Dandelions 2006 - first results

1
Ground-based measurements of tropospheric NO2 -
Comparison of MAX-DOAS results to other
instruments for DANDELIONS-
Folkard Wittrock, Hilke Oetjen, Anja Schönhardt,
Andreas Richter, John P. Burrows - IUP
Bremen Hester Volten, Daan Swart - RIVM Ellen
Brinksma - KNMI
2
Outline

• DANDELIONS campaign
• MAX-DOAS
• Profile Retrieval Algorithm BREAM
• Results focusing on 2nd campaign in 2006
• BREAM / geometric approach
• BREAM / in situ
• BREAM / LIDAR
• BREAM / satellite
• Conclusions and Outlook

3
DANDELIONS
Dutch Aerosol and Nitrogen Dioxide Experiments
for vaLIdation of OMI and SCIAMACHY
CESAR site (Cabauw) located at 51.971N, 4.927E
0.70 m below ASL Dandelions Objectives Validati
on of satellite measurements Intercomparison of
instruments (NO2, O3, aerosols) NO2 ground in
situ, lower tropospheric profile (lidar), total
columns MAX-DOAS intercomparison -gt
ideal for evaluation of MAX-DOAScapabilities for
profile retrieval
4
MAX-DOAS

• Detection
• UV/vis measurement of scattered sunlight measured
  for different lines of sight
• Method Differential Optical Absorption
  Spectroscopy - DOAS
• Result averaged absorption along all
  contributing light paths -gt Slant Column
• Retrieval
• Correction for vertical sensitivity (airmass
  factors) depending on several meteorological
  parameters (e.g. albedo, aerosol)
• Optimal Estimation for profile retrieval from
  ground (BREAM)
• Results
• VC and profile information for the lower
  troposphere

5
Multi Axis (MAX)-DOAS
Zenith Stratosphere
O3
NO2
BrO
IO
OClO
IO
NO2
H2O
BrO
HCHO
O3
CHOCHO
Horizon Troposphere
6
MAX-DOAS during Dandelions 2006 Some Technical
Details

• The Bremen DOAS instruments
• Spectrometer (f275mm) for visible Wavelength
  range 399.5 715.0 nm
• Spectrometer (f500mm) for UV Wavelength range
  324.0 409.0 nm
• One fixed azimuthal viewing direction

The telescope on the roof of the portocabin
E
W

• 11 Lines of sight
• Zenith
• Scan from 0 to 16 in 2 steps
• 30 elevation
• Integration time 1 min
• Full scan duration 12 min or5 scans per hour

7
MAX-DOAS during Dandelions 2006
8
MAX-DOAS during Dandelions 2006
9
Dandelions 2006 NO2 sequence
Sun
Sat
10
Dandelions 2006 NO2 sequence
Sun
Sat
11
Bremian Advanced MAX-DOAS Retrieval Algorithm -
BREAM
Flowchart for BREAM
12
Bremian Advanced MAX-DOAS Retrieval Algorithm -
BREAM

• Example of input data and results for BREAM
• usually 1 hour temporal steps comprising 3 to 5
  scans
• retrieval grid 50 m
• for NO2 two wavelength regions are used to
  benefit from different effective light paths

13
Dandelions 2006 NO2 in the UV and vis
Zenith

• Scans show smaller NO2 values for the UV
• due to different air
• mass factors (effective light paths)
• -gt valuable information for profile retrieval

scans
Wavelength ranges Vis 450-497 nm UV 345-365 nm
14
Bremian Advanced MAX-DOAS Retrieval Algorithm -
BREAM

• Averaging Kernel Matrix
• Only one scan
• for SZA about 35
• trace of A (number of pieces of independent
  information) 2.8
• illustrates the high sensitivity to the
  lowermost layers while for altitudes more than
  one kilometre there is only very little height
  information

15
Bremian Advanced MAX-DOAS Retrieval Algorithm -
BREAM
Influence of some important input parameters on
the retrieved profile (sensitivity study for one
day) lt-gt small effect on derived tropospheric
column, reflecting the fact that some parameter
errors have different effects at different
altitudes (e.g. aerosol)
16
Bremian Advanced MAX-DOAS Retrieval Algorithm -
BREAM
17
Results
18
Results
19
Results
20
Results
21
Results
22
Results
23
Results
24
Results
25
Results
26
Results
27
Results
28
Results
29
Results
30
Results
31
Results
32
Conclusions and Outlook

• MAX-DOAS NO2 profile and tropospheric column data
  for almost all golden days
• Tropospheric vertical column agrees well with
  pure geometric approach lt-gt in agreement with
  previous studiesVC is not very sensitive to
  retrieval parameters
• Comparison to in situ measurements contradictory
  general variation captured well, in particular
  for polluted days in situ much higher lt-gt
  interference problem??
• Comparison with NO2-LIDAR much better in terms of
  absolute values, but profile shape of the LIDAR
  is different
• All in all not very surprising NO2 is very
  variable in time, in space (and in experiment)
  lt-gt in particular more profile measurements needed

33
Results