HOx and NO Observations during INTEX-A

X. Ren J. Mao R. Long R. Lesher W. BruneDepartment of Meteorology

Pennsylvania State University

HOx and NO measurement techniques

• OH and HO2 measurementsATHOS — Airborne Tropospheric Hydrogen Oxides

Sensor– Laser-induced fluorescence (LIF) detection of OH

– Chemically convert HO2 to OH by HO2+NO followed by the detection of OH with LIF

• NO measurementsTEI 42C NO-NOx analyzer

– Chemiluminescence– NO single mode– Online NO span and zero checks

Data Quality

• Data coverage:OH (1 min) - 97%HO2 (1 min) - 95%NO (1 min) - 89%

• Typical uncertainties: HOx ±32% (2σ)NO ±30% (2σ)

• Detection limits: OH 0.01 pptvHO2 0.1 pptvNO 50 ppt

0

5

10

15A

LTP

(km

) ALT P

0

0.2

0.4

0.6

OH

(ppt

v) OH obs

0

20

40

HO

2 (ppt

v) HO2 obs

12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:000

500

1000

NO

(ppt

v)

Hour (UTC)

NO obs

HOx and NO observations (July 22, Flight 11)

HO2 and OH have good precision – sub-minute resolution will be used to examine variability.

Observed & PSS NO vertical profiles

1 10 100 10000

2

4

6

8

10

12

NO (pptv)

ALT

P (k

m)

NO obsNO obs medianNO PSS median

0 1 2 30

2

4

6

8

10

12

NO obs / NO PSS

ALT

P (k

m)

The NO values between 2-6 km are around or below the NO detection limit (~50 pptv).

0 20 400

2

4

6

8

10

12

HO2

Alti

tude

(km

) obsobsmod

0 1 2 3HO2 obs/mod

INTEXINTEXTRACEPPEMTB

• Median observed-to-modeled OH ~ 0.6 at all altitudes.• Median observed-to-modeled HO2 ~ 0.8 up to 8 km.• Behavior is similar to that in TRACE-P.• Large observed-to-modeled OH in PBL correlates to

isoprene (from Jim Crawford) as seen in forests.

0 0.2 0.4 0.60

2

4

6

8

10

12

OH

Alti

tude

(km

)

obsobsmod

0 1 2 3OH obs/mod

INTEXINTEXTRACEPPEMTB

(pptv) (pptv)

50 1001500

2

4

6

8

10

12

CO (ppbv)

Alti

tude

(km

)

101 102 103

NOx (pptv)50 100

O3 (ppbv)0 50 100 150

0

2

4

6

8

10

12

HO2/OH

Alti

tude

(km

)

obsobsmod

0 1 2 3HO2/OH obs/mod

INTEXINTEXTRACEPPEMTB

• NOx in INTEX-A is greater than in TRACE-P & PEM TB; CO and O3 are similar in INTEX-A & TRACE-P.

• Observed-to-modeled HO2/OH is close to 1 below 7 km, but exceeds 2 above ~9 km.

• HO2/OH deviations appear to be NOx related.

0.2

0.4

0.6

OH

obsobsmod

100 1020

1

2

3

OH

obs

/mod

NOx (pptv)

INTEXINTEXTRACEPPEMTB

20

40

HO

2

obsobsmod

100 1020

1

2

3

HO

2 obs

/mod

NOx (pptv)

INTEXINTEXTRACEPPEMTB

• Observed-to-modeled OH shows little NOx-dependence.• Observed-to-modeled HO2 grows for NOx > few 100 pptv.• INTEX-A and TRACE-P dependences on NOx are similar.• Observed-to-modeled HO2 < 1 for NOx < few 100 pptv &

> 1 for NOx > few 100 pptv is usually observed by us and a few others.

HO2 versus (PHOx)1/2

0 2000 4000 6000 80000

2

4

6

8

10

12

14

sqrt [P(HOx)] (sqrt [cm-3 s-1])

HO

2 obs

(x10

8 cm

-3)

• P(HOx) = L(HOx) ∝ [ HO2 ]2, so [ HO2 ] ∝ sqrt {P(HOx)}.• Much HO2 variance can be explained by P(HOx).

HOx observed & modeled comparisons

0.01 0.1 1 100.01

0.1

1

10

OH mod (pptv)

OH

obs

(pp

tv)

median obs/calc =0.58 r2 = 0.64

1 10 1001

10

100

HO2 mod (pptv)

HO

2 obs

(pp

tv)

median obs/calc =0.77 r2 = 0.65

• Solid line: 1:1; dashed lines: obs. uncertainty ±32%.• HOx comparison similar to that in TRACE-P.

Modeled OH production and loss

105 106 107 1080

2

4

6

8

10

12

P(OH) (cm-3)

ALT

P (k

m)

POHPOH medianO1D+H2OPeroxidesHO2+NO

104 106 1080

2

4

6

8

10

12

L(OH) (cm-3)

ALT

P (k

m)

LOHLOH medianOH+COOH+NO2OH+HCHO

Main P(OH) is O1D+H2O (below 5 km) and HO2+NO (above 5 km).Main L(OH) is OH+CO/VOC.

Modeled HO2 production and loss

105 106 107 1080

2

4

6

8

10

12

P(HO2) (cm-3)

ALT

P (k

m)

104 106 1080

2

4

6

8

10

12

L(HO2) (cm-3)

ALT

P (k

m)

PHO2PHO2 medianOH+COHCHO+hvOH+HCHO

LHO2LHO2 medianHO2+NOHO2+O3HO2+RO2

Main P(HO2) is OH+CO. Main L(HO2) is HO2-RO2 self-reactions (below 5 km) & HO2+NO (above 5 km).

O3 budget

0.01 0.1 1 100

2

4

6

8

10

12

P(O3) (ppb hr-1)

ALT

P (

km)

0.01 0.1 1 100

2

4

6

8

10

12

L(O3) (ppb hr-1)

ALT

P (k

m)

PO3PO3 medianPO2-HO2PO3-RO2

LO3LO3 medianLO3-O1DLO3-OHLO3-HO2

-0.5 0 0.5 1 1.50

2

4

6

8

10

12

P(O3), L(O3), P(O3)-L(O3) (ppb hr-1)

ALT

P (

km)

P(O3)L(O3)P(O3)-L(O3)

• Main P(O3): HO2+NO.

• Main L(O3): O1D+H2O (< 5 km) & O3+HO2/OH (> 5 km).

• Net O3 loss at altitudes between 1 km and 5 km.

Science questions we hope to answer • General comparisons between observed and modeled HOx

– Were previous observed-to-modeled anomalies also observed in INTEX-A? (e.g., NOx-dependence of observed-to-modeled HO2)

– Can the HOx heterogeneous effects (or lack thereof) be understood?

• High speed photochemistry – one-to-a-few seconds– What are the effects of scale on calculating P(O3) from HO2 & NO?

– Is HOx behavior understood in urban, forest-fire, and long-range regionally transported plumes?

• HOx behavior in the planetary boundary layer– What is the behavior of HOx and P(O3) and vertical distribution in the

boundary layer?– Is isoprene chemistry in forested regions adequately understood?

• Collaborations with many others on these & other questions.