Crosscomparison of AIRS Cloud Products with ARM and A-train - - PowerPoint PPT Presentation

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Cross–comparison of AIRS Cloud Products with ARM and A-train Measurements

by Brian Kahn1, Amy Braverman1, Annmarie Eldering1, Eric Fetzer1, Evan Fishbein1, Michael Garay1,2, Jonathan Jiang1, Sung-Yung Lee1, and Shaima Nasiri3

1Jet Propulsion Laboratory, Pasadena, CA, USA 2Department of Atmospheric and Oceanic Sciences, UCLA, Los Angeles, CA, USA 3Department of Atmospheric Sciences, Texas A&M University, College Station, TX, USA

Cloud pictures courtesy of australiansevereweather.com

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology AIRS Science Team Meeting, March 7–9, 2006

Why do we care about cirrus clouds?

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology AIRS Science Team Meeting, March 7–9, 2006

Why do we care about cirrus clouds?

• They are fun to look at

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology AIRS Science Team Meeting, March 7–9, 2006

Why do we care about cirrus clouds?

• They are fun to look at
• Help set Earth’s radiative balance

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology AIRS Science Team Meeting, March 7–9, 2006

Why do we care about cirrus clouds?

• They are fun to look at
• Help set Earth’s radiative balance
• Integral part of hydrological cycle

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology AIRS Science Team Meeting, March 7–9, 2006

Why do we care about cirrus clouds?

• They are fun to look at
• Help set Earth’s radiative balance
• Integral part of hydrological cycle
• Feedbacks between radiation, dynamics, and thermodynamics

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology AIRS Science Team Meeting, March 7–9, 2006

Why do we care about cirrus clouds?

• They are fun to look at
• Help set Earth’s radiative balance
• Integral part of hydrological cycle
• Feedbacks between radiation, dynamics, and thermodynamics
• Indirect effects

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology AIRS Science Team Meeting, March 7–9, 2006

Why do we care about cirrus clouds?

• They are fun to look at
• Help set Earth’s radiative balance
• Integral part of hydrological cycle
• Feedbacks between radiation, dynamics, and thermodynamics
• Indirect effects
• Responses to anthropogenic climate change?

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology AIRS Science Team Meeting, March 7–9, 2006

Outline

• How valid are the AIRS V4 cloud fields?
• Focus on upper level CTP
• ARM TWP mm-wave cloud radar (Manus Island) and micropulse lidar (Nauru Island)
• AIRS is sensitive (statistically significant) to thin (and thick) cirrus
• AIRS CTP and Microwave Limb Sounder (MLS) IWC comparisons
• PDFs of AIRS and MODIS agree well…
• …but statistics conditional on MLS level, IWC threshold, AIRS ECF, etc.
• AIRS and MODIS: a “holistic” view
• Use CTP, ECF and Ts to explore consistency in retrievals
• Good agreement for high and opaque clouds
• Some issues within multilayer clouds and cloud edges
• Where to go from here?

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Checking the cloud top height between AIRS and Atmospheric Radiation Measurement (ARM) program observations

0.4 0.3 0.2 0.1 0.0 Hollars 24 min AVG Manus 24 min HIST Manus 24 min MAX Manus 0.4 0.3 0.2 0.1 0.0

• 6
• 4
• 2

2 4 6 Active – Passive Z

CLD

Hawkinson SFOV Hawkinson 3x3 FOV 90 min HIST Nauru Frey

Frequency histogram of the agreement between an active and passive-derived ZCLD obtained from several independent data sources. We compare ARM–AIRS to: Top: ground-based MMCR with GMS-5 (Hollars et al., 2004) Bottom: aircraft lidar and the MODIS Airborne Simulator ZCLD (Frey et al., 1999), ground-based lidar+radar and GOES ZCLD (Hawkinson et al., 2005), and ground-based lidar and AIRS ZCLD.

Kahn et al., 2006a

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Kahn et al., 2006a

Location/Time Time (min) Height Method

• 0. f < .05

.05 f < .15 .15 f < .5 .5 f < .85 .85 f < 1.0 Manus/Night – – N=13 N=9 N=21 N=16 N=16 54 AVG 7.2 ± 7.0 2.1 ± 3.4 0.4 ± 3.7 –0.1 ± 1.5 0.7 ± 1.8 126 AVG 7.1 ± 6.5 1.8 ± 3.2 0.5 ± 3.6 –0.3 ± 1.2 0.7 ± 2.0 186 AVG 7.0 ± 6.5 1.9 ± 3.0 0.4 ± 3.6 –0.4 ± 1.3 0.6 ± 2.0 54 HIST 7.1 ± 7.3 1.1 ± 5.1 –0.9 ± 3.4 –0.5 ± 1.3 –0.1 ± 1.7 126 HIST 4.9 ± 7.4 –0.5 ± 4.5 –0.9 ± 3.4 –1.2 ± 1.0 –0.3 ± 2.0 186 HIST 4.7 ± 7.5 –0.4 ± 4.1 –1.0 ± 3.3 –1.2 ± 1.0 –0.2 ± 2.0 54 MAX 5.3 ± 8.4 0.6 ± 4.9 –2.2 ± 4.0 –1.4 ± 1.3 –0.8 ± 1.9 Manus/Day _ _ N=21 N=12 N=16 N=12 N=16 54 AVG 7.6 ± 5.6 6.3 ± 5.8 1.2 ± 4.2 0.2 ± 2.3 1.1 ± 1.6 126 AVG 7.8 ± 5.6 4.5 ± 4.9 1.3 ± 3.9 0.5 ± 2.3 1.3 ± 1.6 186 AVG 9.0 ± 5.0 4.4 ± 4.7 1.5 ± 3.8 0.7 ± 2.4 1.6 ± 1.7 54 HIST 6.4 ± 8.8 5.4 ± 6.1 –0.4 ± 3.7 –0.1 ± 2.7 0.5 ± 1.6 126 HIST 3.7 ± 9.5 –1.0 ± 8.3 –0.7 ± 3.8 –1.1 ± 2.1 0.4 ± 1.6 186 HIST 1.5 ± 7.8 –1.5 ± 8.5 –0.8 ± 3.8 –1.1 ± 2.1 0.4 ± 1.5 54 MAX 4.8 ± 8.3 3.1 ± 8.1 –0.7 ± 3.8 –1.5 ± 1.7 –0.2 ± 1.4 Nauru/Night _ _ N=32 N=20 _ _ _ 54 AVG 8.2 ± 6.1 2.1 ± 3.9 _ _ _ 126 AVG 7.1 ± 6.1 1.9 ± 3.2 _ _ _ 186 AVG 6.3 ± 5.4 1.9 ± 3.0 _ _ _ 54 HIST 7.4 ± 7.3 0.3 ± 4.1 _ _ _ 126 HIST 5.3 ± 7.8 –0.7 ± 3.7 _ _ _ 186 HIST 3.0 ± 7.3 –1.1 ± 3.1 _ _ _ 54 MAX 7.0 ± 7.5 –0.5 ± 4.5 _ _ _

AIRS Science Team Meeting, March 7–9, 2006

Radar at night Radar at day Lidar at night

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Kahn et al., 2006a

Location/Time Time (min) Height Method

• 0. f < .05

.05 f < .15 .15 f < .5 .5 f < .85 .85 f < 1.0 Manus/Night – – N=13 N=9 N=21 N=16 N=16 54 AVG 7.2 ± 7.0 2.1 ± 3.4 0.4 ± 3.7 –0.1 ± 1.5 0.7 ± 1.8 126 AVG 7.1 ± 6.5 1.8 ± 3.2 0.5 ± 3.6 –0.3 ± 1.2 0.7 ± 2.0 186 AVG 7.0 ± 6.5 1.9 ± 3.0 0.4 ± 3.6 –0.4 ± 1.3 0.6 ± 2.0 54 HIST 7.1 ± 7.3 1.1 ± 5.1 –0.9 ± 3.4 –0.5 ± 1.3 –0.1 ± 1.7 126 HIST 4.9 ± 7.4 –0.5 ± 4.5 –0.9 ± 3.4 –1.2 ± 1.0 –0.3 ± 2.0 186 HIST 4.7 ± 7.5 –0.4 ± 4.1 –1.0 ± 3.3 –1.2 ± 1.0 –0.2 ± 2.0 54 MAX 5.3 ± 8.4 0.6 ± 4.9 –2.2 ± 4.0 –1.4 ± 1.3 –0.8 ± 1.9 Manus/Day _ _ N=21 N=12 N=16 N=12 N=16 54 AVG 7.6 ± 5.6 6.3 ± 5.8 1.2 ± 4.2 0.2 ± 2.3 1.1 ± 1.6 126 AVG 7.8 ± 5.6 4.5 ± 4.9 1.3 ± 3.9 0.5 ± 2.3 1.3 ± 1.6 186 AVG 9.0 ± 5.0 4.4 ± 4.7 1.5 ± 3.8 0.7 ± 2.4 1.6 ± 1.7 54 HIST 6.4 ± 8.8 5.4 ± 6.1 –0.4 ± 3.7 –0.1 ± 2.7 0.5 ± 1.6 126 HIST 3.7 ± 9.5 –1.0 ± 8.3 –0.7 ± 3.8 –1.1 ± 2.1 0.4 ± 1.6 186 HIST 1.5 ± 7.8 –1.5 ± 8.5 –0.8 ± 3.8 –1.1 ± 2.1 0.4 ± 1.5 54 MAX 4.8 ± 8.3 3.1 ± 8.1 –0.7 ± 3.8 –1.5 ± 1.7 –0.2 ± 1.4 Nauru/Night _ _ N=32 N=20 _ _ _ 54 AVG 8.2 ± 6.1 2.1 ± 3.9 _ _ _ 126 AVG 7.1 ± 6.1 1.9 ± 3.2 _ _ _ 186 AVG 6.3 ± 5.4 1.9 ± 3.0 _ _ _ 54 HIST 7.4 ± 7.3 0.3 ± 4.1 _ _ _ 126 HIST 5.3 ± 7.8 –0.7 ± 3.7 _ _ _ 186 HIST 3.0 ± 7.3 –1.1 ± 3.1 _ _ _ 54 MAX 7.0 ± 7.5 –0.5 ± 4.5 _ _ _

Three time averages

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Kahn et al., 2006a

Location/Time Time (min) Height Method

• 0. f < .05

.05 f < .15 .15 f < .5 .5 f < .85 .85 f < 1.0 Manus/Night – – N=13 N=9 N=21 N=16 N=16 54 AVG 7.2 ± 7.0 2.1 ± 3.4 0.4 ± 3.7 –0.1 ± 1.5 0.7 ± 1.8 126 AVG 7.1 ± 6.5 1.8 ± 3.2 0.5 ± 3.6 –0.3 ± 1.2 0.7 ± 2.0 186 AVG 7.0 ± 6.5 1.9 ± 3.0 0.4 ± 3.6 –0.4 ± 1.3 0.6 ± 2.0 54 HIST 7.1 ± 7.3 1.1 ± 5.1 –0.9 ± 3.4 –0.5 ± 1.3 –0.1 ± 1.7 126 HIST 4.9 ± 7.4 –0.5 ± 4.5 –0.9 ± 3.4 –1.2 ± 1.0 –0.3 ± 2.0 186 HIST 4.7 ± 7.5 –0.4 ± 4.1 –1.0 ± 3.3 –1.2 ± 1.0 –0.2 ± 2.0 54 MAX 5.3 ± 8.4 0.6 ± 4.9 –2.2 ± 4.0 –1.4 ± 1.3 –0.8 ± 1.9 Manus/Day _ _ N=21 N=12 N=16 N=12 N=16 54 AVG 7.6 ± 5.6 6.3 ± 5.8 1.2 ± 4.2 0.2 ± 2.3 1.1 ± 1.6 126 AVG 7.8 ± 5.6 4.5 ± 4.9 1.3 ± 3.9 0.5 ± 2.3 1.3 ± 1.6 186 AVG 9.0 ± 5.0 4.4 ± 4.7 1.5 ± 3.8 0.7 ± 2.4 1.6 ± 1.7 54 HIST 6.4 ± 8.8 5.4 ± 6.1 –0.4 ± 3.7 –0.1 ± 2.7 0.5 ± 1.6 126 HIST 3.7 ± 9.5 –1.0 ± 8.3 –0.7 ± 3.8 –1.1 ± 2.1 0.4 ± 1.6 186 HIST 1.5 ± 7.8 –1.5 ± 8.5 –0.8 ± 3.8 –1.1 ± 2.1 0.4 ± 1.5 54 MAX 4.8 ± 8.3 3.1 ± 8.1 –0.7 ± 3.8 –1.5 ± 1.7 –0.2 ± 1.4 Nauru/Night _ _ N=32 N=20 _ _ _ 54 AVG 8.2 ± 6.1 2.1 ± 3.9 _ _ _ 126 AVG 7.1 ± 6.1 1.9 ± 3.2 _ _ _ 186 AVG 6.3 ± 5.4 1.9 ± 3.0 _ _ _ 54 HIST 7.4 ± 7.3 0.3 ± 4.1 _ _ _ 126 HIST 5.3 ± 7.8 –0.7 ± 3.7 _ _ _ 186 HIST 3.0 ± 7.3 –1.1 ± 3.1 _ _ _ 54 MAX 7.0 ± 7.5 –0.5 ± 4.5 _ _ _

AIRS Science Team Meeting, March 7–9, 2006

Three ARM ZCLD averages

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Kahn et al., 2006a

Location/Time Time (min) Height Method

• 0. f < .05

.05 f < .15 .15 f < .5 .5 f < .85 .85 f < 1.0 Manus/Night – – N=13 N=9 N=21 N=16 N=16 54 AVG 7.2 ± 7.0 2.1 ± 3.4 0.4 ± 3.7 –0.1 ± 1.5 0.7 ± 1.8 126 AVG 7.1 ± 6.5 1.8 ± 3.2 0.5 ± 3.6 –0.3 ± 1.2 0.7 ± 2.0 186 AVG 7.0 ± 6.5 1.9 ± 3.0 0.4 ± 3.6 –0.4 ± 1.3 0.6 ± 2.0 54 HIST 7.1 ± 7.3 1.1 ± 5.1 –0.9 ± 3.4 –0.5 ± 1.3 –0.1 ± 1.7 126 HIST 4.9 ± 7.4 –0.5 ± 4.5 –0.9 ± 3.4 –1.2 ± 1.0 –0.3 ± 2.0 186 HIST 4.7 ± 7.5 –0.4 ± 4.1 –1.0 ± 3.3 –1.2 ± 1.0 –0.2 ± 2.0 54 MAX 5.3 ± 8.4 0.6 ± 4.9 –2.2 ± 4.0 –1.4 ± 1.3 –0.8 ± 1.9 Manus/Day _ _ N=21 N=12 N=16 N=12 N=16 54 AVG 7.6 ± 5.6 6.3 ± 5.8 1.2 ± 4.2 0.2 ± 2.3 1.1 ± 1.6 126 AVG 7.8 ± 5.6 4.5 ± 4.9 1.3 ± 3.9 0.5 ± 2.3 1.3 ± 1.6 186 AVG 9.0 ± 5.0 4.4 ± 4.7 1.5 ± 3.8 0.7 ± 2.4 1.6 ± 1.7 54 HIST 6.4 ± 8.8 5.4 ± 6.1 –0.4 ± 3.7 –0.1 ± 2.7 0.5 ± 1.6 126 HIST 3.7 ± 9.5 –1.0 ± 8.3 –0.7 ± 3.8 –1.1 ± 2.1 0.4 ± 1.6 186 HIST 1.5 ± 7.8 –1.5 ± 8.5 –0.8 ± 3.8 –1.1 ± 2.1 0.4 ± 1.5 54 MAX 4.8 ± 8.3 3.1 ± 8.1 –0.7 ± 3.8 –1.5 ± 1.7 –0.2 ± 1.4 Nauru/Night _ _ N=32 N=20 _ _ _ 54 AVG 8.2 ± 6.1 2.1 ± 3.9 _ _ _ 126 AVG 7.1 ± 6.1 1.9 ± 3.2 _ _ _ 186 AVG 6.3 ± 5.4 1.9 ± 3.0 _ _ _ 54 HIST 7.4 ± 7.3 0.3 ± 4.1 _ _ _ 126 HIST 5.3 ± 7.8 –0.7 ± 3.7 _ _ _ 186 HIST 3.0 ± 7.3 –1.1 ± 3.1 _ _ _ 54 MAX 7.0 ± 7.5 –0.5 ± 4.5 _ _ _

AIRS Science Team Meeting, March 7–9, 2006

Five ECF bins

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Kahn et al., 2006a

Location/Time Time (min) Height Method

• 0. f < .05

.05 f < .15 .15 f < .5 .5 f < .85 .85 f < 1.0 Manus/Night – – N=13 N=9 N=21 N=16 N=16 54 AVG 7.2 ± 7.0 2.1 ± 3.4 0.4 ± 3.7 –0.1 ± 1.5 0.7 ± 1.8 126 AVG 7.1 ± 6.5 1.8 ± 3.2 0.5 ± 3.6 –0.3 ± 1.2 0.7 ± 2.0 186 AVG 7.0 ± 6.5 1.9 ± 3.0 0.4 ± 3.6 –0.4 ± 1.3 0.6 ± 2.0 54 HIST 7.1 ± 7.3 1.1 ± 5.1 –0.9 ± 3.4 –0.5 ± 1.3 –0.1 ± 1.7 126 HIST 4.9 ± 7.4 –0.5 ± 4.5 –0.9 ± 3.4 –1.2 ± 1.0 –0.3 ± 2.0 186 HIST 4.7 ± 7.5 –0.4 ± 4.1 –1.0 ± 3.3 –1.2 ± 1.0 –0.2 ± 2.0 54 MAX 5.3 ± 8.4 0.6 ± 4.9 –2.2 ± 4.0 –1.4 ± 1.3 –0.8 ± 1.9 Manus/Day _ _ N=21 N=12 N=16 N=12 N=16 54 AVG 7.6 ± 5.6 6.3 ± 5.8 1.2 ± 4.2 0.2 ± 2.3 1.1 ± 1.6 126 AVG 7.8 ± 5.6 4.5 ± 4.9 1.3 ± 3.9 0.5 ± 2.3 1.3 ± 1.6 186 AVG 9.0 ± 5.0 4.4 ± 4.7 1.5 ± 3.8 0.7 ± 2.4 1.6 ± 1.7 54 HIST 6.4 ± 8.8 5.4 ± 6.1 –0.4 ± 3.7 –0.1 ± 2.7 0.5 ± 1.6 126 HIST 3.7 ± 9.5 –1.0 ± 8.3 –0.7 ± 3.8 –1.1 ± 2.1 0.4 ± 1.6 186 HIST 1.5 ± 7.8 –1.5 ± 8.5 –0.8 ± 3.8 –1.1 ± 2.1 0.4 ± 1.5 54 MAX 4.8 ± 8.3 3.1 ± 8.1 –0.7 ± 3.8 –1.5 ± 1.7 –0.2 ± 1.4 Nauru/Night _ _ N=32 N=20 _ _ _ 54 AVG 8.2 ± 6.1 2.1 ± 3.9 _ _ _ 126 AVG 7.1 ± 6.1 1.9 ± 3.2 _ _ _ 186 AVG 6.3 ± 5.4 1.9 ± 3.0 _ _ _ 54 HIST 7.4 ± 7.3 0.3 ± 4.1 _ _ _ 126 HIST 5.3 ± 7.8 –0.7 ± 3.7 _ _ _ 186 HIST 3.0 ± 7.3 –1.1 ± 3.1 _ _ _ 54 MAX 7.0 ± 7.5 –0.5 ± 4.5 _ _ _

AIRS Science Team Meeting, March 7–9, 2006

# of samples

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Kahn et al., 2006a

Location/Time Time (min) Height Method

• 0. f < .05

.05 f < .15 .15 f < .5 .5 f < .85 .85 f < 1.0 Manus/Night – – N=13 N=9 N=21 N=16 N=16 54 AVG 7.2 ± 7.0 2.1 ± 3.4 0.4 ± 3.7 –0.1 ± 1.5 0.7 ± 1.8 126 AVG 7.1 ± 6.5 1.8 ± 3.2 0.5 ± 3.6 –0.3 ± 1.2 0.7 ± 2.0 186 AVG 7.0 ± 6.5 1.9 ± 3.0 0.4 ± 3.6 –0.4 ± 1.3 0.6 ± 2.0 54 HIST 7.1 ± 7.3 1.1 ± 5.1 –0.9 ± 3.4 –0.5 ± 1.3 –0.1 ± 1.7 126 HIST 4.9 ± 7.4 –0.5 ± 4.5 –0.9 ± 3.4 –1.2 ± 1.0 –0.3 ± 2.0 186 HIST 4.7 ± 7.5 –0.4 ± 4.1 –1.0 ± 3.3 –1.2 ± 1.0 –0.2 ± 2.0 54 MAX 5.3 ± 8.4 0.6 ± 4.9 –2.2 ± 4.0 –1.4 ± 1.3 –0.8 ± 1.9 Manus/Day _ _ N=21 N=12 N=16 N=12 N=16 54 AVG 7.6 ± 5.6 6.3 ± 5.8 1.2 ± 4.2 0.2 ± 2.3 1.1 ± 1.6 126 AVG 7.8 ± 5.6 4.5 ± 4.9 1.3 ± 3.9 0.5 ± 2.3 1.3 ± 1.6 186 AVG 9.0 ± 5.0 4.4 ± 4.7 1.5 ± 3.8 0.7 ± 2.4 1.6 ± 1.7 54 HIST 6.4 ± 8.8 5.4 ± 6.1 –0.4 ± 3.7 –0.1 ± 2.7 0.5 ± 1.6 126 HIST 3.7 ± 9.5 –1.0 ± 8.3 –0.7 ± 3.8 –1.1 ± 2.1 0.4 ± 1.6 186 HIST 1.5 ± 7.8 –1.5 ± 8.5 –0.8 ± 3.8 –1.1 ± 2.1 0.4 ± 1.5 54 MAX 4.8 ± 8.3 3.1 ± 8.1 –0.7 ± 3.8 –1.5 ± 1.7 –0.2 ± 1.4 Nauru/Night _ _ N=32 N=20 _ _ _ 54 AVG 8.2 ± 6.1 2.1 ± 3.9 _ _ _ 126 AVG 7.1 ± 6.1 1.9 ± 3.2 _ _ _ 186 AVG 6.3 ± 5.4 1.9 ± 3.0 _ _ _ 54 HIST 7.4 ± 7.3 0.3 ± 4.1 _ _ _ 126 HIST 5.3 ± 7.8 –0.7 ± 3.7 _ _ _ 186 HIST 3.0 ± 7.3 –1.1 ± 3.1 _ _ _ 54 MAX 7.0 ± 7.5 –0.5 ± 4.5 _ _ _

AIRS Science Team Meeting, March 7–9, 2006

AIRS–ARM ± 1-σ (km) Bold: significant @ 5% Italic: significant @ 1%

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Kahn et al., 2006a

Location/Time Time (min) Height Method

• 0. f < .05

.05 f < .15 .15 f < .5 .5 f < .85 .85 f < 1.0 Manus/Night – – N=13 N=9 N=21 N=16 N=16 54 AVG 7.2 ± 7.0 2.1 ± 3.4 0.4 ± 3.7 –0.1 ± 1.5 0.7 ± 1.8 126 AVG 7.1 ± 6.5 1.8 ± 3.2 0.5 ± 3.6 –0.3 ± 1.2 0.7 ± 2.0 186 AVG 7.0 ± 6.5 1.9 ± 3.0 0.4 ± 3.6 –0.4 ± 1.3 0.6 ± 2.0 54 HIST 7.1 ± 7.3 1.1 ± 5.1 –0.9 ± 3.4 –0.5 ± 1.3 –0.1 ± 1.7 126 HIST 4.9 ± 7.4 –0.5 ± 4.5 –0.9 ± 3.4 –1.2 ± 1.0 –0.3 ± 2.0 186 HIST 4.7 ± 7.5 –0.4 ± 4.1 –1.0 ± 3.3 –1.2 ± 1.0 –0.2 ± 2.0 54 MAX 5.3 ± 8.4 0.6 ± 4.9 –2.2 ± 4.0 –1.4 ± 1.3 –0.8 ± 1.9 Manus/Day _ _ N=21 N=12 N=16 N=12 N=16 54 AVG 7.6 ± 5.6 6.3 ± 5.8 1.2 ± 4.2 0.2 ± 2.3 1.1 ± 1.6 126 AVG 7.8 ± 5.6 4.5 ± 4.9 1.3 ± 3.9 0.5 ± 2.3 1.3 ± 1.6 186 AVG 9.0 ± 5.0 4.4 ± 4.7 1.5 ± 3.8 0.7 ± 2.4 1.6 ± 1.7 54 HIST 6.4 ± 8.8 5.4 ± 6.1 –0.4 ± 3.7 –0.1 ± 2.7 0.5 ± 1.6 126 HIST 3.7 ± 9.5 –1.0 ± 8.3 –0.7 ± 3.8 –1.1 ± 2.1 0.4 ± 1.6 186 HIST 1.5 ± 7.8 –1.5 ± 8.5 –0.8 ± 3.8 –1.1 ± 2.1 0.4 ± 1.5 54 MAX 4.8 ± 8.3 3.1 ± 8.1 –0.7 ± 3.8 –1.5 ± 1.7 –0.2 ± 1.4 Nauru/Night _ _ N=32 N=20 _ _ _ 54 AVG 8.2 ± 6.1 2.1 ± 3.9 _ _ _ 126 AVG 7.1 ± 6.1 1.9 ± 3.2 _ _ _ 186 AVG 6.3 ± 5.4 1.9 ± 3.0 _ _ _ 54 HIST 7.4 ± 7.3 0.3 ± 4.1 _ _ _ 126 HIST 5.3 ± 7.8 –0.7 ± 3.7 _ _ _ 186 HIST 3.0 ± 7.3 –1.1 ± 3.1 _ _ _ 54 MAX 7.0 ± 7.5 –0.5 ± 4.5 _ _ _

AIRS Science Team Meeting, March 7–9, 2006

Some day/night variation – slightly worse during day

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Kahn et al., 2006a

Location/Time Time (min) Height Method

• 0. f < .05

.05 f < .15 .15 f < .5 .5 f < .85 .85 f < 1.0 Manus/Night – – N=13 N=9 N=21 N=16 N=16 54 AVG 7.2 ± 7.0 2.1 ± 3.4 0.4 ± 3.7 –0.1 ± 1.5 0.7 ± 1.8 126 AVG 7.1 ± 6.5 1.8 ± 3.2 0.5 ± 3.6 –0.3 ± 1.2 0.7 ± 2.0 186 AVG 7.0 ± 6.5 1.9 ± 3.0 0.4 ± 3.6 –0.4 ± 1.3 0.6 ± 2.0 54 HIST 7.1 ± 7.3 1.1 ± 5.1 –0.9 ± 3.4 –0.5 ± 1.3 –0.1 ± 1.7 126 HIST 4.9 ± 7.4 –0.5 ± 4.5 –0.9 ± 3.4 –1.2 ± 1.0 –0.3 ± 2.0 186 HIST 4.7 ± 7.5 –0.4 ± 4.1 –1.0 ± 3.3 –1.2 ± 1.0 –0.2 ± 2.0 54 MAX 5.3 ± 8.4 0.6 ± 4.9 –2.2 ± 4.0 –1.4 ± 1.3 –0.8 ± 1.9 Manus/Day _ _ N=21 N=12 N=16 N=12 N=16 54 AVG 7.6 ± 5.6 6.3 ± 5.8 1.2 ± 4.2 0.2 ± 2.3 1.1 ± 1.6 126 AVG 7.8 ± 5.6 4.5 ± 4.9 1.3 ± 3.9 0.5 ± 2.3 1.3 ± 1.6 186 AVG 9.0 ± 5.0 4.4 ± 4.7 1.5 ± 3.8 0.7 ± 2.4 1.6 ± 1.7 54 HIST 6.4 ± 8.8 5.4 ± 6.1 –0.4 ± 3.7 –0.1 ± 2.7 0.5 ± 1.6 126 HIST 3.7 ± 9.5 –1.0 ± 8.3 –0.7 ± 3.8 –1.1 ± 2.1 0.4 ± 1.6 186 HIST 1.5 ± 7.8 –1.5 ± 8.5 –0.8 ± 3.8 –1.1 ± 2.1 0.4 ± 1.5 54 MAX 4.8 ± 8.3 3.1 ± 8.1 –0.7 ± 3.8 –1.5 ± 1.7 –0.2 ± 1.4 Nauru/Night _ _ N=32 N=20 _ _ _ 54 AVG 8.2 ± 6.1 2.1 ± 3.9 _ _ _ 126 AVG 7.1 ± 6.1 1.9 ± 3.2 _ _ _ 186 AVG 6.3 ± 5.4 1.9 ± 3.0 _ _ _ 54 HIST 7.4 ± 7.3 0.3 ± 4.1 _ _ _ 126 HIST 5.3 ± 7.8 –0.7 ± 3.7 _ _ _ 186 HIST 3.0 ± 7.3 –1.1 ± 3.1 _ _ _ 54 MAX 7.0 ± 7.5 –0.5 ± 4.5 _ _ _

AIRS Science Team Meeting, March 7–9, 2006

Variation w.r.t. method of ARM Z definition

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Kahn et al., 2006a

Location/Time Time (min) Height Method

• 0. f < .05

.05 f < .15 .15 f < .5 .5 f < .85 .85 f < 1.0 Manus/Night – – N=13 N=9 N=21 N=16 N=16 54 AVG 7.2 ± 7.0 2.1 ± 3.4 0.4 ± 3.7 –0.1 ± 1.5 0.7 ± 1.8 126 AVG 7.1 ± 6.5 1.8 ± 3.2 0.5 ± 3.6 –0.3 ± 1.2 0.7 ± 2.0 186 AVG 7.0 ± 6.5 1.9 ± 3.0 0.4 ± 3.6 –0.4 ± 1.3 0.6 ± 2.0 54 HIST 7.1 ± 7.3 1.1 ± 5.1 –0.9 ± 3.4 –0.5 ± 1.3 –0.1 ± 1.7 126 HIST 4.9 ± 7.4 –0.5 ± 4.5 –0.9 ± 3.4 –1.2 ± 1.0 –0.3 ± 2.0 186 HIST 4.7 ± 7.5 –0.4 ± 4.1 –1.0 ± 3.3 –1.2 ± 1.0 –0.2 ± 2.0 54 MAX 5.3 ± 8.4 0.6 ± 4.9 –2.2 ± 4.0 –1.4 ± 1.3 –0.8 ± 1.9 Manus/Day _ _ N=21 N=12 N=16 N=12 N=16 54 AVG 7.6 ± 5.6 6.3 ± 5.8 1.2 ± 4.2 0.2 ± 2.3 1.1 ± 1.6 126 AVG 7.8 ± 5.6 4.5 ± 4.9 1.3 ± 3.9 0.5 ± 2.3 1.3 ± 1.6 186 AVG 9.0 ± 5.0 4.4 ± 4.7 1.5 ± 3.8 0.7 ± 2.4 1.6 ± 1.7 54 HIST 6.4 ± 8.8 5.4 ± 6.1 –0.4 ± 3.7 –0.1 ± 2.7 0.5 ± 1.6 126 HIST 3.7 ± 9.5 –1.0 ± 8.3 –0.7 ± 3.8 –1.1 ± 2.1 0.4 ± 1.6 186 HIST 1.5 ± 7.8 –1.5 ± 8.5 –0.8 ± 3.8 –1.1 ± 2.1 0.4 ± 1.5 54 MAX 4.8 ± 8.3 3.1 ± 8.1 –0.7 ± 3.8 –1.5 ± 1.7 –0.2 ± 1.4 Nauru/Night _ _ N=32 N=20 _ _ _ 54 AVG 8.2 ± 6.1 2.1 ± 3.9 _ _ _ 126 AVG 7.1 ± 6.1 1.9 ± 3.2 _ _ _ 186 AVG 6.3 ± 5.4 1.9 ± 3.0 _ _ _ 54 HIST 7.4 ± 7.3 0.3 ± 4.1 _ _ _ 126 HIST 5.3 ± 7.8 –0.7 ± 3.7 _ _ _ 186 HIST 3.0 ± 7.3 –1.1 ± 3.1 _ _ _ 54 MAX 7.0 ± 7.5 –0.5 ± 4.5 _ _ _

AIRS Science Team Meeting, March 7–9, 2006

Agreement strong function of f

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Kahn et al., 2006a

Location/Time Time (min) Height Method

• 0. f < .05

.05 f < .15 .15 f < .5 .5 f < .85 .85 f < 1.0 Manus/Night – – N=13 N=9 N=21 N=16 N=16 54 AVG 7.2 ± 7.0 2.1 ± 3.4 0.4 ± 3.7 –0.1 ± 1.5 0.7 ± 1.8 126 AVG 7.1 ± 6.5 1.8 ± 3.2 0.5 ± 3.6 –0.3 ± 1.2 0.7 ± 2.0 186 AVG 7.0 ± 6.5 1.9 ± 3.0 0.4 ± 3.6 –0.4 ± 1.3 0.6 ± 2.0 54 HIST 7.1 ± 7.3 1.1 ± 5.1 –0.9 ± 3.4 –0.5 ± 1.3 –0.1 ± 1.7 126 HIST 4.9 ± 7.4 –0.5 ± 4.5 –0.9 ± 3.4 –1.2 ± 1.0 –0.3 ± 2.0 186 HIST 4.7 ± 7.5 –0.4 ± 4.1 –1.0 ± 3.3 –1.2 ± 1.0 –0.2 ± 2.0 54 MAX 5.3 ± 8.4 0.6 ± 4.9 –2.2 ± 4.0 –1.4 ± 1.3 –0.8 ± 1.9 Manus/Day _ _ N=21 N=12 N=16 N=12 N=16 54 AVG 7.6 ± 5.6 6.3 ± 5.8 1.2 ± 4.2 0.2 ± 2.3 1.1 ± 1.6 126 AVG 7.8 ± 5.6 4.5 ± 4.9 1.3 ± 3.9 0.5 ± 2.3 1.3 ± 1.6 186 AVG 9.0 ± 5.0 4.4 ± 4.7 1.5 ± 3.8 0.7 ± 2.4 1.6 ± 1.7 54 HIST 6.4 ± 8.8 5.4 ± 6.1 –0.4 ± 3.7 –0.1 ± 2.7 0.5 ± 1.6 126 HIST 3.7 ± 9.5 –1.0 ± 8.3 –0.7 ± 3.8 –1.1 ± 2.1 0.4 ± 1.6 186 HIST 1.5 ± 7.8 –1.5 ± 8.5 –0.8 ± 3.8 –1.1 ± 2.1 0.4 ± 1.5 54 MAX 4.8 ± 8.3 3.1 ± 8.1 –0.7 ± 3.8 –1.5 ± 1.7 –0.2 ± 1.4 Nauru/Night _ _ N=32 N=20 _ _ _ 54 AVG 8.2 ± 6.1 2.1 ± 3.9 _ _ _ 126 AVG 7.1 ± 6.1 1.9 ± 3.2 _ _ _ 186 AVG 6.3 ± 5.4 1.9 ± 3.0 _ _ _ 54 HIST 7.4 ± 7.3 0.3 ± 4.1 _ _ _ 126 HIST 5.3 ± 7.8 –0.7 ± 3.7 _ _ _ 186 HIST 3.0 ± 7.3 –1.1 ± 3.1 _ _ _ 54 MAX 7.0 ± 7.5 –0.5 ± 4.5 _ _ _

AIRS Science Team Meeting, March 7–9, 2006

Agreement much improved for thin ci using lidar

• ver radar

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology AIRS Science Team Meeting, March 7–9, 2006

What about AIRS and MLS?

• MLS is a passive microwave limb sounder
• Reports IWC at 11 altitudes from 46 to 316 hPa
• “Pixel” size roughly 165 × 7 × 3 km (along-track, cross-track, and vertical)
• Use nonzero IWC as a proxy to CTP
• Highest altitude of occurrence of IWC > 0 defined to be CTP
• Lowest values of IWC “similar” to clear sky
• Define AIRS CTP two ways:
• “High”: lowest CTP from 3 nearest along-track
• “Avg”: average CTP from 3 nearest along-track
• Different “views” of similar clouds

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

• Frequency of coincident AIRS and MLS PCLD. The AIRS

values in 20 hPa bins, and MLS reported at the MLS standard pressure levels.

• When we use all AIRS and MLS clouds, PDFs vary

substantially

• When we exclude MLS max IWC < 1.0 mg m–3, the

agreement is similar

• When we exclude MLS first IWC < 1.0 mg m–3, the

agreement is much improved Used ~20 days in January 2005 ± 30 deg latitude Kahn et al., 2006a

AIRS Science Team Meeting, March 7–9, 2006

4 5 6 7 8 9

100

2 3 4 5

Pressure (hPa) 0.20 0.15 0.10 0.05 0.00 Normalized Counts PMLS , all IWC PMLS , max IWC 1.0 PMLS , first nonzero IWC 1.0 PAIRS (high), fUP 0.1

4 5 6 7 8 9

100

2 3 4 5

Pressure (hPa) PAIRS (avg), fUP 0.0 PAIRS (hi), fUP 0.0 PAIRS (hi), fUP 0.1

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Difference between AIRS and MLS PCLD per MLS pressure level: AIRS “hi” approach at top, “avg” approach at bottom Some MLS pressure levels agree much more poorly than others For lowest MLS pressure levels, AIRS and MLS cloud distributions statistically different Lesson: the cloud morphology might look good after averaging, but individual match-ups can have large disagreement Kahn et al., 2006a

120 100 80 60 40 20 Number of Counts

• 200
• 100

100 200 HIGH AIRS – MLS PCLD (hPa) 316 hPa 215 hPa 177 hPa 146 hPa 121 hPa 100 hPa 82 hPa 68 hPa 120 100 80 60 40 20 Number of Counts

• 200
• 100

100 200 AVG AIRS – MLS PCLD (hPa) 316 hPa 215 hPa 177 hPa 146 hPa 121 hPa 100 hPa 82 hPa 68 hPa

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Coincident AIRS and MODIS Cloud Products

• Many cloud products from AIRS and MODIS: focus on operational ECF and CTP
• AIRS reports up to two cloud layers of CTP and ECF, MODIS only one
• MODIS reports ~ 5 km, while AIRS ~ 15 km for ECF, ~45 km for CTP
• Need to collocate AIRS and MODIS: not trivial
• How do we compare similar quantities from different instruments?

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Consistency between AIRS and MODIS cloud products ?

300 280 260 240 220 200 AIRS TCLD (Upper Layer) (K) 300 280 260 240 220 200 MODIS T

CLD (K)

Category 1: all bits in 1st byte = 0 Category 2: one or more bits in 2nd byte > 0 Category 3: one or more bits in 1st byte > 0

1.0 0.8 0.6 0.4 0.2 0.0 AIRS f (Upper Layer) 1.0 0.8 0.6 0.4 0.2 0.0 MODIS f 100

2 3 4 5 6 7 8 9

1000 AIRS PCLD (Upper Layer) (hPa) 100

2 3 4 5 6 7 8 9

1000 MODIS PCLD (hPa)

Kahn et al., 2006b

Left: September 6th, 2002, Granule 11, North-Central subtropical Pacific Ocean Right: Agreement between AIRS and MODIS TCLD, PCLD, and f as a function of AIRS retrieval type. Bottom line: When clouds are thin and broken: bad agreement. When clouds are high and thick: good agreement.

35 30 25 20 15 Latitude 5

• 5

(B) 35 30 25 20 15 Latitude 300 280 260 240 220 (A) 35 30 25 20 15 Latitude 190 185 180 175 170 Longitude 300 280 260 240 220 (C)

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Should we think of cloud products in terms of “a whole” ?

Kahn et al., 2006b

BTAIRS = f1T

1 + f2 T2 + (1 f1 f2)Tsfc

BTMODIS = fcld Tcld + (1 fcld )Tsfc

f1 f2 1–f1–f2

• “Re-build” BT from MODIS and AIRS cloud and surface products
• Replace Planck function by T of emitting layer or surface
• First-order means of comparison: does not guarantee that T or f agree individually ,

but shows if the “sum of the whole” agrees or not

• All products averaged to AMSU scale (~ 45 km)

Bottom line: A way to look at “consistency” of cloud products between AIRS and MODIS AIRS footprint

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Should we think of cloud products in terms of “a whole” ?

Kahn et al., 2006b

Bottom line: BTR is consistent, except near Ci edges – many possible reasons for disagreement

300 280 260 240 220 BTR AIRS (K)

"Cat 1": all bits in 1st byte = 0 "Cat 2": one or more bits in 2nd byte > 0 "Cat 3": one or more bits in 1st byte > 0

300 280 260 240 220 BTR AIRS (K) 300 280 260 240 220 BTR MODIS (K) 1.0 0.8 0.6 0.4 0.2 0.0 % MODIS CTP = Type 6

22 20 18 16 14

• 176
• 175
• 174
• 173
• 172
• 171
• 170
• 15
• 10
• 5

5 10 15

AIRS–MODIS BTR

22 20 18 16 14

• 176
• 175
• 174
• 173
• 172
• 171
• 170

120 100 80 60 40 20 eff_emis

MODIS Effective Emissivity

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Should we think of cloud products in terms of “a whole” ?

Kahn et al., 2006b

Bottom line: BTR is consistent, except near Ci edges – many possible reasons for disagreement

300 280 260 240 220 BTR AIRS (K)

"Cat 1": all bits in 1st byte = 0 "Cat 2": one or more bits in 2nd byte > 0 "Cat 3": one or more bits in 1st byte > 0

300 280 260 240 220 BTR AIRS (K) 300 280 260 240 220 BTR MODIS (K) 1.0 0.8 0.6 0.4 0.2 0.0 % MODIS CTP = Type 6

22 20 18 16 14

• 176
• 175
• 174
• 173
• 172
• 171
• 170
• 15
• 10
• 5

5 10 15

AIRS–MODIS BTR

22 20 18 16 14

• 176
• 175
• 174
• 173
• 172
• 171
• 170

120 100 80 60 40 20 eff_emis

MODIS Effective Emissivity

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Why are there differences?

• MODIS and AIRS look at different clouds: collocation not perfect
• “Misplaced” MODIS cirrus as low cloud
• MODIS cloud mask misses Ci w/ τ < 0.2–0.3
• Multilayered clouds: errors in inferred cloud properties [Baum and Wielicki 1994]
• Method of averaging MODIS to AIRS footprint
• Lessons learned from AIRS/ARM comparisons
• Nonlinearity in BT
• Misfits of MODIS and AIRS radiances, use of different channels
• Systematic errors in retrieval algorithms?
• 3-D IR effects [Liou and Ou 1979; Harshvardhan and Weinman 1982; Cornet et al. 2005]
• BT differences in plane-parallel and cubic clouds ~ 2–5 K or more at TOA
• Look at background picture: Ci is not plane-parallel

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

• 10

10 AIRS–MODIS BTR (B)

• 50
• 45
• 40
• 35
• 30

Latitude 320 280 240 200 BT960 cm

–1 (K)

(A)

• 25
• 20
• 15
• 10

Latitude (D) (E) 300 280 260 240 220 200 AIRS BTR (K) (F) 300 280 260 240 220 200 AIRS BTR (K) 100 80 60 40 20 % 11 µm Retrievals (C)

• 10
• 5

5 10 Latitude

• 92
• 90
• 88
• 86
• 84
• 82
• 80
• 78
• 76

Longitude (J)

• 92
• 90
• 88
• 86
• 84
• 82
• 80
• 78
• 76

Longitude (K) 300 280 260 240 220 200 AIRS BTR (K) 300 280 260 240 220 200 MODIS BTR (K) (L) 55 50 45 40 35 Latitude (G) (H) 300 280 260 240 220 200 AIRS BTR (K) (I)

Midlatitude SH Subtropical/tropical SH Midlatitude NH Equatorial East Pacific

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology AIRS Science Team Meeting, March 7–9, 2006

Summary and Conclusions

• AIRS upper level CTP agrees well with ARM CTH, even for thin cirrus
• Lidar comparisons imply AIRS CTP locates thin cirrus better than MMCR
• Implications for studies of thin ci – AIRS has excellent coverage
• AIRS and MLS cloud placement similar when thin, tenuous cases discarded
• However, height-dependence on agreement
• Holistic view of AIRS and MODIS more consistent than individual comparisons
• Disagreement in reconstructed BT associated with cloud edges, multilayer clouds
• Other possible reasons too
• Useful diagnostic tool
• Confidence in AIRS Version 4.0 clouds, despite large pixel size (~45 km CTP, ~15 km ECF)
• Useful for quantitative analyses, such as cirrus mapping and frequency, and τ and De retrievals

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Towards the Retrieval of Cirrus Particle Size and Optical Depth with AIRS

by Brian Kahn1, Annmarie Eldering1, Kuo Nan Liou2, Omar Mussa2, Shaima Nasiri3, and Qing Yue2

1Jet Propulsion Laboratory, Pasadena, CA, USA 2Department of Atmospheric and Oceanic Sciences, UCLA, Los Angeles, CA, USA 3Department of Atmospheric Sciences, Texas A&M University, College Station, TX, USA

Cloud pictures courtesy of australiansevereweather.com

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Outline

AIRS Science Team Meeting, March 7–9, 2006

• Cirrus frequency from AIRS: how does it compare to other climatologies?
• Multilayered clouds in V4.0
• Mixed phase clouds: simulations of AIRS versus MODIS
• Retrieving thin cirrus De and τ with AIRS radiances
• An example footprint at the Manus Island ARM site
• An example granule in the Tropical Pacific

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

• 20

20

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0

AIRS June/July 2005 HIRS 4-year JJA ISCCP IR only JJA (1983-2001) ISCCP IR/VIS only JJA (1983-2001) Significant differences exist between different platforms!

Where is the cirrus?

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

• 30
• 20
• 10

10 20 30

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0 SON 2004 SON 2004

• 20

20

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0 JJA 2004 JJA 2004

• 20

20

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0 MAM 2004 MAM 2004

• 30
• 20
• 10

10 20 30

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0 JJA 2005 JJA 2005

• 30
• 20
• 10

10 20 30

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0 MAM 2005 MAM 2005

• 30
• 20
• 10

10 20 30

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0 DJF 2005 DJF 2005

• Seasonal maps of Ci frequency from MAM 2004 until JJA 2005 using AIRS V4.0
• Cloud mask from Kahn et al. [2005] + a threshold for “missed” clouds, using BT960 < 273 K.
• A conservative cloud mask which misses many thin cirrus clouds with τIR < 0.1–0.15.
• Despite the conservative thresholds, the frequency exceeds 80–90% over much of the tropics

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

• 30
• 20
• 10

10 20 30

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0 SEP 2004 – AUG 2005 SEP 2004 – AUG 2005

• 30
• 20
• 10

10 20 30 0.4 0.2 0.0

• 0.2
• 0.4

[MAR 2004 to FEB 2005] – [MAR 2004 to FEB 2005] – [SEP 2004 to AUG 2005] [SEP 2004 to AUG 2005]

Top: Yearly average from Sep 2004 – Aug 2005 Note ragged features smooth greatly with a longer time average The maximum frequency decreases to the neighborhood of 80–85%. Bottom: % difference in annual frequency of cirrus between 03/2004 – 02/2005 & 09/2004 – 08/2005 Large Interannual variability is noted in particular regions of interest

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

How realistic are AIRS cloud fields?

• 20

20

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0 JAN 05 ECF > 0.01 JAN 05 ECF > 0.01

• 20

20

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0 JAN 05 ECF > 0.05 JAN 05 ECF > 0.05

• 20

20

• 150
• 100
• 50

50 100 150 JAN 05 ECF > 0.1 JAN 05 ECF > 0.1 1.0 0.8 0.6 0.4 0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0

• 40
• 20

20 40

Figure 1 ECF > 0.01 ECF > 0.03 ECF > 0.05 ECF > 0.1 ECF > 0.15

Jan 05 global zonal avgerage Jan 05 global zonal avgerage

Bottom line: Using f as a “cloud mask” produces reasonable cloud fields compared to Kahn et al. (2005), JGR, and Wylie et al., (1994) J. Climate

1.0 0.8 0.6 0.4 0.2 0.0

• 30
• 20
• 10

10 20 30 DJF 2005 JJA 2005 SEP 2004 – AUG 2005 JJA 1989–1993 Wylie HIRS DJF 1989–1993 Wylie HIRS

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

• 50

50

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0

• 50

50

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0

• 50

50

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0

Multilayer Clouds: January 2005

Clear Single Layer Multilayer

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Nasiri and Kahn, 2006

What about more complicated cloud configurations?

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Nasiri and Kahn, 2006

• 50

50

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0

• 50

50

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0

• 50

50

• 150
• 100
• 50

50 100 150 1.0 0.8 0.6 0.4 0.2 0.0

January 2005 AIRS L2 Version 4.0 January 2005 AIRS L2 Version 4.0

BT 960 cm

–1

< 255 K 255 K < BT 960 cm

–1

< 265 K BT 960 cm

–1

> 265 K

The “uncertain” clouds: A large minority in polar oceans!

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Optical thickness at 11 µm

MODIS sims from DISORT AIRS sims from CHARTS

Nasiri and Kahn, 2006

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Optical thickness at 11 µm

MODIS sims from DISORT AIRS sims from CHARTS

9 km, T = 226 K 7 km, T = 238 K 3 km, T = 262 K 2 km, T = 265 K 1 km, T = 269 K

Nasiri and Kahn, 2006

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Optical thickness at 11 µm

MODIS sims from DISORT AIRS sims from CHARTS

Nasiri and Kahn, 2006

AIRS Science Team Meeting, March 7–9, 2006

Δ = 1 K

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Z3_water

What about the harder cases from 3–5 km? 5 km, T = 256 K 4 km, T = 262 K 3 km, T = 265 K

Nasiri and Kahn, 2006

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Z3_water

What about the harder cases from 3–5 km?

Δ = 0.5 K

5 km, T = 256 K 4 km, T = 262 K 3 km, T = 265 K

Nasiri and Kahn, 2006

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Retrieving cirrus properties

• (Faster) RT model (OPTRAN) + parameterized thin Ci [Yue et al., 2006, JAS, submitted]

– Calculate τVIS and De from AIRS (no scattering… yet)

• (Slower) RT model + multiple scattering (CHARTS)

– complicated atmospheric configurations [e.g., Kahn et al., 2003, GRL]

• AIRS provides: cloud detection [e.g., Kahn et al., 2005, JGR], ZCLD, TCLD, f (up to 2

layers), T(z), RH(z), etc.

• ARM sites provide accurate cloud location, independent validation of τVIS and De, T(z)

and RH(z), etc. Bottom line: Use CHARTS to validate parameterized OPTRAN RT model w.r.t. Ci characterization over ARM sites, then use AIRS data alone to expand beyond ARM sites

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Answer: AIRS cloud products are consistent with other measurements

Ci detection Atmospheric state and surface properties, scattering models OPTRAN + Ci parameterization RT model Best guess τVIS and De AIRS–derived ARM–derived CHARTS Best guess τVIS and De ARM MMCR and lidar Best guess τVIS and De

When are τVIS and De consistent? Dependence on cloud configuration? Can we use AIRS (and MODIS?)

• utside of ARM

sites reliably?

Need consistent scattering properties for all pathways to τVIS and De

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

The “fast” RT approach: OPTRAN + ci parameterization

• Combine OPTRAN clear-sky radiances with a thin cirrus parameterization
• Cirrus represented by series of De and habit distributions
• Fit AIRS radiance to best τ and De and habit distributions: the Ci “retrieval”

Yue et al., 2006

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

The “fast” RT approach: OPTRAN + ci parameterization

• Combine OPTRAN clear-sky radiances with a thin cirrus parameterization
• Cirrus represented by series of De and habit distributions
• Fit AIRS radiance to best τ and De and habit distributions: the Ci “retrieval”

Size and habit models impact here From AIRS L2 retrieval

Yue et al., 2006

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

The “fast” RT approach: OPTRAN + ci parameterization

Yue et al., 2006

Sensitive to De, habit distribution, and τVIS

AIRS Science Team Meeting, March 7–9, 2006

• 9 size distributions
• 11 habit distributions
• 100 τVIS from 0–1.0

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

An illustrative example on June 20, 2003, at Manus Island

Kahn et al., 2005, J. Geophys. Res. BT 960 cm

–1

BT 2616 cm

–1

– BT 960 cm

–1

“cloud mask” Total column precipitable water vapor Manus Island Arm site (blue cross)

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

14x10

3

12 10 8 6 4 2 MMCR–derived height (m) 17.5 17.0 16.5 16.0 15.5 15.0 14.5 Time (UTC)

Param RT Model CHARTS MMCR τVIS

0.26

0.28 0.13 De (µm) 91.5 71.4 106

1

2 4 6

10

2 4 6

100

2 4 6

1000 Chi–Squared 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 Tau_VIS

40 µm (Hex Cyl) 80 µm (Hex Cyl) 102 µm (Hex Cyl) 122 µm (Hex Cyl) 44 µm (Aggregates) 71 µm (Aggregates) 92 µm (Aggregates)

CHARTS and parameterized RTM retrievals have larger τVIS and smaller De than MMCR: indicative of missed small particles by MMCR? Tempting to say… but need more cases, and add in MPL!

An illustrative example on June 20, 2003, at Manus Island

ARM cloud height from radar χ2 model–

• bs

fit

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

An illustrative granule on July 1st, 2003

BT960 (K) Upper CTP (hPa)

AIRS Science Team Meeting, March 7–9, 2006

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

An illustrative granule on July 1st, 2003

BT960 (K) Upper CTP (hPa)

AIRS Science Team Meeting, March 7–9, 2006

Cloud top increases away from convective towers Is this the elusive “upper peak” in the AIRS-MLS comparisons?

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

An illustrative granule on July 1st, 2003

ECF (Upper) τVIS

AIRS Science Team Meeting, March 7–9, 2006

15 10 5 185 180 175 170 165 1.0 0.8 0.6 0.4 0.2 0.0 UCF 15 10 5 185 180 175 170 165 1.0 0.8 0.6 0.4 0.2 0.0 Tau_VIS

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

An illustrative granule on July 1st, 2003

De (microns) Habit Distribution

AIRS Science Team Meeting, March 7–9, 2006

15 10 5 185 180 175 170 165 10 8 6 4 2 Habit Distribution 15 10 5 185 180 175 170 165 120 100 80 60 40 20 De (microns)

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

ECF (Upper) versus τVIS Frequency of size dist and habit dist

AIRS Science Team Meeting, March 7–9, 2006

An illustrative granule on July 1st, 2003

8 6 4 2 10 8 6 4 2 600 400 200

0.4 0.3 0.2 0.1 0.0 Upper Level ECF 1.0 0.8 0.6 0.4 0.2 0.0 Tau_VIS

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

ECF (Upper) versus τVIS Frequency of size dist and habit dist

AIRS Science Team Meeting, March 7–9, 2006

An illustrative granule on July 1st, 2003

8 6 4 2 10 8 6 4 2 600 400 200

0.4 0.3 0.2 0.1 0.0 Upper Level ECF 1.0 0.8 0.6 0.4 0.2 0.0 Tau_VIS

Thinnest Ci: 33.7% SC, 24.7% BR, 41.6% A: McFarquhar et al. [1999] Slightly thicker Ci: 100% SC: Baum et al. [2005]

National Aeronautics and Space Administration Jet Propulsion Laboratory California Institute of Technology

Summary and Conclusions

AIRS Science Team Meeting, March 7–9, 2006

• AIRS maps out cirrus realistically
• Limits of thin cirrus detection have not been reached
• AIRS may be useful for more complicated cloud configurations
• Cloud phase – tradeoff between sensitivity and footprint size when compared to MODIS
• Multilayered clouds – V4.0 clouds have coherent patterns
• Fast RT approach to retrieve thin cirrus De and τ with AIRS radiances
• Future modifications with 4-stream approximation…thicker Ci
• Efficiency of calculation work in progress
• Further comparisons to ARM site-derived and MODIS-derived De and τ are necessary