Forgiving of Inadequate Sensor Performance than Power SpectraBecause there is little w-c correlation in the inertial subrange

Co-Spectra is a Function of Atmospheric Stability:Shifts to Shorter Wavelengths under Stable ConditionsShifts to Longer Wavelengths under Unstable Conditions

Detto, Baldocchi and Katul, Boundary Layer Meteorology 2010

ESPM 228 Adv Topics Micromet & Biomet

F ws w s w w wa a c c c ' ' ' '

Formal Definition of Eddy Covariance, V2

Most Sensors Measure Mole Density, Not Mixing Ratio

ESPM 228 Adv Topics Micromet & Biomet

F w mm

w mm T

w Tc ca

v

c

av

v a

a v

c ' ' ' ( ) ' ''

1

Webb, Pearman, Leuning Algorithm:‘Correction’ for Density Fluctuations when

using Open-Path Sensors

ESPM 228 Adv Topics Micromet & Biomet

dead annual grassland

Day

180 181 182 183 184 185

F (

mol

m-2

s-1

)

-20

-15

-10

-5

0

5

10

Fwpl

<w'c'>

Raw <w’c’> signal, without density ‘corrections’, will infer Carbon Uptake when the system is Dead and Respiring

See new Theory by Gu et al concerning da/dt

ESPM 228 Adv Topics Micromet & Biomet

Hanslwanter et al 2009 AgForMet

Annual Time Scale, Open vs Closed sensors

Ope

n Pa

th

Closed path

Towards Annual SumsAccounting for Systematic and Random Bias Errors

• Advection/Flux Divergence• U* correction, or some alternative, for lack of adequate turbulent

mixing at night• QA/QC for Improper Sensor Performance

– Calibration drift (slope and intercept), spikes/noise, a/d off-range– Signal Filtering

• Software Processing Errors• Lack of Fetch/Spatial Biases

– Sorting by Appropriate Flux Footprint

• Change in Storage• Gaps and Gap-Filling

ESPM 228 Adv Topic Micromet & Biomet

Oak Ridge, TNTemperate Deciduous Forest1997

Day-Hour

0 50 100 150 200 250 300 350

F wpl

(m

ol m

-2 s

-1)

-40

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

Vaira Grassland 2001

Day/Hour

0 50 100 150 200 250 300 350

Fc (

mol

m-2

s-1

)

-25

-20

-15

-10

-5

0

5

10

15

Tall Vegetation, Undulating Terrain

Short Vegetation, Flat Terrain

time 0 1 2 3 4 5 6 7

f(t)

-3 -2 -1 0 1 2 3

signal random error

time 0 1 2 3 4 5 6 7

f(t)

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0

signal bias error

time 0 1 2 3 4 5 6 7

f(t)

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

signal systematic bias

Systematic and Random Errors

ESPM 228 Adv Topic Micromet & Biomet

Random Errors Diminish as We Measure Fluxes Annually and Increase the Sample Size, n

Systematic Biases and Flux Resolution: A Perspective

• FCO2: +/- 0.3 mol m-2 s-1 => +/- 113 gC m-2 y-1

• 1 sheet of Computer paper 1 m by 1 m: ~70 gC m-2 y-1

• Net Global Land Source/Sink of 1PgC (1015g y-1): 6.7 gC m-2 y-1

The Real World is Not Kansas, which is Flatter than a Pancake

ESPM 228 Adv Topics Micromet & Biomet

Cloud

Cloud

Eddy Covariance in the Real World

ESPM 228 Adv Topics Micromet & Biomet

' ' ' ' ' ' ' 'j

j

u cdc c c c c u c v c w cu v wdt t x y z x x y z

I: Time Rate of ChangeII: AdvectionIII: Flux Divergence

I II III

Our Master ---To Design and Conduct Eddy Flux Measurements under Non-ideal Conditions

Conservation Equation for C for Turbulent Flow

Test Representativeness:

Sampling Error with Two Towers

Hollinger et al GCB, 2004

Estimating Flux Uncertainties:Two Towers over Rice

Detto, Anderson, Verfaillie, Baldocchi unpublished

Test for AdvectionDaytime and Nightime Footprints over an Ideal, Flat Paddock

Detto et al. 2010 Boundary Layer Meteorology

0Fz

FF dz Constz

Examine Flux Divergence

Detto, Baldocchi and Katul, Boundary Layer Meteorology 2010

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-2

0

2

4

6

8

10

12NEE at Night, Peatland Pasture

CO2 [

mol

m-2

s-1

]

u* m/s

Losses of CO2 Flux at Night: u* correction

U* Corrections are not the Best, But we have Few Better Alternatives

time (hours)

0 4 8 12 16 20 24

CO

2 Flu

x D

ensi

ty m

ol m

-2 s

-1

-25

-20

-15

-10

-5

0

5

10

Ne: measured (-4.84 gC m-2 day-1) Ne: computed (-5.09 gC m-2 day-1)

Fwpl+Storage: measured (-5.96 gC m-2 day-1) Fwpl: measured (-6.12 gC m-2 day-1)

Baldocchi et al., 2000 BLM

Underestimating C efflux at Night, Under Tall Forests, in Undulating Terrain

Van Gorsel et al 2007, Tellus

Consider Storage under Low, Winds

Systematic Biases are an Artifact of Low Nocturnal Wind Velocity

Friction Velocity, m/s

Moffat et al., 2007, AgForMet

Gap-Filling Inter-comparison Bias Errors

ESPM 228 Adv Topic Micromet & Biomet

ESPM 228 Adv Topics Micromet & Biomet

Annual Sums comparing Open and Closed Path Irgas

Hanslwanter et al 2009 AgForMet

Biometric and Eddy Covariance C Balances Converge after Multiple Years

Gough et al. 2008, AgForMet

Is There an Energy Balance Closure Problem?:

Evidence from FLUXNET

Wilson et al, 2002 AgForMet

Timing/SeasonInstrument/Canopy Roughness

-100 0 100 200 300 400 500 600-100

0

100

200

300

400

500

600

slope=1.05r2=0.98

wheat

H+L

E (W

m-2

)

Rn-G (W m-2)

-100 0 100 200 300 400 500 600 700 800-100

0

100

200

300

400

500

600

700

800

r2=0.93slope=0.93

Boreas 1994Hourly averages

Old Jack Pine

LE+H

+S+G

(W

m-2

)

Rn (W m-2)

Rnet (W m-2)

0 200 400 600 800

E +

H +

G +

S +

Ps

(W m-2)

0

200

400

600

800

Coefficients:b[0] 3.474b[1] 1.005r ² 0.923

Temperate Deciduous Forest

Contrary Evidence from Personal Experience:Crops, Grasslands and Forests

Rnet (W m-2)

-100 0 100 200 300 400

LE+H

+G

-100

0

100

200

300

400coefficients:b[0] 5.55b[1] 0.94r ² 0.926

Vaira Grassland, D296-366, 2000

ESPM 228 Adv Topic Micromet & Biomet

Many Studies Don’t Consider Heat Storage of Forests Well, or at All, and Close Energy Balance when they Do

Lindroth et al 2010 Biogeoscience Haverd et al 2007 AgForMet

Methane Measurements in California

Measuring Methane with Off-Axis Infrared Laser Spectrometer

Los Gatos Research

Closed pathModerate Cell Volume, 400 ccLong path length, kilometersHigh power Use:Sensor, 80 WPump, 1000 W; 30-50 lpmLow noise: 1 ppb at 1 HzStable Calibration

Closed Path Systems Need access to AC powerTo power 1000 W pumps and sensors

LI-7700 Methane Sensor, variant of frequency modulation spectroscopy

Open path, 0.5 mShort optical path length, 30 mLow Power Use: 8 W, no pumpModerate Noise: 5 ppb at 10 HzStable Calibration

Open Path Sensors Work off the Grid

ESPM 228 Adv Topics Micromet & Biomet

Zero-Flux Detection Limit, Detecting Signal from Noise

' ' wc w cF w c r rwc ~ 0.5ch4 ~ 0.84 ppbco2 ~ 0.11 ppm

Methane Lab Calibration

Time

260 280 300 320 340 360 380 400

Met

hane

Sen

sor

1880

1890

1900

1910

1920

Mean: 1897.4277StdDev: 0.8411Std Err 0.0219

U* w Fmin, CH4 Fmin, CO2m/s m/s nmol m-2 s-1 mol m-2 s-1

0.1 0.125 2.1 0.2750.2 0.25 4.2 0.550.3 0.375 6.3 0.8250.4 0.5 8.4 1.10.5 0.625 10.5 1.375

Detto et al 2012 AgForestMet

Flux Detection Limit, v2Based on 95% CI that the Correlation between

W and C that is non-zero

0.035 mmol m-2 s-1, 0.31 mol m-2 s-1 and 3.78 nmol m-2 s-1 for water vapour, carbon dioxide and methane flux,

Lags are Introduced when Sampling through a Tube

Closed Path Methane Sensors Attenuate Fluctuations

Detto et al. 2012 AgForMet

Spectral Losses Scale with Wind Speed

Open Path Sensors are Sensitive toWPL Density Correction

Detto et al. 2012 AgForMet

0 2 4 6 8 100

2

4

6

8

10

12

14

16

18

20

wq mmol m-2 s-1

met

hane

cor

rect

ion

term

for w

ater

vap

or, n

mol

m-2

s-1

Density ‘Corrections’ Are More Severe for CH4 and N2O:This Imposes a Need for Accurate and Concurrent Flux Measurements of H and LE

0 0.1 0.2 0.3 0.40

20

40

60

80

100

120

wt K m s-1

met

hane

cor

rect

ion

term

for <

wt>

, nm

ol m

-2 s

-1

Open-Closed Methane Flux Comparison

Detto et al. 2012 AgForMet

Time

0 4 8 12 16 20 24

CH

4 (pp

b)

1900

2000

2100

2200

2300

2400

2500

Days 100-300Days < 100 or > 300

Time

0 4 8 12 16 20 24

CH

4 Fl

ux (n

mol

m-2

s-1)

0

10

20

30

40

50

60

70

Days 100-300 Days < 100 or > 300

Sherman Island, 2007-2010, Westerly Winds

Baldocchi et al. 2012 AgForMet

Interpreting Methane Flux Measurements

8:15 9:15 10:4510:15

8:15

11:15 11:45 12:15 13:4513:15

14:15 14:45 15:15 15:45 16:45

8:45

12:45

16:15

9:45

Watch Out for Vacche

UpScaling Tower Based C Fluxes with Remote Sensing

LIDAR derived map of Tree location and Height

Hemispherical Camera Upward Looking Camera

Web Camera

ESPM 111 Ecosystem Ecology

Annual Grassland, 2004-2005

Wavelength (nm)

400 500 600 700 800 900 1000

Ref

lect

ance

0.0

0.2

0.4

0.6

0.8

1.0

Oct 13, 2004Oct 27, 2004Nov 11, 2004Jan 5, 2005Feb 2, 2005Apr 1, 2005Mar 9, 2005May 11, 2005Dec 29, 2005

Falk, Ma and Baldocchi, unpublished

Ground Based, Time Series of Hyper-Spectral Reflectance Measurements, in Conjunction with Flux Measurements Can be Used to Design Future Satellites

Remote Sensing of NPP:Up and down PAR, LED, Pyranometer, 4 band Net Radiometer

LED-based sensors are Cheap, Easy to Replicate and Can be Designed for a Number of Spectral Bands

Spectrally-Selective Vegetation Indices Track Seasonality of C Fluxes Well

Ryu et al. Agricultural and Forest Meteorology, 2010

Ryu et al. Agricultural and Forest Meteorology, 2010

Vegetation Indices can be Used to Predict GPP with Light Use Efficiency Models

Take-Home Message for Application of Eddy Covariance Method under Non-Ideal

Conditions

•Routine Flux Measurements Must Comply with Governing Principles of Conservation Equation•Design Experiment that measures Flux Divergence and Storage, in addition to Covariance•Networks need more Sites in Tropics and Distinguish C3/C4 crops•Networks need Sites that Cover a Range of Disturbance History•Network of Flux Towers, in conjunction with Remote Sensing, Climate Networks and Machine Learning Algorithms has Potential to Produce Carbon Flux Maps for Carbon Monitoring for Treaty, with Caveats and Accepted Errors

Moffat et al., 2007, AgForMet

ESPM 228 Adv Topic Micromet & Biomet

Root Mean Square Errors with Different Gap Filling Methods

Non-Dispersive Infrared Spectrometer, CO2 and H2O

LI 7500

Measures mole density, not mixing ratio

Open-path , 12.5 cmLow Power, 10 WLow noise, CO2: 0.16 ppm; H2O: 0.0047 ppthLow drift, stable calibrationLow temperature sensitivity: 0.02%/degree C