Carbon, soil moisture and fAPAR assimilation

1
Carbon, soil moisture and fAPAR assimilation
Wolfgang Knorr Max-Planck Institute of
Biogeochemistry Jena, Germany 1 Acknowledgments
Nadine Gobron 2, Marko Scholze 3, Peter Rayner 4,
Thomas Kaminski 5, Ralf Giering 5, Heinrich
Widmann1
2
3
QUEST /
LSCE
IES/JRC
FastOpt
4
5
2
Overview

• CO2 climate linkages
• Satellite fAPAR as soil moisture indicator
• Assimilation of fAPAR

3
Atmospheric CO2 Measurements
4
Carbon Cycle Data Assimilation System (CCDAS)
Assimilated
Prescribed
Assimilated
CO2 Uncert.
satellite fAPAR Uncert.
Phenology Hydrology
CCDAS Step 2 BETHYTM2 only Photosynthesis,
EnergyCarbon Balance
CCDAS Step 1 full BETHY
Background CO2 fluxes
Optimized Params Uncert.
Diagnostics Uncert.

ocean Takahashi et al. (1999), LeQuere et al.
(2000) emissions Marland et al. (2001), Andres
et al. (1996) land use Houghton et al. (1990)
5
Terr. biosphereatmosphere CO2 fluxes
6
ENSO and global climate normalized anomalies
ENSO
temperature
precipitation
7
(No Transcript)
8
(No Transcript)
9
for more information see http//www.CCDAS.org
10
Overview

• CO2 climate linkages
• Satellite fAPAR as soil moisture indicator
• Assimilation of fAPAR

11
Remotely Sensed Vegetation Activity
12
SeaWiFS fAPAR archive

• developed by Nadine Gobron, Bernard Pinty,
  Frédéric Melin, IES/JRC, Ispra
• 3-channel algorithm taylored to SeaWiFS ocean
  color instrument (blue, red, near-infrared)
• cloud screening algorithm
• requires no atmospheric correction
• starts 10/1997, continuing...
• being extended by same product for MERIS

13
Precipitation  fAPAR
precipitation
gridded station data
soil moisture
BETHY simulations
leaf area index
fAPAR
satellite observations BETHY simulations
14
precipitation vs. fAPAR from SeaWiFS satellite
obs.
percent area with 99 significant correlation
0.5x0.5, 50 cloud free, 75 temporal coverage
15
precipitation vs. fAPAR satellite and model
SeaWiFS fAPAR
percent area with 99 significant correlation
16
precipitation vs. fAPAR satellite and model
4-month lag
SeaWiFS fAPAR
percent area with 99 significant correlation
BETHY simulated fAPAR
17
precipitation vs. satellite fAPAR and simulated
soil moisture
SeaWiFS fAPAR
percent area with 99 significant correlation
18
precipitation vs. satellite fAPAR and simulated
soil moisture
4-month lag
SeaWiFS fAPAR
percent area with 99 significant correlation
19
ENSO SeaWiFS fAPARlagged correlation
3-month lag
20
Overview

• CO2 climate linkages
• Satellite fAPAR as soil moisture indicator
• Assimilation of fAPAR

21
fAPAR Assimilation
PFT distribution
ecosystem model parameters
Prescribed
climate soils data
BETHY
model-derived fAPAR
carbon and water fluxes
22
The Cost Function
Measure of the mismatch (cost function)
23
The Parameters
vector of prior parameter values m0
represents
parameter vector mm1,m2,m3
m1
?Tf
shift of leaf onset/shedding temperature
temperature limitation
DTf??0
m2
wmax,0 (derived from FAO soil map)
wmax
maximum soil water holding capacity
water limitation
m3
fc
fraction of grid cell covered with vegetation
residual, unmodelled limitations (nitrogen, land
use)
fc,0 (function of P/PET and Temp. of warmest
month)
24
Prior Parameter 1
prior values
note each 0.5x0.5 has mixture of up to 6 PFTs
map reflects presence of crops red unvegetated
25
Prior Parameter 2
26
Prior Parameter 3
27
Prior Parameter Errors
error covariance matrix of parameters Cm0

• off-diagonal elements assumed 0 here
• no prior correlation between errors of
  different parameters

28
The Assimilated Data
model diagnostics vector yy1,y2,...,y12
yi
modelled fAPAR of month i
satellite-derived diagnostics vector
y0y0,1,y0,2,...,y0,12
y0,i
SeaWiFS derived fAPAR of month i
29
Prior Errors of Measurements
error covariance matrix of measurements Cy
ij

• off-diagonal elements again 0
• no prior correlation between errors of
  different months

30
Parameter 2 (regional)
soil water-holding capacity
prior
optimized
31
Local Simulations
Paragominas 3S 48W 63 m
precipitation mm/month
fAPAR
1992
no remote sens. data
fAPAR prescribed
evapotranspiration mm/month
NPP gC/(m2 month)
fAPAR assimilated
-
-
32
Measured Soil Moisture
Paragominas 3S 48W 63 m
precipitation mm/month
fAPAR
1992
0...8m depth
0...2m depth
-
-
1992
1992
33
evapotranspiration (regional)
prior
optimized
mm/year
mm/year
34
July soil moisture (regional, dry season)
prior
optimized
mm
mm
35
Conclusions

• The carbon cycle is highly sensitive to climate
  fluctuations
• Vegetation can be quantified reliably from space
• fAPAR lags precipitation by 14(?) months
• seems to behave similar to soil moisture
• assimilation of fAPAR can deliver valuable
  information on soil moisture status

36
Conclusions

• Need to improve phenology model
• Implement sequential 2-D var assimilation scheme
• Assimilate fAPAR into coupled ECHAM5-BETHY model

(hope not too distant) goal make fAPAR what SST
is for ocean-atmosphere interactions... and
improve seasonal forecasts
37
Thank You For Your Attention!