The impact of cumulus congestus clouds on mid

METHODS
INTRODUCTION
Ozone (O3) is rapidly destroyed in the remote tropical marine
boundary layer [6, 13]. This ozone depleted marine boundary
layer air is injected into the free troposphere by convection, where
it is chemically produced at a rate of 1 - 2 ppbv per day [5]. This
rate of production gives rise to a chemical replacement timescale
which is comparable with the overturning timescale for the Hadley
circulation. Ozone is therefore a useful tracer for diagnosing
convective motions in the tropics.
Ozone measurements have been used to demonstrate the
existence of strong convective outflow in the upper troposphere
from deep convection [3, 4, 9, 12]. At present, however, the
evidence for the existence of preferred convective detrainment
near the melting level due to congestus (towering) clouds (Figure
1) arises mainly from measurements of cloud top heights [1, 17],
mass divergence [10, 11, 15], and relative humidity [7]. Here, we
present additional evidence for the existence of mid-level
cumulus congestus outflow using O3 measurements.
We also use the O3 simulation in a chemical transport model
(GEOS-Chem) to examine the representation of convective
outflow in a widely-used assimilated meteorological dataset
(GEOS-4 and GEOS-5).
Negative mid-tropospheric ozone
anomalies (>8 ppbv) are temporally
confined within 10-hour belt about peak
rainfall. The divergence anomaly
patterns indicate that the negative midlevel ozone anomaly preceding peak
rainfall can be attributed to outflow from
cumulus congestus clouds.
We also examined the ozone and
divergence anomaly patterns about
high rain events in the GEOS-4 and
GEOS-5 meteorology in the GEOSChem model.
In addition, we examined the
divergence anomaly pattern about high
rainfall events of a higher resolution
GEOS-5 YOTC meteorological dataset
[16] (0.25x0.33 deg.) at 12 locations in
the tropical Pacific.
Toni Mitovski,
Dalhousie University, Halifax, Canada
Email: [email protected]
Figure 1. The boundary layer air is detrained near the melting level
(congestus) and near the tropopause (deep convection).
Figure 2. Area of interest.
Data
availability
temporal
resolution
horizontal
resolution
TRMM 3B42
1998-2010
3-hrs
2.0°
SHADOZ
1998-2010
bi-weekly
2 stations
691
IGRA
1998-2010
12-hrs
2 arrays
>10,000
GEOS-4 &
GEOS-5
2004-2006
2D (3-hrs)
3D (6-hrs)
2.0°x2.5°
2009
2D (1-hr)
3D (3-hrs)
0.25°x0.33°
GEOS-5 YOTC
Figure 3. Rain frequency
distribution for various datasets
Figure 4. TRMM 3B42 rainfall
over Samoa for Oct. 2008. We
used all rain events above a
rain threshold of 1.5 mm/h
(red line). We stored only
those SHADOZ soundings
that had occurred within 24
hrs about the rainfall events.
RESULTS
a)
Table 1. Data and models used in this research.
Poster Design & Printing by Genigraphics® - 800.790.4001
The composite anomaly patterns at both Fiji and Samoa were constructed using the
following procedure:
  High rainfall events were defined as those which exceeded a threshold of 1.5 mm/hr;
  We then searched for SHADOZ soundings that had occurred within 24 hours of each
rain event (Figure 4). For the chosen threshold of 1.5 mm/hr, there were approximately
300 soundings available from both Fiji and Samoa;
  At both Fiji and Samoa, we defined seasonal mean ozone profiles (DJF, MAM, JJA,
SON) using the 12 years available (1998 – 2010). The appropriate seasonal mean was
then subtracted from each of the 300 ozonesondes within 24 hours of a rain event to
generate an instantaneous deviation of the ozone mixing ratio from the seasonal
climatology;
  Using the difference between the ozonesonde launch time and the time of each
rainfall event, the ozone deviations were grouped into 3 hour time bins. The anomaly
patterns for Fiji and Samoa were similar to each other and were therefore combined
into one shown in Figure 5(a).
Divergence anomaly patterns (Figure 6(a)) were constructed using simultaneous wind
measurements from two triangular arrays (Figure 2) of radiosondes and the same
procedure as described before with exception that here, we used an array averaged
TRMM rainfall. There were 805 divergence profiles within 24 hours of a rain event from
Array 1, and 1339 profiles within 24 hours of a rain event from Array 2.
DATA
The data and models used in
this research are summarized
in Table 1. We used data
from 2 SHADOZ locations
(Fiji and Samoa) and from 2
IGRA arrays to construct
ozone and divergence
anomaly patterns about high
rain events. Figure 2 shows
the SHADOZ and the IGRA
locations.
DISCUSSION
The selection of a rain event threshold for Fiji and Samoa requires making some
compromise between the number of rain events and their intensity. We wanted the rain
event threshold to be sufficiently large that the rain events would be associated with deep
convection, and give rise to significant deviations in ozone mixing ratios from background.
On the other hand, the rainfall event threshold had to be sufficiently small that a statistically
representative number of ozonesonde profiles would be available to construct the ozone
anomaly patterns. Figure 3 shows the frequency distribution of the rainfall using a variety of
data. Due to more frequent light precipitation in the models than in the TRMM, we used a
smaller rain threshold of 1.0 mm/h, rather than 1.5 mm/h.
c)
b)
a)
b)
d)
c)
d)
number of
soundings
Figure 5. Ozone anomalies about peak rainfall (t=0) calculated
using: a) SHADOZ, and GEOS Chem with b) GEOS-4 and d)
GEOS-5 meteorology. Panel c) shows the strength of the signal
and the significance (within the boxes) of the ozone anomalies in
panel a). The rain variability about t=0 is plotted under the panels.
Figure 6. Divergence anomalies about peak rainfall (time=0)
calculated using: a) radiosonde arrays, b) GEOS-4, c) GEOS 5
(YOTC), d) GEOS-5 meteorology. The rain variability about t=0
is plotted under the panels.
In SHADOZ (Figure 5(a)), negative ozone anomalies start to
develop at mid-levels (4-8 km) about 16 hours prior to maximum
rainfall. The mid-level ozone anomaly intensifies as peak rainfall
approaches. At peak rainfall (t = 0), the ozone anomaly extends from 4
km to 15 km.
The divergence pattern (Figure 6(a)) exhibit convergence near
the surface prior to peak rainfall, an upper level divergence near 13 km
associated with deep convection, and an antisymmetric divergence
dipole near 6 km. The divergence dipole consists of a congestus
divergence roughly 8 hours prior to peak rainfall, and a stratiform
convergence 8 hours after peak rainfall.
With GEOS-4 meteorology (Figure 5(b)), convection gives rise to
two broad negative ozone anomalies, centered at 6 km and at 15 km.
The anomalies are more persistent and are about 5-fold weaker than
the observations, extending from more than 24h prior to peak rainfall,
to more than 24h after peak rainfall. A broad upper tropospheric
divergence is about 3-fold weaker than in the observations (Figure 6
(b)).
With GEOS-5 meteorology, the ozone (Figure 5(d)) and
divergence (Figure 6(d)) anomaly patterns are almost identical as with
GEOS-4. However, the ozone anomalies and the upper level
divergence with GEOS-5 are stronger than with GEOS-4, presumably
due to stronger rain at t=0.
In GEOS-5 YOTC (Table 1), overall structure of the YOTC
divergence anomaly pattern (Figure 6(c)) is quite similar to the IGRA
divergence anomaly pattern. However, there is a vertical offset
between the congestus and stratiform divergence features that is not
observed, and both features occur too close to peak rainfall.
CONCLUSIONS
1.  The negative mid-level ozone anomalies that develop prior to peak
rainfall observed in the SHADOZ dataset are likely to be a result of
the vertical transport of ozone depleted boundary layer air within
congestus clouds, followed by detrainment (divergence) at mid-levels.
2.  Two negative ozone anomaly features in GEOS Chem are much
broader than observed, likely due to lower rainfall variability (about
t=0) in GEOS-4 and GEOS-5. Quantitatively, the anomalies are
comparable to the observations in GEOS-5, and are 5-fold weaker in
GEOS-4.
3.  In GEOS-4 and GEOS-5 meteorology, including GEOS-5 (YOTC),
there is no evidence of a congestus outflow mode that precedes deep
outflow, as occurs in the observations.
REFERENCES
1. Dessler, A. E., S. P. Palm, and J. D. Spinhirne (2006), Tropical cloud-top height distributions revealed by the Ice, Cloud, and Land Elevation Satellite (ICESat)/
Geoscience Laser Altimeter System (GLAS), J. Geophys. Res., 111, D12215, doi:10.1029/2005JD006705.
2. Durre I., R. S. Vose and D. B. Wuertz (2006), Overview of the Integrated Global Radiosonde Archive, J. Climate, Vol. 19, No. 1, pp. 53-68.
3. Folkins, I., C. Braun, A. M. Thompson, and J. C. Witte (2002),Tropical ozone as an indicator of deep convection, J. Geophys. Res., 107(D13), 4184, doi:
10.1029/2001JD001178.
4. Folkins, I., et al. (2006), Testing convective parameterizations with tropical measurements of HNO3, CO, H2O, and O3: Implications for the water vapor budget, J.
Geophys. Res., 111, D23304, doi: 10.1029/2006JD007325.
5. Jacob, D. J., et al. (1996), Origin of ozone and NOx in the tropical troposphere: A photochemical analysis of aircraft observations over the south Atlantic basin, J.
Geophys. Res., 101, 24,235–24,250.
6. Johnson, J. E., et al. (1990), Ozone in the marine boundary layer over the Pacific and Indian Oceans: Latitudinal gradients and diurnal cycles, J. Geophys. Res., 95,
11847-11856.
7. Johnson, R.H., et al. (1999), Trimodal characteristics of tropical convection. J. Climate, 12, 2397-2418.
8. Kummerow, C., W. Barnes, T. Kozu, J. Shiue, and J. Simpson, 1998: The Tropical Rainfall Measuring Mission (TRMM) Sensor Package. J. Atmos. Oceanic Technol.,
15(3), 809-817.
9. Lawrence, M. G., von Kuhlmann, R., Salzmann, M., and Rasch, P. J. (2003), The balance of effects of deep convective mixing on tropospheric ozone, Geophys. Res.
Lett., 30(18), 1940, doi:10.1029.2003GL017644.
10. Mapes, B. E., S. Tulich, J. Lin, and P. Zuidema (2006), The mesoscale convection life cycle: Building block or prototype for large-scale tropical waves?, Dyn. Atmos.
Oceans, 42, 3-29.
11. Mitovski, T., Folkins, I., von Salzen, K., and Sigmond, M. (2010), Temperature, relative humidity, and divergence response to high rainfall events in the Tropics:
observations and models, J. Climate, 23, 3613 - 3625, 2010.
12. Pickering, K. E., Thompson, A. M., Tao, W.-K., and Kucsera, T. L. (1993), Upper tropospheric ozone production following mesoscale convection during STEP/EMEX,
J. Geophys. Res., 98, 8737–8749.
13. Thompson, A. M., et al. (1993), SAGA-3 ozone observations and a photochemical model analysis of the marine boundary layer during SAGA-3, J. Geophys. Res.,
98, 16955-16968.
14. Thompson, A. M., et al. (2003), Southern Hemisphere Additional Ozonesondes (SHADOZ) 1998-2000 tropical ozone climatology 2. Tropospheric variability and the
zonal wave-one, J. Geophys. Res., Vol. 108 No. D2,8241, doi: 10.1029/2002JD002241.
15. Thompson, R. M., Jr., S. W. Payne, E. E. Recker, and R. J. Reed (1979), Structure and properties of synoptic-scale wave disturbances in the intertropical
convergence zone of the eastern Atlantic, J. Atmos. Sci., 36, 53-72.
16. Waliser, D. E., and M. Moncrieff (2008), The Year of Tropical Convection (YOTC) Science Plan: A joint WCRP - W WRP/THORPEX International Initiative. WMO/TD
No. 1452, WCRP - 130, WWRP/THORPEX - No 9. WMO, Geneva, Switzerland.
17. Zuidema, P. (1998), The 600 – 800 mb minimum in tropical cloudiness observed during TOGA COARE, J. Atmos. Sci., 55, 2220 – 2228.