Greenland%20Environmental%20Observatory%20~%20Summit

1
Greenland Environmental Observatory Summit
Cooperating Agencies

• GEOSummit Science Coordination Office
• Jack Dibb
• John Burkhart
• Roger Bales

2
Summit Camp
72o34N, 38o29W 3,250 m.a.s.l.

• Manned year-round since 2003 by at least 4
• Operated by US NSF OPP under MOU with Denmark and
  Greenland
• Site of GISP2 and GRIP ice cores drilled 1989 -
  1993
• Special issue of Atmos. Env. on recent Summit
  studies in 2007

Dome protecting the ice core drill
3
Year Round Sampling of Trace Gases
NOAA ESRL/GMD sampling at Summit demonstrates the
value of the site as a unique, high latitude,
free troposphere observatory.

• Opposing secular trends in CO2 and d13C reflect
  anthropogenic input, antiphase seasonality due to
  C uptake by terrestrial biosphere.
• Methane and CO show little recent change, yet
  clear seasonality.
• N2O and SF6 (minor GH gases) show anthropogenic
  increases.

4
Hydrocarbon Seasonality

• Seasonal cycles of hydrocarbons mainly reflect
  anthropogenic emissions and a weaker winter OH
  sink
• Reproducibility of cycles over 8 years indicates
  source/sink balance, providing sensitive baseline
  to detect/quantify future changes
• (Whole air samples analyzed by the D. R. Blake
  group at UC Irvine)

5
All Summit data 2000-2004
Profiles at 4 polar stations are the mean of all
available sonde data 1995-2004, X marks the
median mixing ratio at Summit.
From Helmig et al., 2007
2004 at Summit
6
7Be and 210Pb Radionuclides

• Summer peak in 7Be unique at Summit among Arctic
  sites, indicating significant seasonal influence
  from the upper troposphere/lower stratosphere
• 210Pb concentrations decrease with increasing
  elevation to Summit. Distinct lack of a winter
  peak confirms that Arctic Haze rarely extends to
  3 km altitude over Summit
• (From Dibb, 2007)

7
Filter sampling with 48 hour integration time for
the radio-tracers was conducted at Summit in the
1997-98 pilot overwinter campaign, again for
the 2000-2002 experiment, and since August 2003
to present. Dibb (2007) reported on results
through July 2005. Here are observations during
the first phase of ARCTAS/ARCIONS. Note that the
peak of 7Be in late April is the highest seen
since the 2-day sampling interval was initiated
in 1997! Unfortunately, NASA airplanes went home,
and I believe ARCIONS took a break, on 19 or 20
April.
8
Everyone came back in the summer for phase 2.
Coupling between 7Be and 210Pb even more
striking than in April. What does bromide have
to do with any of this? Mostly, means I am
getting ahead of myself, but for reactive gases
it is now well established that Summit is not
really a great free trop observatory.
9
In 1998 production of NOx in snow was discovered
at Summit (also Alert and South Pole)
Snow Photochemistry is now a hot topic at
Summit, and elsewhere.
10
We have learned... ..that the way impurities get
incorporated into snow (and ice cores) is much a
more complex process than we thought!
More like this model
Than this!
11
Formation Mechanisms??
Studies in both polar regions, and seasonal
snowpacks have confirmed that photolysis of
nitrate in snow releases NOx into the firn air.
Clearly, OH is enhanced above the snow and is
likely even higher in the upper part of the pack.

It is not yet clear whether subsequent cycling of
N oxides is dominated by homogeneous reactions
(in and above the snow) or is mediated by surface
chemistry in the porous snowpack.
Oxidation of the abundant (but poorly
characterized) supply of organic compounds in the
snow appears to result in production of a slew of
reactive hydrocarbons.
12
HOx Measurements vs. Model
HOx measurements above polar snow are limited, so
far, to South Pole, Summit, and Halley. Levels
highest at Summit, particularly for OH (summer
median values of OH about 4 times higher at
Summit than South Pole). Standard models,
constrained by observations of presumed main
precursors, reasonably simulate HO2 at Summit,
but can not capture OH enhancements. Note that
model underestimates OH most severely in high
winds. From Sjostedt et al. 2007 (in the special
issue)
13
HO Measurements vs. Model (Sjostedt et al., 2007)
Obs. HO (1E6
molec./cm3)
Obs. HO (1E6
molec./cm3)
Note the different scales! Obs. gt Model
consistently
14
Composite 24 hr measurements and model
predictions of OH during 2003 summer campaign.
Predicted values obtained from NMHC model
constrained by H2O2, CH2O and HONO from Sjostedt
et al., 2007.
15
A couple more puzzling observations at Summit,
particularly when compared to South Pole. Note
that Detlev has much more O3 flux data now,
confirms the snow/surface is often a sink at
Summit.
RO2 NO RONO2
BL
O3 vertical profile measured at Summit,
Greenland by Helmig D.et al.,2007
Ozone production rate due to standard
photochemistry from Summit 2003 data
16
Effect of Bromine Chemistry on Summit
Photochemistry

• BrO can significantly shift the HO2 and OH
  partition while has a limited impact on total
  peroxy radical levels.
• BrO does not have significant impact on NO2/NO
  ratio and is a minor sink for CH2O.
• Model estimated net ozone production rates ranges
  from 0.6, -0.8, to -6 ppbv/day, corresponding to
  2, 4, and 10 pptv of BrO.
• Near constant O3 observations is consistent with
  a few pptv of BrO.

17
2. Key bromine chemical cycles in polar BL
BrO
Br2
OH
Bromine chemical cycles that induce ozone
depletion, mercury depletion and impact the
RO2/OH ratio.
18
G-SHOx May-June 07, June-July 08 Greenland
Summit Halogens and HOx
Species Method Research Group
Hg (GEM, RGM, FPM) Tekran NOAA - Brooks
OH, RO2, H2SO4 CIMS Ga. Tech - Huey
HCl, HO2NO2, BrO,SO2, etc CIMS Ga. Tech - Huey
BrO, HCHO DOAS UCLA - Stutz
Soluble Ions (Br-, Cl-, etc.) Aerosol , area distribution MC/IC, Filters, Snow SMPS UNH - Dibb
Radiation, J-values BL profiling (met and O3) Spectrometer Tower/Tether Sonde UH - Lefer
Whole Air Samples GC-MS UCI Blake
NO, O3, Dewpoint, etc commercial Ga. Tech - Huey
19
Comparison of OH predictions from the HOx-NOx-CH4
model to OH measurements, plotted with RGM and O3
concentrations. Note the model has trouble when
RGM (Br?) is enhanced.
20
2007 LP-DOAS/CIMS intercomparison
21
BrO observation by CIMS in 2008
22
Gas phase soluble Br- in the air just above snow
was also generally lower in 2008 than 2007. May
be a seasonal feature, but can not rule out
interannual variability in transport and/or
ambient conditions. Note that soluble Br- shows
modest increase in first 2008 CIMS BrO event, but
is quite low 20 - 21 June.
23
Simultaneous measurements above the snow and at
10 cm depth in the snowpack were made on 2 days
in 2007. Both cases the mid-day peak was more
pronounced in the firn.
24
In 2008 such measurements were made on 2/3 of all
sampling days. Mixing ratios in the firn air
were lower in 2008 than 2007, but tended to be
higher than those above the snow. Again, firn
air enhancement largest in early afternoon. Is
the snow one of the sources of Br at Summit?
25
In the surface layer of snow bromide was always
quantifiable, albeit low compared to major
ions. If 5 nmol Br kg-1 in the top cm of snow
was released to a well mixed 100 m deep boundary
layer as a gas the mixing ratio would be 165 ppt
(compare to soluble Br- mainly lt 2 ppt). Note
that bromide in snow was greatly in excess of the
sea salt ratio (based on sodium).
26
FT
partly halogen processed air partly clean partly
marine partly BB

• strong diurnal variation
• long range air mass changes
• intermittent BL venting

aerosol, gases
300m
T
gases
inversion
blowing snow
gases trapped at night mixed during day
fog
Br-
Br reservoir is snow atmosphere 1
summer, year -1
Br reactivity
Hg
Cl reactivity
summer, year -2
27
no Br/Cl influence on VOC no Arctic Haze
Early spring
no transport
processes in blue are critical, need checking!!
no elevated Br- in snow pack
autumn/winter - no Br/Cl influence on VOC
- no elevated Br- in snow pack
Bromine explosion
Arctic Haze
1st year sea ice frost flowers
deposition of Bry via snow fall - Br- peak
in summer snow pack - ambient gas and aerosol
Bry is low
Late spring/early summer
active photochemistry
long-range transport
snow pack photochemistry - release of NOx to
air - high snow pack Br- - high Br
VOC-reactivity in snow
cold production of RGM warm no
RGM production
Convection (open leads)
in low elevation sites, snow-bound RGM might get
into aquatic ecosystems during snowmelt
Bromine explosion
reduction and release of RGM during summer ?
no summer peak in Hg deeper in firn
28
Coincident DC-8 spiral down over Summit and
ARCIONS sonde launch, 5 April, 2008.
29
Coincident DC-8 spiral down over Summit and
ARCIONS sonde launch, 7 July, 2008.