Chuixiang (Tree) Yi School of Earth and Environmental Sciences Queens College, City University of New York.

Chuixiang (Tree) YiChuixiang (Tree) Yi

School of Earth and Environmental SciencesSchool of Earth and Environmental SciencesQueens College, City University of New YorkQueens College, City University of New York

Acknowledgements to

Russ Monson, University of ColoradoDean Anderson, USGS

Brian Lamb, Washington State UniversityJohn Zhai, University of Colorado

Andrew Turnipseed, NCARSean Burn, University of Colorado

3

5

0 0 0

r r r

r

Z Z Z

Z

S T HA VA

c c cNEE dz w c u dz w dz

t x zF F F F

Eddy Flux TowerEddy Flux Tower

COCO22 HH22OO

CHCH44 T T

COCO22 HH22OO

CHCH44 T TCOCO22 HH22OO

CHCH44 T T

Super-stable layer, flow separation (Yi et al., 2005)

Courtesy of Jielun Sun Canopy flows are separated by a

superstable layer in calm nights.

Davis et al. 2004

West West towertower

South South towertower

North North towertower

East East towertower

Vertical advectionVertical advection

0 0

r s r

s

Z Z Z

vadv Z

c c cF w dz w dz w dz

z z z

00

hWind SpeedWind Speed Air TemperatureAir Temperature

Super-stable layerSuper-stable layer

ZZss

ZZrr

w++

--

Nighttime dc/dz is always negativeNighttime dc/dz is always negative

Vertical exchange zoneVertical exchange zone

Horizontal exchange zoneHorizontal exchange zone

Yi et al., 2008Yi et al., 2008

sZ

rr s

r r s s

ZZ Zvadv Z Z Z Z

r s

w wF w c w c cdz

Z Z

Vertical advectionVertical advection

s sZ Z =

r r r

s

Z Z Z

vadv Z

c wc wF w dz dz c dz

z z z

=

=

= ( )

rr

r rs

r r r

r r

ZZvadv Z Z Z

r s

Z Z Z

Z Z

wF w c cdz

Z Z

w c w c

w c c

Lee’s assumptionLee’s assumption

r sZ Z

r s

w ww

z Z Z

0 at super-stable layersZ

w ,r rZ Zw c

0sZw

ZZss

ZZrr

Super-stable Super-stable layerlayer

1 r

s

Z

Zr s

c cdzZ Z

Yi et al., 2008Yi et al., 2008

sZ

rr s

r r s s

ZZ Zvadv Z Z Z Z

r s

w wF w c w c cdz

Z Z

Vertical advectionVertical advection

s sZ Z =

r r r

s

Z Z Z

vadv Z

c wc wF w dz dz c dz

z z z

=

=

= ( )

rr

r rs

r r r

r r

ZZvadv Z Z Z

r s

Z Z Z

Z Z

wF w c cdz

Z Z

w c w c

w c c

Lee’s assumptionLee’s assumption

r sZ Z

r s

w ww

z Z Z

0 at super-stable layersZ

w ,r rZ Zw c

0sZw

ZZss

ZZrr

Super-stable Super-stable layerlayer

1 r

s

Z

Zr s

c cdzZ Z

Yi et al., 2008Yi et al., 2008

Longitude (degree)

-105.5485 -105.5475 -105.5465

Latit

ude

(deg

ree)

40.0315

40.0320

40.0325

40.0330

40.0335

WT

NT

ET

ST

136

10

21.5

136

10

31

1

6

1

6

wind direction

NT

ET

ST

Longitude (degree)

-105.5485 -105.5475 -105.5465

Latit

ude

(deg

ree)

40.0315

40.0320

40.0325

40.0330

40.0335

WT

NT

ET

ST

136

10

21.5

136

10

21.5

136

10

31

1

6

1

6

wind directionwind direction

NT

ET

ST

Horizontal advection calculationHorizontal advection calculation

0

rZ

hadv

cF u dz

x

0cos

rZ

iT

cu dz

r

WT ET

height

c

r

Horizontal CO2 gradient

Horizontal CO2 gradient

Ho

rizo

nta

l CO

2 g

rad

ien

t

Ho

rizo

nta

l CO

2 g

rad

ien

tWT NT

height

c

r

u

WT ST

height

c

r

Horizontal CO2 gradient

Horizontal CO2 gradientYi et al., 2008Yi et al., 2008

Horizontal advection calculationHorizontal advection calculation

Longitude (degree)

-105.5485 -105.5475 -105.5465

La

titu

de

(d

eg

ree

)

40.0315

40.0320

40.0325

40.0330

40.0335

WT

NT

ET

ST

136

10

21.5

136

10

31

1

6

1

6

wind direction

NT

ET

ST

Longitude (degree)

-105.5485 -105.5475 -105.5465

La

titu

de

(d

eg

ree

)

40.0315

40.0320

40.0325

40.0330

40.0335

WT

NT

ET

ST

136

10

21.5

136

10

21.5

136

10

31

1

6

1

6

wind directionwind direction

NT

ET

ST

WT ET

height

c

r

Horizontal CO2 gradient

Horizontal CO2 gradient

WT NT

height

c

r

u

WT ST

height

c

r

12

iTiT0

1 31 3

6 106 10

cos

cos *2 cos *2

cos *4 cos *4,

hadv

WT iT WT ET

m iT m ETm m

WT iT WT ET

m iT m ETm m

cF u dz

r

c cu u

r r

c cu u

r r

The CO2 gradient The CO2 gradient that is closer to that is closer to wind direction is wind direction is first choice to first choice to use. For example, use. For example, the angle can be the angle can be limited bylimited by

cos 0.8

Yi et al., 2008Yi et al., 2008

Another horizontal advection calculationAnother horizontal advection calculation

( ) ( )c c c c

c Ui Vj i j U Vx y x y

U

cos( )iTiT

cc u

r

U

Projection of CO2 gradient into wind directionProjection of CO2 gradient into wind direction

Projection of CO2 gradient into x and y directionsProjection of CO2 gradient into x and y directions

UU

VV

COCO 22 g

radient

gra

dientWind direction

Wind direction

UU

VV

/c x

/c y

/c r

dc/dr dc/dr is CO2 gradient along a pair towersis CO2 gradient along a pair towers

x-y coordinate system x-y coordinate system is fixed on an instrumentis fixed on an instrument

Yi et al., 2008Yi et al., 2008

udc/drudc/drcos(cos())

Udc

/dx+

Vdc

/dy

Udc

/dx+

Vdc

/dy

Comparison between two Comparison between two methods methods

Good agreement betweenGood agreement betweenTwo algorithmsTwo algorithms

Yi et al., 2008Yi et al., 2008

Yi et al., 2008Yi et al., 2008

Yi et al., 2008Yi et al., 2008

NEE+Fhadv

NEE+Ftadv

NEE+u*filter

NEE

NEE+Fvadv

0 0

rZ

cz

NEE s dz w c

Yi et al., 2008Yi et al., 2008

Vertical advection

Horizontal advection

u* correction

Soil temperature (Soil temperature (ooC)C)

Ad

vect

ion

flu

xA

dve

ctio

n fl

ux

Yi et al., 2008Yi et al., 2008

Advection Advection correctioncorrection

BiologicalBiologicalcorrectioncorrection

TurbulenceTurbulence

*u

BacterialBacterial

soilT

SummarySummary

Advection fluxes have no correlation to Advection fluxes have no correlation to drivers of biological activity. Conversely, drivers of biological activity. Conversely, the biological correction (u* filter) is the biological correction (u* filter) is based on climatic responses of based on climatic responses of organisms, and has no physical organisms, and has no physical connection to aerodynamic processes.connection to aerodynamic processes.

ConclusionConclusion

We must accept the fact that the We must accept the fact that the biological correction has no physical biological correction has no physical

basesbases

24

Super-stable layer, flow separation (Yi et al., 2005)

Courtesy of Jielun Sun Canopy flows are separated by a

superstable layer in calm nights.

hWind SpeedWind Speed Air TemperatureAir Temperature

2

g TT zRiuz

(1) Slow mean airflow;(1) Slow mean airflow; (2) Maximum drag elements;(2) Maximum drag elements;

(3) Minimum vertical exchange;(3) Minimum vertical exchange;

(4) Maximum horizontal CO2 (or other scalar) gradient;(4) Maximum horizontal CO2 (or other scalar) gradient;

(5) Maximum ratio of wake and shear production rate.(5) Maximum ratio of wake and shear production rate.

Vertical exchange zoneVertical exchange zone

Horizontal exchange zoneHorizontal exchange zone

X Distance (m)

Yi et al. 2005Yi et al. 2005

Yi et al. 2008Yi et al. 2008

Horizontal CO2 gradients are maximum at the super stable layer

VPD (kPa)0 1 2 3 4

-32

-30

-28

-26

-24

-22

Niwot RidgeBowling et al.

R (

‰)

0.05 0.1 0.15 0.2-27

-26

-25

-24

-23

(m3 m-3)

R (

‰)

Whole Whole canopy layercanopy layer

(Schaeffer et al., 2008)

0 1 2 3 4-32

-30

-28

-26

-24

-22

VPD (kPa)

NWR Canopy InletsBowling et al. R

-ca

no

py (

‰)

0.05 0.1 0.15 0.2-32

-30

-28

-26

-24

-22

NWR Ground Inlets R

-gro

un

d (

‰)

θ (m3 m-3)

Super-stable Super-stable layerlayer

After dividedAfter divided by by Super-stable layerSuper-stable layer

Above SSLAbove SSL

Below SSLBelow SSL

RemindingReminding