PAN

From Geos-chem
Revision as of 04:55, 17 September 2013 by Emily Fischer (Talk | contribs)

Jump to: navigation, search

Emily Fischer has updated the PAN simulation. The code has been merged with v9.02.h at this point, and a publication (Fischer et al., 2013) has been submitted to ACP.

Updates to Chemistry

1. I replaced the isoprene chemical mechanism with the Paulot scheme. This scheme has already been implemented into the standard code. Here are some bullets on the effect of the addition of nighttime chemistry:

- Nighttime reactions of NO3 increase PAN formation

- The difference in monthly mean simulated PAN is largest in regions with the largest PAN production (Eastern U.S. and East Asia)

- The addition of nighttime chemistry increases monthly mean surface PAN mixing ratios in these regions by ~50 pptv in spring and summer.

- This does propagate to higher altitudes, but the changes are very modest. Springtime PAN mixing ratios increase broadly over the Arctic by 5-10 pptv (~5%). The impact in summer is more heterogeneous at higher altitudes, but the overall increase is on the order of ~2% over most of the Arctic.

- These changes also DO impact ozone.

- This chemistry increases ozone over the Arctic during spring. Surface ozone increases by ~0.5 ppbv, and the impact is larger aloft (0.75 ppbv at 4 km)

2. I added nighttime chemistry from reactions of organic peroxy radicals with NO3 following Stone et al. (2013). This may not be incorporated into the standard chemistry, and needs to be discussed by the GEOS-Chem Steering Committee.

3. I updated the rate coefficients for the reactions of HO2 with the >C2 peroxy radicals to Equation (iv) in Saunders et al. (2003). These changes have been incorporated into the standard code.

4. I added several new NMVOCs. The extended mechanism includes ethanol, benzene, toluene and ethylbenzene (lumped), xylenes and trimethyl benzenes(lumped), and monoterpenes (lumped). Hydroxyacetone and methylglyoxal, which are species in past versions, are treated as tracers in this simulation. Hydroxyacetone has a 1-2 day lifetime. Methylglyoxal is treated as a tracer so that it can be emitted from biomass burning plumes and so I could track PAN production via this pathway.

- The ethanol code was provided by Dylan Millet.

- The inclusion and treatment of aromatics was motivated by Liu et al. (2010). I calculated the associated yield of methylglyoxal using recommended values for the individual aromatic species(toluene, o-xylene, m-xylene, p-xylene, 1,2,3-trimethylbenzene, 1,2,4-trimethylbenzene, and 1,3,5-trimethylbenzene) from Nishino et al. (2010) and the observed mean aromatic speciation for Chinese cities from Barletta et al. (2006). Thus the treatment is particular to the limited observations of aromatic speciation in China. It would be good to determine how different these ratios would be for the US, Europe or other developed/developing regions.

- We adopted the treatment of monoterpene oxidation from theRACM2 chemical mechanism (Goliff et al., 2013), lumping terpenes with one double bond (alpha-pinene, beta-pinene, sabinene and delta-3-carene) into one proxy. Unlike Ito et al. (2007), hydroxyacetone is not a product of terpene oxidation in the revised RACM2 mechanism used here. So I expect that the mechanism used here will make less PAN than that in Ito et al., 2007. This code was provided by Jingqiu Mao.

- I also added emissions of acetaldehyde from the ocean. This code was also provided by Dylan Millet.

5. I increased the deposition flux of PAN by assuming PAN reactivity with surfaces is more similar to O3 than NO2. (I changed F0 from 0.1 to 1.0 in drydep_mod.F. F0 is a ‘reactivity factor for oxidation of biological substances’)

- 0.1 is used for slightly reactive species (NO2, HNO2)

- 1 is used for highly reactive species (O3, H2O2, and recently OVOCs)

This change increases the global annual PAN deposition from 6.5 Tg/year to 10.7 Tg/year. This is a change of 0.47 Tg N/year. During the springtime PAN maximum, faster deposition decreased PAN at the surface by 15 – 20% (30 – 50 pptv) throughout the northern hemisphere. At higher altitudes, there is also a hemispheric impact, with PAN decreases of ~10% (20 – 40 pptv). In summer the surface decrease in PAN is most notable over the Arctic, but here it is on the order og 25%. At 4 km, the PAN decrease is about 10%, on the order of 5 – 10 pptv. These changes do propagate to ozone. The faster deposition causes a springtime decrease of ozone of almost a ppb at the surface and in the lower free troposphere. In summer, the changes to ozone are most pronounced for the Arctic, but still 0.5 ppbv over the eastern U.S. The limited evidence from the literature suggests faster PAN deposition than previously assumed. In our model this means f0=1 is probably better than f0=0.1, though a different deposition model would probably be ideal in the longer term.

To test the sensitivity of both species to a likely underestimate in the deposition velocity of PAN, I changed F0 from 0.1 to 1.0 in drydep_mod.F. F0 is a ‘reactivity factor for oxidation of biological substances’

0.1 is used for slightly reactive species (NO2, HNO2) 1 is used for highly reactive species (O3, H2O2, and recently OVOCs)

This change increases the global annual PAN deposition from 6.5 Tg/year to 10.7 Tg/year. This is a change of 0.47 Tg N/year. During the springtime PAN maximum, faster deposition decreased PAN at the surface by 15 – 20% (30 – 50 pptv) throughout the northern hemisphere. At higher altitudes, there is also a hemispheric impact, with PAN decreases of ~10% (20 – 40 pptv). In summer the surface decrease in PAN is most notable over the Arctic, but here it is on the order og 25%. At 4 km, the PAN decrease is about 10%, on the order of 5 – 10 pptv. These changes do propagate to ozone. The faster deposition causes a springtime decrease of ozone of almost a ppb at the surface and in the lower free troposphere. In summer, the changes to ozone are most pronounced for the Arctic, but still 0.5 ppbv over the eastern U.S. The limited evidence from the literature suggests faster PAN deposition than previously assumed. In our model this means f0=1 is probably better than f0=0.1, though a different deposition model would probably be ideal in the longer term.

Updates to Emissions

The PAN simulation uses the RETRO emission inventory as global default for anthropogenic NMVOC emissions aside from ethane (C2H6) and propane (C3H8). Ethane and propane emissions in RETRO were far too low compared to the GEOS-Chem inventories from Xiao et al. (2008), which are unbiased relative to observations. Emissions of both species appeared to be missing from the major natural gas production region in Russia. I didn't catch this when I did the acetone updates because these emissions don't produce biased results downwind of major anthropogenic source regions (China for example) compared to the existing data. I used the offline ethane emission inventories developed as in Xiao et al. (2008). I have not been able to find the full reference for the propane inventory that corresponds to the Xiao et al. (2008) ethane inventory. I have tried unsuccessfully to make contact with Yaping Xiao.

RETRO includes anthropogenic emissions for benzene (BENZ), xylene (XYLE) and toluene (TOLU). Based on the observed CO to benzene ratio for TRACE-P, I increased benzene emissions over China by 25%. The CO/benzene ratios were examined by Chris Miller. I then scaled xylene and toluene emissions to benzene based on measurements from 43 Chinese cities from Barletta et al. (2006). In summary, RETRO emissions of toluene were increased by a factor of 4 over China to create the lumped toluene (TOLU), and RETRO emissions xylene were increased by a factor of 8 over China to create the lumped xylene species (XYLE). I suspect that upward scaling may also be appropriate in other developing cities, but this would need to be carefully examined and I am not sure what data is available.

I made several updates to the way fire emissions are incorporated into the model.

1) I updated the emission factors for NMVOCs and NOx from extratropical forests, savannas and agricultural fires from Akagi et al. (2011). These are the only categories that overlap with GFED3, and this is why these are the only ones to be updated. The updated NOx emission factor for extratropical fires is approximately a factor of three lower, and the emission factors for the NMVOCs are generally higher. Sometimes they are much higher.

2) Following Alvarado et al. (2010) I directly partition 40% and 20% NO emissions from fires directly to PAN and HNO3 respectively. The Alvarado et al. (2010) partitioning is based on an observation of fresh boreal fire plumes over North America in summer, but I have applied it to all fire types. To support PA radical formation on faster timescales I also added GFED3 emissions of several shorter lived hydrocarbons to the suite of species emitted from fires: terpenes (APINE), aromatics(XYLE, TOLU) along with additional oxygenated species (hydroxyacetone (HAC), and methylgloyxal(MGLY)).

3) The standard version of GEOS-Chem releases all fire emissions in the boundary layer, but injecting a fraction of the emissions above the boundary layer is important for PAN because temperatures above the boundary layer enhance its stability. I distribute 35% of biomass burning emissions by mass in the 10 sigma layers (4 km) above the boundary layer, and this improves the comparison with PAN observations at high latitudes.

4) Finally, I increased wild fire emissions by 60% in North Asia (30 – 75°N, 60 – 190°E), 25% in Canada and 50% in Alaska. The increase over Asia was motivated by Kaiser et al. (2012). This region is one region of comparison in this paper. For Canada and Alaska, I deferred to the analysis of fire records by Xu Yue.