Supplement of Atmos. Chem. Phys., 17, 3553–3572, 2017 http://www.atmos-chem-phys.net/17/3553/2017/ doi:10.5194/acp-17-3553-2017-supplement © Author(s) 2017. CC Attribution 3.0 License.
Supplement of Methane fluxes in the high northern latitudes for 2005–2013 estimated using a Bayesian atmospheric inversion Rona L. Thompson et al. Correspondence to: Rona L. Thompson (
[email protected])
The copyright of individual parts of the supplement might differ from the CC-BY 3.0 licence.
Table 1. Inversion cost a priori and a posteriori for the three different scenarios shown for 2009 (other years were analogous). Scenario S1 S2 S3
Cost a priori 57737 57278 51472
Cost a posteriori 17531 17378 13352
Table 2. Comparison of the prior and posterior mixing ratios (scenario S1) with observations shown for 2009 (other years were analogous). Note the mean error is calculated as the mean of the prior (or posterior) minus the observed mixing ratio. Site ID ZEP PAL IGR YAK DEM KRS AZV VGN ZOT CHL LLB ETL FSD CHM ESP CBA MHD BAL
Prior R 0.36 0.65 0.49 0.35 0.68 0.63 0.38 0.49 0.55 0.51 0.66 0.49 0.35 0.10 0.08 0.18 0.45 0.44
NSD 0.95 1.25 0.34 0.68 0.78 0.64 1.39 0.89 0.75 0.70 0.21 0.29 0.71 0.64 0.44 0.42 1.25 0.92
mean error -4.66 2.29 -81.55 -6.96 -33.02 -38.11 -20.15 -13.10 -12.26 -4.97 -68.10 -24.41 -13.95 -10.89 -5.40 -6.66 1.81 3.89
Posterior R 0.40 0.68 0.67 0.68 0.75 0.82 0.42 0.74 0.72 0.74 0.80 0.74 0.57 0.35 0.20 0.54 0.45 0.62
NSD 0.89 1.05 0.83 0.83 1.13 1.00 1.39 0.98 1.14 1.04 0.67 0.89 0.80 0.67 1.10 0.58 0.93 0.88
mean error -3.71 0.51 -27.93 -0.83 -5.89 -7.11 -14.21 -1.94 0.54 -5.90 -21.48 -0.64 -8.97 -7.55 -1.20 -2.09 -0.37 0.20
Figure 1a. Comparison of atmospheric CH4 mixing ratios in winter 2008 – 2009 (left) and in summer 2009 (right). The error bars show the calculated observation uncertainty used in the inversions (observations = black, background = green, S1 prior = blue, S1 posterior = red). The full site names and details are provided in Table 1.
2200
2300
CHL
CH4 (ppb)
CH4 (ppb)
2300 2100 2000 1900 1800 2200 2000 1900 1800
2000 1900 1800
2000 1900 1800
2000 1900 1800
2100 2000 1900
2200
CBA
2100 2000 1900
2200
LLB
2100 2000 1900 2300
2100 2000 1900 1800 01/12
BAL
1800
VGN
CH4 (ppb)
CH4 (ppb)
2200
2200
2300
LLB
2100
2300
1900
1800
CH4 (ppb)
CH4 (ppb)
2200
2000
2300
CBA
2100
2300
KRS
2100
1800
CH4 (ppb)
CH4 (ppb)
2200
2200
2300
2100
2300
1900
1800
BAL
CH4 (ppb)
CH4 (ppb)
2200
2000
2300
KRS
2100
2300
CHL
2100
1800
CH4 (ppb)
CH4 (ppb)
2300
2200
2200
VGN
2100 2000 1900 1800
11/12
21/12
31/12
10/01
20/01
30/01
09/02
19/02
01/06
11/06
21/06
01/07
11/07
21/07
31/07
10/08
20/08
30/08
Figure 1b. Same as Fig. 1a but for a different set of sites.
2200
2300
ETL
CH4 (ppb)
CH4 (ppb)
2300 2100 2000 1900 1800 2200 2000 1900 1800
2000 1900 1800
2000 1900 1800
2000 1900 1800
2100 2000 1900
2200
CHM
2100 2000 1900
2200
ESP
2100 2000 1900 2300
2100 2000 1900 1800 01/12
FSD
1800
ZOT
CH4 (ppb)
CH4 (ppb)
2200
2200
2300
ESP
2100
2300
1900
1800
CH4 (ppb)
CH4 (ppb)
2200
2000
2300
CHM
2100
2300
MHD
2100
1800
CH4 (ppb)
CH4 (ppb)
2200
2200
2300
2100
2300
1900
1800
FSD
CH4 (ppb)
CH4 (ppb)
2200
2000
2300
MHD
2100
2300
ETL
2100
1800
CH4 (ppb)
CH4 (ppb)
2300
2200
2200
ZOT
2100 2000 1900 1800
11/12
21/12
31/12
10/01
20/01
30/01
09/02
19/02
01/06
11/06
21/06
01/07
11/07
21/07
31/07
10/08
20/08
30/08
Figure S1c. Same as Fig. 1a but for a different set of sites. 2200
2300
ZEP
CH4 (ppb)
CH4 (ppb)
2300 2100 2000 1900 1800 2200 2000 1900 1800
2000 1900 1800
2000 1900 1800
2000 1900 1800
PAL
2100 2000 1900
2200
IGR
2100 2000 1900
2200
YAK
2100 2000 1900 2300
2100 2000 1900 1800 01/12
2200
1800
DEM
CH4 (ppb)
CH4 (ppb)
2200
1900
2300
YAK
2100
2300
2000
1800
CH4 (ppb)
CH4 (ppb)
2200
TER
2100
2300
IGR
2100
2300
2200
1800
CH4 (ppb)
CH4 (ppb)
2200
1900
2300
PAL
2100
2300
2000
1800
CH4 (ppb)
CH4 (ppb)
2200
2100
2300
TER
2100
2300
ZEP
1800
CH4 (ppb)
CH4 (ppb)
2300
2200
2200
DEM
2100 2000 1900 1800
11/12
21/12
31/12
10/01
20/01
30/01
09/02
19/02
01/06
11/06
21/06
01/07
11/07
21/07
31/07
10/08
20/08
30/08
Figure 2. Calculated uncertainties by site and month (units of ppb) for the transport within the domain (a) and for the background mixing ratios (b).
VGN AZV KRS DEM YAK IGR NOY ZOT TER PAL ZEP ESP CHM FSD CDL ETL LLB CHL MHD CBA BAL TIK
70
a)
50 40 30 20
uncertainty (ppb)
60
10 0 0
VGN AZV KRS DEM YAK IGR NOY ZOT TER PAL ZEP ESP CHM FSD CDL ETL LLB CHL MHD CBA BAL TIK
2
4
6
8
10
12 70
b)
50 40 30 20 10 0 0
2
4
6
8
10
12
uncertainty (ppb)
60
Figure 3. Distribution of the observation-model CH4 mixing ratio mismatches a priori (blue) and a posteriori (red). Also shown are the assumed observation uncertainty distributions (grey). 12 10
CHL
60 50
8
40
6
30
4
20
2
10
0 −100
0
100
200
FSD
30 20 10 0
12
−100
0
100
CBA
CHM
20
40
10
20
−100
0
100
200
−100
0
100
200
ZEP
20
20
0
0 −200
−100
0
100
200
YAK
30
0
100
200
PAL
25
−100
0
100
200
−100
0
100
200
DEM
2
0
100
200
0
100
200
−200 40
−100
100
200
0
100
200
ZOT
30 20 10
0 −100
0
KRS
0 −100
10
−200
200
10
1 0
100
20
20
3
−100
30
30
4
0
IGR
−200 40
VGN
5
200
0
−200
AZV
100
5
0 −200
0
10
5
0
−100
15
10
2
200
20
15
4
−200 30
20
6
100
0 −100
25
8
0
5
−200 35
−100
10
40
40
200
BAL
60
60
100
15
−200 80
0
ESP
−200
MHD
0 −200
−100
0 −200
10
0
−200
80
20
2
6
200
30
4
7
100
40
6
10
0
60
50
8
12
0 −100
30
200
10
80
20
0 −200
ETL
40
−200
40
40
14
60
0 −200
50
LLB
0 −200
−100
0
100
200
obs - model (ppb)
−200
−100
Figure 4. Temporal distribution of atmospheric observations (number of observations per month).
ZEP TIK TER PAL NOY IGR YAK ZOT DEM CHL KRS BAL CBA LLB AZV VGN ETL CDL MHD FSD CHM 30 2005
2006
2007
2008
2009
2010
Year
2011
2012
2013
2014
Figure 5. Correlation of CH4 fluxes with different environmental parameters for 2005 to 2013. Shown are the correlations with soil temperature (a), soil water volume (b), precipitation (c) and snow depth (d). (Note white means no significant correlation).
a)
b) 1.0
0.0
Correlation
0.5
−0.5 −1.0
d) 1.0 0.5 0.0 −0.5 −1.0
Correlation
c)
