Multiscale Approaches to BiosphereAtmosphere Exchange Rates of H2O and CO2

1
Multi-scale Approaches to Biosphere-Atmosphere
Exchange Rates of H2O and CO2 

• By
• Gabriel Katul
• Nicholas School of the Environment and Earth
  Sciences,
• Duke University

2
National Research Priorities

• The role of the biosphere as a terrestrial C
  sink.
• How can we determine whether ecosystems are C
  sources or sinks from climate drivers and
  ecophysiological properties?

• Couple C to H2O (other biogeochemical cycles).
• To what extent does knowledge of water
  availability improve our ability to predict
  terrestrial CO2 sources and sinks?

3
Spatial Scales in Water Transport
4
Temporal Scales in Water Transport
Diurnal
Turbulent
Seasonal
Water vapor flux variability exists on time
scales ranging from fractions of seconds to
multiple decades
5
Existing models on a complexity scale
6
The Mechanisms in the Models
7
Nested Scheme

• Revisions to 3PG
• Foliage Display Allometric relation for
  Stem/Foliage (amount), GDD (timing)
• Litterfall Fraction of biomass at a reference
  month, two cohorts
• Reserve Pool foliage and respiration costs

8
Multilayer Model Calculations
Meteorological Forcing (30 min)
Ta, RH, Ca
a(z)
PAR
ltUgt
S
dz
Soil
9
Time-averaged Equations
Time averaging 30 minutes
10
Linking concentration to source
strength Lagrangian Fluid Mechanics
Determined from Thomsons (1987) model
11

Channel Bottom
12
Scalar Transport Thomson (1987)
Model
Measured
13
CO2
Fluxes at z/h1
Modeled Sc
Model ecophysiological parameters are
independently measured using porometry (leaf
scale).
H2O
T
Fluxes shown are measured at the canopy scales
14
Comparison between measured and modeled mean CO2
Concentration
CO2 measured by a 10 level profiling system
sampled every 30 minutes.
Sources and sinks and transport mechanics are
solved iteratively to compute mean scalar
concentration
15
Testing the Model Nesting
2000-2004
2002
16
Results Leaf area dynamics

• N-CANVEG slow processes

Measured
Modeled
17
Results LAI
18
Criteria in Model Evaluation

• Emphasis on variability not only long-term mean
  values Tool - Spectral Comparisons for NEE and
  LE
• Measured spectrum from EC data

19
Results Spectral Comparisons

• All models NEE spectrum

20
Disagreements Short Time Scales

• Dynamic Footprint of measurements
• Nocturnal canopy turbulence
• Parameterization of canopy turbulence
• In the deeper layers of the canopy

21
Footprint shifts
22
Gravity Waves Stable Boundary Layer at z/h
1.12 (Duke Forest)
23
Night-Time Nonstationarity
24
(No Transcript)
25
Richardson (1925) Cascade 3-D turbulenceBig
whirls have little whirls,which feed on their
velocity.Little whirls have lesser whirls,and
so on to viscosity
Kolmogorov (K41)
26
A poem describing 2D turbulence When little
whirls meet little whirls,they show a strong
affectionelope, or form a bigger whirl,and so
on by advection.
by G.J.F. van Heijst
Batchelor-Kraichnan
27
Formation of organized and stable vortices
Initial Conditions
Simulations from Farge et al. (1994)
28
(No Transcript)
29
NEE Spectral Comparisons
30
(No Transcript)
31
Electric equivalence models - ODE
Porous media models PDE
Soil-tree water flux
3. Simple beam model
1. 2R model
2. RC model
4. Branching model linked to dynamic turbulent
closure
32

• 1-D Richards equation with H2O sinks
• Solved by Finite Element Methods
• ? Stem hydraulics based on cavitation
• curves from Sperry et al. (2000)
• Maximum H2O loss from CANVEG
• Stomatal shut-down
• From Sperry - Critical pressure

33
(No Transcript)
34
H2O Pressure Simulations
Mean conductive Xylem
Poorly conductive Xylem
35
Conclusions

• The newly proposed nested carbon allocation
  scheme captured most of the intra-annual
  variations of NEE and LE in this non-equilibrium
  stand.
• The model comparisons suggest strong spectral
  linkage between carbon production and LAI,
  especially at seasonal time scales.
• None of the models tested here captured well the
  drought effects on water and CO2 fluxes.

36

• Plant hydraulics a starting point porous media
  models. Time scales of storm/water redistribution
  may be needed non-trivial computational
  expense.
• Effects of Nitrogen amendments elevated CO2 on
  wood composition and plant hydraulics has been
  documented in FACE.

37

• Below ground
• Respiration
• Soil moisture
• Another mess

38
Building on the French Success, Nicaragua
released a stamp with K.F. Gauss.
39
Thanks to
Davide Poggi (Torino) Daniela Cava
(CNR-Lecce) Brani Vidakovic (GaTech) David
Ellsworth (U. Michigan) John Sperry (Utah) Ram
Oren (Duke) Amilcare Porporato (Duke) Heather
McCarthy (Duke) Gil Bohrer (Duke) Mario Siqueira
(Duke) Paul Stoy (Duke) Edo Daly (Duke) Jehn-Yih
Juang (Duke)
40
Results Spectral Comparisons

• All models LE spectrum