Impacts of Large-scale Controls and Internal Processes on Low Clouds Observational and Numerical Studies Xue Zheng RSMAS, University of Miami Acknowledgement:

Impacts of Large-scale Controls and Internal Processes on Low Clouds

Observational and Numerical Studies

Xue Zheng

RSMAS, University of Miami

Acknowledgement:Bruce Albrecht, Virendra Ghate, and coauthorsAmy Clement, Ping Zhu, and Paquita Zuidema

Low clouds over the world

From VOCLAS, RICO website

Stratocumulus

Cumulus

Cumulus

The climatological importance

Hahn and Warren 2007; Ackerman et al. 1993; Warren et al. 1988; Hartmann et al. 1992; Slingo 1990; Stephens and Greenwald 1991; etc.

Cloud-controlling factors

• Large-scale controls– lower tropospheric static stability(LTS),

SST and SST advection, large scale subsidence, free-troposphere humidity

(Albrecht et al. 1995)

• Internal processes– precipitation and cool pool– entrainment– decoupling– aerosol-induced processes

(Rauber et al. 2007) (NASA)

The uncertainty of low-cloud feedback

“…, process studies leading to a better assessment of the behaviour of MBL clouds … will have the potential to reduce substantially the uncertainty in model predictions of tropical cloud feedbacks and climate sensitivity. ”- Bony and Dufresne 2005

Low clouds

Better understand the cloud-controlling factors and related mechanisms

Hopefully, provide ideas to improve the low-cloud simulation in climate models

Motivation

Data and methodology

• ARM Nauru cumulus observation and VOCALS stratocumulus observation

Important factors for cloud variations

• Large eddy simulations with observed large-scale forcing

Process-oriented simulations

• Nested WRF simulations for stratocumulus cases

Further test in more realistic simulations

Cumulus clouds

Seasonal variability of low-cloud amount

Spring Case: Mar-Apr, 2000 20%Summer Case: Aug-Spt, 2000 10%

Spring Case

Summer Case

From Bruce Albrecht

Summer Case• Warmer SST (302.4 K)• Weaker surface wind (4.1 m/s)

Spring Case• Colder SST (300.4 K)• Stronger surface wind (5.6 m/s)

SummerSpring

Composite large-scale profiles

Remote sensing measurements

Radar reflectivity (dBz)

3/30/2000

Vertical Velocity (m/s)

Spring Case

Summer Case

Color contours:Negative buoyancy (m2/s1)

More active

More stratiform

19% cloudiness

9% cloudiness

Observation-simulation comparison

Obs. Obs.

LES LES

SummerSpring

SummerSpring

Simulations based on the two observed states capture the cloudiness and cloud structure differences

Nudging SST (=1K)

Rain scheme

Wind LTS

, qt

U,V

Mar-Apr +Aug-Spt-

No rain Exchange U profiles

Exchange the inversion strength

, qt

U,V

Same as above

Same as above

Same as above

Same as above

The impact of large-scale forcing

16 sensitivity cases: 8 cases for Spring Case, 8 cases for Summer Case

Control

SST

Wind

Strong constrainsWeak constrains

Summary of all 20 cases • Spring Case is more

sensitive to large-scale forcing

• Most sensitive to lower tropospheric static stability (inversion)

• If the inversion strength is constrained, both cases are insensitive to SST

• Wind profiles also have impacts on cloudiness

(Zheng et al. 2013a)

LTS

The impacts of precipitation and cold pools

Summer

Spring

Summary for cumulus study

High Cloudiness Low Cloudiness

Shallow cumulus, very weak cool pool

Stratocumulus clouds

CIRPAS Twin Otter Instrumentation

Oct 16 – Nov 13, 200815 out of 18 flights were around 8 AM local time

Observed CCN and LWP relationships

(Zheng et al. ACP 2011)

Observed CCN and LWP relationships

Precip. suppression

Observed CCN and LWP relationships

Precip. suppression

Large-scale controls

Observed CCN and LWP relationships

• A strong positive correlation between the LWP and the BL CCN

(Zheng et al. GRL 2010)

• What about sedimentation/entrainment feedback?

• Could be caused by earlier cloud history?

Precip. suppression

Large-scale controls

?

Case CCN (cm-3)zi0

(m)

LWP0

(g m-2)Comments

A0 2001055 117

Constant BL with thinning cloud layerA1 2000

B0 2001055 18

Deepening BL with deepening cloud layer

B1 2000

C0 200

900 47Deepening BL with constant cloud layerC0.5 400

C1 2000

The impact of CCN on non-drizzling stratocumulus

• Entrainment instability index k

• The cloud top interface of polluted cloud is not unstable compared with clean clouds

k =

(Lilly 2002)

Case A0 A1 B0 B1 C0 C0.5 C1

LWP (g m-2) 53 47 42 38 47 45 43

k 1.03 1.02 1.24 1.23 1.14 1.13 1.12

The LWP of the polluted clouds ↓ <10% (5%) after 12h

(Zheng 2012)

WRF SIMULATIONS

• Initialized with the Naval Research Laboratory’s COAMPS real-time forecasts

• 4 nested domains:– 4km, 1.4km, 450m, 150m – 91 (/128) levels < 850hPa

• 10/19/2008 Case: High LWP• 10/27/2008 Case: Low LWP• No aerosol indirect effects• 60-hour simulation• Diurnal cycle

(Zheng et al. 2013b)

GOES visible images during the similar time

Maximum cloud liquid water mixing ratio

Summary for stratocumulus study

• The aerosol indirect effect on non-drizzling stratocumulus is limited.

• The observed LWP difference is captured by WRF simulation in the absence of aerosol indirect effect.

• The large-scale factors and internal processes can have large impacts on the cloud LWP variability.

• The LWP increases with the CCN concentrations in spite of lack of precipitation.

Conclusions

• Large-scale atmospheric pattern, including large-scale wind pattern might play the lead role in the low-cloud variability: low cloud is closely tied to large-scale circulations

• Internal processes (e.g. precipitation) responding to the large-scale forcing can also play an important role in the low-cloud variability depending on the cloud regime

THANK YOU!