Recent changes in atmospheric methane

Transcription

1 Recent changes in atmospheric methane

2 Recent changes in atmospheric methane

3 Recent changes in atmospheric methane

4 Recent changes in atmospheric methane

5 Recent changes in atmospheric methane

6 Recent changes in atmospheric methane

7 Recent changes in atmospheric methane

8 Recent changes in atmospheric methane

9

10 Another look at it CH 4 (ppb) Bootstrapped Mean Northern Hemisphere: Southern Hemisphere: De-seasonalize observational records from each site - Split observations based on hemisphere - Bootstrap hemispheric averages and uncertainties Data from NOAA/ESRL 1,2,3, U. Heidelberg 2, UCI 2, UW 2, & GAGE/AGAGE 1 = CH4, 2 = δ 13 CH4, 3 = CH3CCl3

11 Methylchloroform Used as solvent and insecticide prior to the Montreal Protocol (is also ozone depleting substance) Emissions phased out in 1989

12 Ambiguity in the causes for decadal trends in atmospheric methane and hydroxyl Turner, Frankenberg, Wennberg, Jacob (in press, PNAS) Can use a simple 2-box S = E X,N (t) k [X] [OH] N (t)[x] N (t)+ [X] S(t) [X] N (t) NS = E X,S (t) k [X] [OH] S (t)[x] S (t)+ [X] N (t) [X] S (t) NS

13 Jacobians

14 Jacobians

15 Jacobians

16 Fits

17 Formal Inversion Minimize T -1 T 1 ( x xa) Sa ( x xa) + ( y Kx) Sε ( y Kx) State vector x (size 2940) includes: NH and SH emissions each month isotopic composition of those OH scaling factor for each month Measurement vector y (size 2280) size includes: NH and SH abundances NH and SH isotopic composition NH and SH MCF abundances

18 Prior run

19 Posterior (non-linear inversion) CH 3 CCl 3 (ppt) CH 4 (ppb) Observations Northern Hemisphere Southern Hemisphere δ 13 CH 4 ( )

20 Flux Inversion CH4 lifetime fixed CH4 Emissions (Tg/y) CH4 emissions 13 C of source (permil) Average Isotope source OH anomaly (%) OH Anomaly NH Fixed OH SH Fixed OH Average Fixed time

21 With OH fitted (through MCF measurements) CH4 Emissions (Tg/y) CH4 emissions 13 C of source (permil) Average Isotope source OH anomaly (%) OH Anomaly NH Var. OH SH Var. OH Average Var. OH NH Fixed OH SH Fixed OH Average Fixed time

22 With OH fitted (through MCF measurements) CH4 Emissions (Tg/y) CH4 emissions 13 C of source (permil) Average Isotope source OH anomaly (%) OH Anomaly NH Var. OH SH Var. OH Average Var. OH NH Fixed OH SH Fixed OH Average Fixed time

23 With OH fitted (through MCF measurements) CH4 Emissions (Tg/y) CH4 emissions 13 C of source (permil) Average Isotope source OH anomaly (%) OH Anomaly NH Var. OH SH Var. OH Average Var. OH NH Fixed OH SH Fixed OH Average Fixed time

24 Is OH variation realistic?

25 Previous publications find a similar behavior Montzka et al., Rigby et al., & McNorton et al. find similar OH anomalies

26 Holmes et al 10.0 R 2 = 0.90 R 2 = 0.88 τ CH OH, y CTM 5-Parameter Model UCI CTM Oslo CTM3 GEOS-Chem/GEOS-5 GEOS-Chem/MERRA R 2 = R 2 = Fig. 1. Methane lifetime due to oxidation by tropospheric OH ( CH4 OH) simulated by each CTM (solid lines) and reconstructed from the 5-parameter model (dashed lines). The parameters are air temperature, water vapor, ozone column, lightning NO x emission, and biomass burning emission. Parameter values for each CTM are

27 Mode times Prather Assume that V is a solution for a generic operator A, dv/dtza[v ], then the continuity equation for a perturbation dv can be derived as ddv d½v CdV Š Z K dv dt dt dt Z A½V CdV ŠKA½V Š Z J V $dv CorderðdV 2 Þ; where the Jacobian matrix J of the operator A is the first term in the Taylor expansion of A[VCdV] evaluated at V, i.e. the linearized system (e.g. Prather 1996, 2002; Manning 1999). The Jacobian matrix has dimension m!m, where m is the number of variables. Species are inherently discrete, and adopting a discrete spatial grid gives us a finite number of variables across (x, y, z, n). For discussion of the continuum, see 8. If the perturbation dv is an eigenvector of J V with eigenvalue l, then ddv Z J dt V $dv Z ldv ; Z

28 Coupled OH Chemistry Prather it to carbon isotopes, but in this example, the original box model is retained. Consider a simplified model of the CH 4 CO OH chemical system as having three reactions ðiþ CH 4 COH0/0CO R 5 Z k 5 ½CH 4 Š½OHŠ k 5 Z 1:266!10 K7 s K1 ppt K1 ðiiþ CO COH0/ R 6 Z k 6 ½COŠ½OHŠ k 6 Z 5:08!10 K6 s K1 ppt K1 ðiiiþ OH CX0/ R 7 Z k 7 ½XŠ½OHŠ k 7 ½XŠ Z 1:062 s K1 ; d½ch 4 Š dt d½coš dt d½ohš dt Z S CH4 K R 5 Z S CO CR 5 K R 6 Z S OH K R 5 K R 6 K R 7 : O 3 + hv ( < 330nm)! O( 1 D) + O 2, O( 1 D) + H 2 O! 2OH.

29 Prather d½ch 4 ŠðtÞzC0:995 e Kt=13:6 C0:005 e Kt=0:285 ppb;

30 Prather pre-inclustria J S(CH4) (ppb/yr) o E 30 E,"', 20 o o 10 E looo CH4 (ppb)

31 Lelieveld et al, 2016 eactivity near the Earth s surface in s 1. tropopause and the poles CH4 is he mean CH4 in the extratropics nner tropics 6.1 years. The mean in the FT 9.1 years. The effective ncentration and CH4 between the troposphere is about a factor of OH and HO 2 range between the h latitudes. This is much smaller de gradients and the seasonal cycle indicative of the important role of. 6). The NH / SH ratio of CH4 is nd latitude contrasts are found for ic CO ( CO ) due to reaction with ut 38 days in the tropics, 65 days d 86 days in the SH extratropics, CO is ecycling probability ary of global annual mean HO x oposphere, also listed in Table 1, f sources and sinks. Primary OH 1) and (R2) (P, purple) amounts h about 85 % takes place in the H formation (G) and HO 2 procount for about 85 % of the tro- Table 1. Global, annual mean tropospheric source and sink fluxes of OH (Tmol yr 1 ). Sources and sinks are also specified for the boundary layer and free troposphere. Sources/sinks BL FT Troposphere O( 1 D)+H 2 O (33 %) NO+HO (30 %) O 3 +HO (14 %) H 2 O 2 + hv (10 %) OVOCs, ROOH+hv (13 %) Total OH sources OH+HO 1 y (18 %) OH+NO 2 y (1.5 %) OH+CH (12 %) OH+CO (39 %) OH+other C 1 VOC (15 %) OH+C 2+ VOC (14 %) Rest (0.5 %) Total OH sinks H 2,O 3,H 2 O 2, radical radical reactions. 2 NO, NO 2, HNO 2, HNO 3, HNO 4, ammonia, N-reaction products. 3 VOC with one C atom (excl. CH 4 ), incl. CH 3 OH, C 1 -reaction products. 4 VOC with 2 C atoms, C 2+ -reaction products. is rather evenly distributed between different environments

32 Lelieveld et al, Unit: 10 molecules cm OH OH in 10 5 molecules cm 3. Left: tropospheric, annual mean. Right: zonal annual mean up to 10 hpa. verage boundary layer height, the upper dashed line the mean tropopause and the solid lines the annua use height.

33 Lelieveld et al, 2016 Figure 11. Zonal annual mean OH concentrations calculated in the reference simulation (black) and by successively excluding OH recycling through the NO x,o x and OVOC mechanisms.

34 Holmes et al Table 2. Sensitivity of CH4 OH to climate variables and emissions a. Variable b UCI CTM Oslo CTM3 GEOS-Chem Literature c Adopted d Chemistry-climate interactions Air temperature e ± 0.8 Water vapor e ± 0.03 Ozone column, 40 S 40 N f g [1] ± h [2] g [3] Lightning NO x emissions [4] 0.16 ± 0.06 Biomass burning emissions i ± j CH 4 abundance k [5] ± ± 0.03 [6] (f = 1.34 ± 0.06) l Convective mass flux N Optical depth, ice clouds N Optical depth, water clouds [2] N [3] Anthropogenic emissions Land NO m x [5] 0.14 ± ± [7] Ship NO x ± 0.01 [8] 0.03 ± ± [9] [10] Aviation NO x ± [11] ± CO [5] ± ± [7] VOC [5] ± ± 0.01 [7] a Sensitivities are reported as d ln( CH4 OH)/d ln(f ) for each variable F, based on perturbation tests described in Sect. 3.2.

35 Murray et al, 2013 e 7. (a) Monthly mean percent anomalies in global mean OH that inferred from methyl chl

36 Murray et al, 2013 ure 8. (a) Interannual variability in OH contributed from chemical sources and sinks in th

37 Murray et al, 2013 Table 2. Sensitivity of Simulated IAV in OH to Different Reaction Pathways, Climate Variables, and Emissions Slope of doh/dp b in Monthly Anomalies Parameter, P RWith OH Anomalies a (%/%) of P c (%) OH anomalies Production anomalies Primary production Water vapor Stratospheric ozone Tropospheric ozone HO x recycling Loss anomalies k CH4 +OH(T)[CH 4 ] k CO+OH (T)[CO] Global emission rates Lightning Biomass burning d a Pearson correlation coefficient, R, between time series of OH percent anomalies and the forcers. b Slope of reduced major axis (RMA) regression between monthly percent anomalies in OH and the forcers. Range gives 95% confidence intervals calculated from a bootstrap ensemble with 10 3 members. c Standard deviation of monthly percent anomalies in tropospheric mean climate variables, reaction rates, and emissions. All values calculated from the simulation using IAV in lightning from LIS. d Statistics for fire emissions are calculated using time series of NO x emission; results are very similar for CO emissions, and negative because fires act as a net sink for OH.

38 D24306 OMAN ET AL.: 21ST CENTURY OZONE EVOLUTION D24 Oman et al Ozone recovery Figure 1. Annual average (a c) total and (d i) partial column ozone amounts over three regions including the tropics (25 S 25 N) and midlatitudes of each hemisphere (35 60 S and N). The partial