1. Satellite and model analysis of wildfire NOx emissions in Siberia: Links to interannual variability of surface ozone, 1998–2004 Hiroshi Tanimoto National Institute for Environmental Studies, Japan In collaboration with K. Folkert Boersma, Ronald van der A, Kiyoshi Matsumoto, Mitsuo Uematsu, Philippe Le Sager, Bob Yantosca Photo: Siberia (courtesy of T. Machida)

2. Background & Motivation Role of boreal forest fires in interannual variations of trop. O 3 in N.H. (especially, Rishiri Island, northern Japan) Impacts on CO is well-known (Yurganov, Edwards, etc), but O 3 is less known. Still lots of discussions on magnitude of O 3 increase. O 3 enhancement in “fire plumes” Cheeka Peak (Jaffe et al.), Pico- NARE (Val Martin, Honrath, et al.), Rishiri (Tanimoto et al.), etc O 3 “background-levels” Mace Head (Simmonds et al.), Pico- NARE (Lapina et al.) O 3 enhancements are subtle, “several ppbv” – challenge NOx is a key species, but not extensively examined, because of limited measurements nearby fires Rishiri Island, Japan

3. Questions & Tools GEOS-Chem Version: v8-01-01 Met. Field: GEOS4 Horizontal Grid: 4x5 deg Vertical Layers: 30 layers Tracers: 43 species Emissions: GFEDv2, monthly Period: Apr – Sep, 1998, 1999, 2000, 2001, 2002, 2003, 2004 GOME 1998-2002, 10:30LT, 40x320 km SCIAMACHY 2002-2004, 10:00LT, 30x60 km TEMIS (KNMI, The Netherlands) monthly grid data cloud-free & nearly cloud-free (cloud radiance <50%) Can satellites see NOx (NO 2 ) enhancement due to boreal fires? Does the GFED-driven model reproduce the NOx enhancement, and predict O 3 enhancement? Are satellites, model, and surface data consistent with each other?

4. Distributions of NOx emissions in GFEDv2 Rishiri Island KAMFESESWS + Target region FES (Far-Eastern Siberia) & ES (Eastern Siberia) “low-fire-year”: 1999, 2000, 2001, 2004 = reference “high-fire-year”: 1998, 2002, 2003

5. Anomalies in trop. NO 2 column 1998 JAS 2002 JAS 2003 AMJ GFEDv2 NOx emis. GOME/SCIA NO 2 GEOS-Chem NOx Locations of NOx enhancement are consistent for satellites and G-C

6. GOME/SCIA vs. GEOS-Chem Satellites can detect NOx emissions from “large” and “medium” scale fires GEOS-Chem agrees well with satellites in a qualitative manner Satellites / GEOS-Chem

7. Anomalies in ozone by G-C, surface level (L=1) Rishiri

8. Anomalies in surface ozone, obs vs. G-C G-C did a good job in reproducing anomalies observed at Rishiri Rishiri Island, Japan (45N, 141E) Rishiri Is. Summary GOME/SCIA can detect NO 2 enhancement from boreal fires in Siberia G-C reproduces NOx enhancement – need more quantitative evaluations (sampling time, 2x2.5 grid, etc) G-C well reproduces O 3 anomalies at Rishiri Island in BB-years

10. Anomalies in ozone by G-C, alt.-lat cross section 120 degE140 degE (Rishiri)155 degW (Barrow) 1998 2002 2003 Rishiri Barrow

11. Trop. NO 2 column over Siberia viewed from space Mean (summer, 1999, 2000, 2001, 2004) In general, NO 2 is very low over Siberia, due to small anth. activities Enhancement of NO 2 is negligible in “low-fire-year”

12. Trop. NO 2 columns in 1998, 2002, 2003 Weak but significant enhancement of NO 2 in 1998, 2002, 2003 Locations of NO 2 enhancements differ depending on year 1998 summer 2003 spring 2002 summer