Water balance

1
Water balance
?S P D - ET
drainage
Change in water content of volume of soil
precipitation
By Dr Marcy Litvak Dept of Biological
Sciences University of Texas at Austin
2
Energy budgeting approach
Latent Heat flux
How do you partition H and ?E??
Can directly measure each of these variables
Sensible Heat flux
3
Net Ecosystem Production
Eddy Covariance Directly measure how much CO2
or H2O vapor blows in or out of a site in wind
gusts.
Integrated measure of ecosystem fluxes
Link changes in CO2 or H2O in the air above a
canopy with the upward or downward movement of
that air
4
Net Ecosystem Exchange
30 minute timescale
Updraft CO2 gt downdraft CO2
Flux gt0 carbon source
Updraft CO2 lt downdraft CO2
Flux lt 0 carbon sink
5
1000
Sunlight
800
600
Sunlight (Wm-2)

• The net CO2 flux is calculated for each half hour
  from the measurements of vertical wind and CO2
  concentration.
• A positive flux indicates a net loss of CO2 from
  the surface (respiration) and a negative flux
  indicates the net uptake of CO2 (photosynthesis)

400
200
0
146.0
146.5
147.0
147.5
148.0
5
0
-5
CO2 Exchange (mmol m-2 s-1)
-10
-15
CO2 Exchange
-20
12 AM
12PM
12AM
12PM
12AM
May 26, 2000
May 27, 2000
6
CO2 Exchange (mmol m-2 s-1)

• A years worth of half-hour data can be summed to
  determine how much Carbon the ecosystem gained or
  lost

5
4
Annual C accumulation (Tons C ha-1)
3
2
1
0
1999 2000
7
ET -Eddy covariance method

• Measurement of vertical transfer of water vapor
  driven by convective motion
• Directly measure flux by sensing properties of
  eddies as they pass through a measurement level
  on an instantaneous basis
• Statistical tool

8
Basic Theory
Instantaneous Perturbation from The mean
Instantaneous signal
Time averaged property
All atmospheric entities show short-period
fluctuations about their long term mean value
9
Turbulent mixing
Propterties carried by eddies Mass, density
? Vertical velocity w Volumetric content ?

1) Expand 2) Simplify a) remove all terms
with single primed entity b) remove terms
with fluctuations in c) remove terms
containing mean vertical velocity
10
Eddy Covariance
11
Eddy covariance
Average vertical flux of entity over 30 minute
period
Fluctuation of entity about its mean g kg air-1
Density of air kg air m-3
?
F
w
x
Velocity of air being moved upwards or
downwards m s-1
At any given instant, multiply velocity of
air being moved upwards or downwards at a speed
of m s-1, by the fluctuation of the entitiy about
its mean
12
Eddy covariance
m g s kg
kg m3
g m-2 s-1
Resultvertical speed of transfer of entity
measured in m s-1 and at a concentration of g
per kg of air
g of entity transferred vertically, per square
meter of surface area per second
13
Latent heat of vaporization (J kg-1 C-1)
Mean density of air
QE
?
Lv
Fluctuation about the mean of vertical wind
speed
Fluctuation about the mean of density of water
vapor in air
m kg s m2
kg m3
J m2s
W m2


14
Specific heat of air at constant pressure (J kg-1
C-1)
Mean density of air
QH
?
Cp
Fluctuation about the mean of vertical wind
speed
Fluctuation about the mean of air temperature
m ?C s
kg m3
J m2s
W m2


15
Instrumentation Requirements
16
3-D Sonic anemometer
Quantum sensor
Pyrronometer
IRGA
Net radiometer
17
Instrumentation Requirements
18
Challenges of operating eddy flux systems in
remote locations!
19
Advantages of eddy covariance

• Inherently averages small-scale variability of
  fluxes over a surface area that increaes with
  measurement height
• Measurements are continuous and in high temporal
  resolution
• Fluxes are determined without disturbing the
  surface being monitored
• Great tool to look at ecosystem physiology

20
Disadvantages

• Need turbulence!
• Gap filling issues
• Relatively Expensive
• Stationarity issues
• Open-path IRGA issues
• The eddy covariance method is most accurate when
  the atmospheric conditions (wind, temperature,
  humidity, CO2) are steady, the underlying
  vegetation is homogeneous and it is situated on
  flat terrain for an extended distance upwind.

21
Stationiarity
Advection
Horizontal concentration gradients may also lead
to perturbation calculation errors
22
(No Transcript)
23
Issue of energy balance closure
24
Impact of encroachment of Ashe juniper and Honey
mesquite on carbon and water cycling in central
Texas savannas
Marcy Litvak Section of Integrative
Biology University of Texas, Austin
Collaboration with James Heilman, Kevin McInnes,
James Kjelgaard, Texas AM Melba Crawford,
Roberto Gutierrez, Amy Neuenschwander, UT Freeman
Ranch - Texas State University
25
Figure 1. Location and geographical extent of
Edwards Plateau
26
Extensive areas of Edwards Plateau historically
were dominated by fairly open live-oak
savannas
27
Due to overgrazing and fire suppression
policies.grasslands are disappearing as woody
species increase
28
Research Objectives

• Determine sink strength for carbon associated
  with woody encroachment and analyze the variables
  that determine gains/losses of carbon from key
  central Texas ecosystems
• Determine change in ET, energy balance and
  potential groundwater recharge associated with
  woody encroachment
• Provide objective data for validation of land
  surface process models (CLM2 Liang Yang, UT)
  related to growth, primary production, water
  cycling, hydrology
• Aid in regional scale modeling efforts

Carbon/water tradeoff
29
Study site
30
Experimental design

• 3 stages of woody
  encroachment
• Open grassland, transition site, closed canopy
  woodland
• -NEE carbon, water, energy open-path eddy
  covariance
• (net radiation, solar radiation (incoming,
  upwelling), PAR, air temperature, relative
  humidity, precipitation)
• -physiological measures of ecosystem component
  fluxes
• leaf-level gas exchange, sap-flow,
  bole-respiration rates, herbaceous NEE
• -soil carbon, soil microclimate, soil respiration
  rates
• Ecosystem structure
• biomass, LAI, species composition

31
open grassland May 2004
(TAMU)
Transition site July 2004 15-20 year old
juniper,mesquite
Live Oak-Ashe juniper woodland July 2004
(TAMU)
32
(No Transcript)
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
(No Transcript)
37
(No Transcript)
38
How carbon cycling and energy balance in boreal
forest stands changes through succession
following wildfire
Collaboration with Michael Goulden, Greg Winston,
Andrew McMilaan UC-Irvine Sue Trumbore, Claudia
Czimczik, UC-Irvine Jennifer Harden, Kristen
Mainies, USGS-Menlo Park Hugo Veldhuis, Pascal
Cyr Environment Canada Tom Gower, Ben
Bond-Lamberty UW-Madison Modeling
work Guo-Yue Niu, Liang Yang Jackson School
Geosciences UT-Austin
39
Boreal forest
Savanna
13.8 million km2 between 46ºN and 66 ºN 8-10
Earths terrestrial surface 88 Gt C in biomass
(19 global total) 471 Gt C in soil organic
matter (23 global total)
40
Fire plays an integral role in boreal forest
ecosystems
July 23, 1989 (Manitoba)
Lightning-induced fire in a stand of black
spruce between BOREAS NSA and Churchill in 1994
Dominant disturbance regime Fire cycle
typically 50-200 years
41
BOREAS-Northern Study Area Fire history- TM July
25, 1990
Fire maintains boreal landscape as spatial mosaic
of forest patches in different successional
stages
Motivation for our research
42
Succession following wildfire in black spruce
mature forest
2-3 weeks following burn
43
5 years post-burn
14 years post-burn
44
23 years post-burn
40 years post-burn
45
150 years post-burn
70 years post-burn
46
Study site Thompson, Manitoba Hub of the North
Chronosequence substitute space for time
1850
1930
1964
1998
1989
1981
47
Net ecosystem exchange measured in 6 sites for
almost 3 years
2002
2003
2004
48
Weather patterns above all 5 stands
49
Results
Ecosystem structure
Tree biomass
Ground cover
Biomass g C m-2
50
Ecosystem structure
Total foliage, seedlings and trees
51
1989
1963
1989
1850
1989
1963
1989
1850
52
1989
1963
1989
1850
1989
1963
1989
1850
53
1989
1963
1989
1850
1989
1963
1989
1850
54
H/ (?E)
Bowen Ratio
Energy balance approach to estimating convective
fluxes Seeks to partition energy available into
sensible and latent heat terms
Typical values 0.1-
0.3 tropical rainforests soil wet
year-round 0.4 0.8 temperate forests and
grasslands 2-6 semi-arid regions extremely dry
soils gt 10 deserts
55
Bowen Ratio
Bowen (1926)
B can be approximated as a function of vertical
differences of temperature and vapor pressure in
the air, or , B g (t2- t1 ) / ( e2 e1 )

vapor pressures measured at the same two points
air temperatures measured at two points at
different heights above the land surface
Psychrometer Constant F(T,P)
56
Bowen Ratio
Bowen Ratio
Average values of the air-temperature differences
(t2 - t1) and vapor-pressure differences (e2 -
e1), taken every 30 seconds for a 30-minute
period are used to determine ? .
Specific heat capacity

• QH
• QE

?T
Ca

Lv
??v
Latent heat Of vaporization
57
Bowen Ratio
The energy budget can then be solved for LE LE
( Rn G W) / ( 1? )

• Uses gradients of heat and water to partition
• available energy into SH and LE
• Assumptions
• One-dimensional heat and vapor flow, only
  vertical
• No transfer to/from measurement area from
  adjacent area
• No significant heat storage in plant canopy
• 2 fluxes originate from same point on land
  surface
• Atmosphere equally able to transfer heat and
  water vapor,
• so turbulence need not be considered

58
Needs large tract of uniform vegetation
Sensors to measure air temperature and humidity
Determine average differentials for 15-minutes,
then switch sensors, and determine average
differentials for another 15 minutes to avoid
sensor bias