Robert G. Ellingson and the ARESE II Science Team

1
ARESE II Description and Initial Results (ARM
Enhanced Shortwave Experiment)

• Robert G. Ellingson and the ARESE II Science Team
• Department of Meteorology
• University of Maryland
• College Park, MD

2
Motivation

• Knowledge of the amount and location of solar
  energy absorption is key to understanding the
  general circulation of the ocean and atmosphere
  and to our understanding and prediction of
  climate change.
• Measurements of the amount of solar radiation
  absorbed within clouds have yielded conflicting
  results. Many studies show much more absorption
  than can be explained by theory.
• If excess or enhanced absorption is true - we
  must
• reexamine our knowledge of the basic physics
• modify climate models, AND
• change remote sensing techniques.

3
The ARESE Experiments - Objectives

• Directly measure the absorption of solar
  radiation by the clear and cloudy atmosphere and
  
• investigate the causes of any absorption in
  excess of model predictions.

ARESE I - 25 September - 1 November 1995 ARESE
II - 15 February - 15 April 2000
4
ARESE I - A Thumbnail Sketch

• Used three aircraft platforms, as well as
  satellites and the ARM central and extended
  facilities in North Central Oklahoma
• Measured solar radiative fluxes at different
  altitudes and at the surface with spectral
  broadband, partial bandpass, and narrow bandpass
  filters
• Measurements obtained from aircraft flying in
  stacked formation over horizontal legs extending
  over several hundred kilometers

25 September - 1 November 1995
5
ARESE I - broadband absorptance increases with
cloud fraction
Courtesy of R. Cess - SUNY Stonybrook
6
ARESE-II Conducted During Feb -Apr 00

• Major Features
• Unique sampling strategy - single aircraft
  overflying SGP CART site on only overcast days
• Multiple independent instruments making same
  measurements with different technologies
  (aircraft and ground)
• Extensive pre- and post- experiment calibrations
• Long duration during a period of climatologically
  high frequency of extensive overcast ( 6 cases)
• Science Team with considerably different
  pre-experiment views

Simulation by A. Marshak
ARESE-II was coordinated with an ARM Cloud IOP -
insitu measurements of cloud microphysics
7
The ARESE-II Measurement Strategy Differed
Significantly From ARESE-I
Ingress

• Used single aircraft (Twin Otter) repeatedly
  overflying surface instruments
• Single aircraft reduced cost, makes long
  deployment possible
• ARESE-I showed thick stratus approach uniform
  case
• 2 years CART data 4-6 uniform stratus cases in
  6-week period
• Consistent with simulations by R. Cess and by A.
  Marshak

Central facility
Bluedata flight leg Reddata not valid
Continue on
One of 2 flight patterns (6 min revisit, 83 duty
cycle)
8
DHC-6 Twin-Otter

Photos courtesy of Tim Tooman
9
Conditional Sampling (theory) from Marshak et
al., 1999 On the Removal of the Effect of
Horizontal Fluxes in Two-Aircraft Measurements of
Cloud Absorption. Quart. J. Roy. Meteor. Soc.,
558, 2153-2170.
Courtesy of Alexander Marshak
10
Extensive Spectral and Broadband Calibrations
Were Performed Before and After ARESE II

• Spectral calibrations at Ponca City airport using
  lamps traceable to the ARM working standard
• Broadband calibrations at Blackwell-Tonkawa
  airport
• Broadband calibrations at SGP site
• Surface measurements at SGP 19 February through 6
  April 2000

11
ARESE II Broadband Calibration Facility at
Blackwell-Tonkawa Airport
Photos courtesy of Joe Michalsky PNNL/SUNY
12
Direct Measurement Uncertainty ? 3 W/m2
Diffuse Measurement Uncertainty ? 5 W/m2 But ...
Slide courtesy of Joe Michalsky PNNL/SUNY
13
The Twin Otter Payload Was Significantly Enhanced
Over ARESE I

• 3 sets of spectral and broadband nadir and zenith
  viewing radiometers
• Scripps, RAMS total solar broadband hemispheric
  (224-3910 nm) Valero
• Scripps, RAMS fractional solar broadband
  hemispheric (680-3300 nm)
• Scripps, RAMS total direct-diffuse hemispheric
  seven bands (495-505 400-450 450-500
  500-550 550-600 600-650 650-700 nm)
• NASA ARC SSFR (300-2500 nm in 300
  channels) Pilewskie
• CSU SSP2 (400-2500 nm in 100 channels) Stephen
  s
• MRI, CM21 broadband hemispheric (3350-2200
  nm) Asano
• SNL, CM22 broadband hemispheric (3350-2200
  nm) Tooman
• Cloud and meteorological measurements
• JPL/UMASS ACR nadir viewing radar Sekelsky
• BNL total temperature Tooman
• BNL static pressure
• BNL chilled mirror hygrometer

14
ARESE II Summary

• High quality data obtained on several clear and
  overcast days (03/03, 03/17, 03/18, 03/21, 03/29
  best ones)
• b1 data released by instrument PIs to ARESE II
  Science Team in Sept 2000 (some data in better
  state than others)
• ARESE II ST data discussion meeting 24-26 Oct
  2000
• Reprocessing with common calibration ? Nov-Dec
  2000
• ARESE II ST meeting 8-9 Feb 2001 - preliminary
  findings
• Data released to science community 17 March 2001
• Publication of ARESE II Science Team papers in
  progress

For additional information see the ARM UAV
Homepage http//armuav.atmos.colostate.edu/
15
Looking for the Right StuffMarch 29, 2000 - An
Excellent Example
16
NCEP Forecasts Are A Must!!!!
17
(No Transcript)
18
What Did We See From Space?
19
(No Transcript)
20
(No Transcript)
21
What Did We See From the Aircraft and Ground?
22
Diffuse Field Camera
23
(No Transcript)
24
Spectral Distribution of Fluxes From the SSFR
25
SSFR Upwelling Fluxes - 03/29/00
Wavelength (nm)
Time (hours)
26
Preliminary SSFR Data From 29 March 2000
1930 UTC
Data courtesy of Peter Pilewskie, NASA Ames
27
Broadband Fluxes
28
March 29, 2000
29
Absorptance
Defined as the layer absorption divided by the
downwelling solar flux at the top of layer
(aircraft level)
Slide courtesy of Tom Ackerman
30
Conditional Sampling(March 29)
Courtesy of Alexander Marshak
31
Analysis courtesy of Bob Cess using data from
10/00
32
Analysis courtesy of Bob Cess using data from
10/00
33
These data are from the five days for which
absorptance measurements from the CM22
radiometers and the TSBR radiometers can be
compared. There are two clear days, 0227 and
0320, with low absorptance values, and three
cloudy days with higher absorptances. Agreement
between the two types of radiometers is very
good. (A 3 difference in the upwelling at 7 km
on 0303 accounts for the offset in absorptance on
that day.) Conclusion The two different types
of radiometers yield the same measured
absorptance in both clear and cloudy conditions.
Results courtesy of Pope et al.
34
Preliminary Comparisons of Model Calculations
with Observations
35
Bars represent leg to leg variability
Results courtesy of Ackerman et al.
36
Ignore CM21 results shown here
Results courtesy of Ackerman et al.
37
Day averages of absorptance (from TSBR
measurements) show values of 0.10 to 0.12 for the
clear days and values of 0.20 to 0.23 for the
cloudy days. A standard model gives absorptance
values ranging from 0.10 for clear sky to 0.15
for cloudy sky (optical depth 60). Conclusion
observed cloudy-sky absorptances are
significantly greater than model predictions.
Results courtesy of Pope et al.
38
OHirok and Gautier, 2001
MWR
BLC
MMCR
MPL
RAMS
SMOS
CM21/22s
AOS
MFRSR
MFR
TDDR
SSP
SB3D
SSFR
MIE
SBMOD
aerosol
atm.
SHORTWAVE ABSORPTION COMPARISON
surface
cloud
model
? obs.
? obs.
39
OHirok and Gautier, 2001
40
03/29/2000
9
structure
km
0
120
optical thickness
? 55 re 7.5
60
0
within 2.5 km of cart site
1730

2130 UTC 0

75 km
OHirok and Gautier, 2001
41
March 03 2000
0.5
Model
Absorptance
RAMS
Cm22
0.4
0.3
/-10
0.2
/-5
0.1
0.0
visible
near-ir
total
OHirok and Gautier, 2001
42
March 21 2000
0.5
Model
Absorptance
RAMS
Cm22
0.4
/-10
/-5
0.3
0.2
0.1
0.0
near-ir
visible
total
OHirok and Gautier, 2001
43
March 29 2000
0.5
Model
Absorptance
RAMS
Cm22
0.4
/-10
/-5
0.3
0.2
0.1
0.0
visible
near-ir
total
OHirok and Gautier, 2001
44
March 29 2000 model sensitivity
0.06
RAMS
visible absorptance
Cm22
0.03
0.4
near-ir absorptance
0.3
0.25
total absorptance
0.15
ref rnd ipa re x2 drz ice x4
OHirok and Gautier, 2001
45
Summary conclusions to date

• Ackerman et al. - Differences between modeled and
  observed absorption on cloudy days are order 10
• Pope et al. - observed cloudy-sky absorptances
  are significantly greater than model predictions.
• OHirok and Gautier - major differences between
  observations and calculations are in the near IR,
  but total differences are within the order 10
  range.

• Common to all
• Observed absorption is greater than calculated
• Smaller absorption and smaller discrepancies
  than ARESE I

46

• Problems and Paths Forward
• Apparent disagreement between different models -
  use ICRCCM as an arbiter
• Causes of the discrepancies not yet identified -
  expanded use of the spectral data and extensive
  examination of all the data by the ARM Science
  Team and the general science community
• The data are there - Have at them!!!

47
ARESE-II has a broad-based Science Team

• Ackerman, Tom (PNNL) Marshak, Sasha (Univ
  Maryland)
• Asano, Shoji (Tohoku Univ) Michalsky, Joe (SUNY
  Albany)
• Cahalan, Bob (NASA GSFC) Minnis, Pat (NASA LRC)
• Cess, Bob (SUNY, Stony Brook) Sekelsky Steve
  (Univ Mass)
• Ellingson, Bob (Univ Maryland) Stephens, Graeme
  (CSU)
• Gautier, Catherine (UCSB) Tooman, Tim (SNL)
• Long, Chuck (PSU) Valero, Francisco (Scripps)
• Mace, Jay (Univ Utah) Vitko, John (SNL)
• Marchand, Roger (PNNL) Wiscombe, Warren (NASA
  GSFC)

Mission Scientist
48
Pre- and Post- ARESE II Boadband Calibrations Data
Pre-
Post-
courtesy of Joe Michalsky PNNL/SUNY
Post-
Pre-