Duane E. Waliser 1, Baijun Tian 12, and Xianan Jiang 12 1 Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 2 Joint Institute for Regional Earth System Science and Engineering, University of California, Los Angeles, CA Vertical Structure And Processes Revealed With Recent Satellite Data BIRS, 2009

Figures: E. Maloney, PMEL/TAO, M. Wheeler, J. Lin, D. Waliser Kelvin Waves Rossby Waves MJOs The MJO is the dominant form of intraseasonal variability in the Tropics, with impacts a wide range of phenomena. Our weather & climate models have a relatively poor representation Aspects of Vertical Structure – which may be important to initiation/maintenance – have been difficult to evaluate via observations. Space-based observations now make it possible to examine aspects of vertical structure of the MJO hydrological cycle. Motivation

Question? Using space-based observations, what can be said about the hydrological cycle of the MJO?

Hydrological Data  CMAP Rainfall : global, 2.5°x2.5° lat-long, pentad, 01/01/ /22/2007. Xie and Arkin (1997)  TRMM 3B42 Rainfall: 40S-40N, 0.25° x 0.25°, 3-hourly, 01/01/ /30/2007. Huffman et al. (2007)  AIRS H2OVapMMR & TotH2OVap V4, L3, global, 1.0° x 1.0°, 2Xdaily, 09/01/ /30/2007. Chahine et al. (2006)  QuikSCAT & TMI Moisture Transport 40S-40N, 0.25° x 0.25°, 2Xdaily, 08/ /31/2005. Liu and Tang (2005)  OAFlux Evaporation 65S-65N, 1.0° x 1.0°, daily, 01/01/ /31/2002. Yu and Weller (2007)  SSMI Total Column H2O Vapor & Total Cloud Liquid H2O V6, DMSP F13, global, 0.25° x 0.25°, 2Xdaily, 01/01/ /30/2007. Wentz (1997), Wentz and Spencer (1998)  MLS Ice Water Content 80S-80N, 4° x 8° lat-long, 2Xdaily, 08/26/ /22/2007. Wu et al. (2006)

Spatial-temporal Pattern of the 1 st EEOF Mode of Rainfall Anomaly MJO Event Selection

MJO Events in Hydrological Time Series TRMM: 18 CMAP: 57 AIRS:11 QuikSCAT&TMI: 13 OAFlux: 44 SSMI: 23 MLS: 5 Principal Component Time Series of 1st EEOF Mode of Rainfall Anomaly

Rainfall Pattern & Data Sensitivity

Rainfall & Moisture Convergence -20 Days -10 Days 0 Days +10 Days +20 Days Rainfall and Total Column Moisture Convergence tend to be Correlated throughout Tropics - except maybe over S. America

-20 Days -10 Days 0 Days +10 Days +20 Days Rainfall & Surface Evaporation Largest Evap anomalies in the subtropics in association with Rossby grye modulations of tradewind regimes Near-equatorial Evap anomalies tend to lag precipitation anomalies

Composite Hydrological Cycle Vertical Structure Water Vapor Cloud Ice

-0.5 mg/m 3 +3 mm/day 600 hPa 900 hPa 300 hPa -3 mm/day -0.3 gm/kg ~ 45 days Surface Upper Troposphere - See Other Diagram +0.3 gm/kg +0.1 gm/kg +0.5 mg/m 3 -3 mm/day+3 mm/day +2 mm -2 mm mm mm MJO Hydrological Cycle - Troposphere -0.1 gm/kg -0.2 mm/day+0.2 mm/day Column Integrated Values

+1 mg/m mg/m 3 +3 mm/day -1 mg/m mg/m 3 ~150 hPa ~250 hPa ~100 hPa -3 mm/day +100 ppmv ppmv +1 ppmv +100 ppmv ppmv +1 ppmv ~ 45 days +0.5 K -0.5 K Surface Lower-Middle Troposphere - See Other Diagram MJO Hydrological Cycle - UTLS Schwartz, M. J., D. E. Waliser, B. Tian, J. F. Li, D. L. Wu, J. H. Jiang, and W. G. Read, 2008: MJO in EOS MLS cloud ice and water vapor. GRL.

Total-column Moisture Budget Surface Rainfall Surface Evaporation Moisture Convergence due to large- scale moisture transport Total column Moisture change Moistening (>0) Drying (<0)

Summary: I Satellite Observations are now able to provide an estimate of the chief components of the Hydrological Cycle Associated with the MJO, in some cases with vertical structure information. However, calcululations of the Residual Term of the column-integrated values indicates closing the budget with current generation of satellite retrievals is difficult. Within the levels of uncertainty, Future plans involve applying the observed Hydrological Cycle of the MJO as a means to diagnose, evaluate and validate GCM simulations of the MJO or Evaluate Theoretical considerations.

Question? What Physical or Dynamical Mechanism is Responsible for the Lower-tropospheric Moisture Preconditioning of the MJO?

17

18

19

20

21

22

23

24

25 Summary: II significant moisture anomalies are located in the lower troposphere with maxima around 700 hPa during the transition phase; total-column and lower-tropospheric moisture change anomalies are positively correlated. moisture change anomalies are positively correlated with moisture convergence anomalies but negatively correlated with rainfall and surface evaporation anomalies. moisture change anomaly is highly & positively correlated with the difference between moisture convergence and rainfall anomalies. implication: lower-tropospheric moisture preconditioning of the MJO is due to the small difference between moisture convergence and rainfall anomalies instead of surface evaporation anomaly.

Question? What types of clouds and cloud processes play a role in the moist pre-conditioning? Considered w.r.t. to boreal summer.

Dataset Cloudsat (Jun– Sep 2006, 2007) Horizontal resolution: 1x1 degs Variables: Cloud liquid water content (LWC) Ice water content (IWC) Cloud types High: Cirrus Middle: Altocumulus (Ac), Altostratus (As) Low: Stratocumulus (Sc), Stratus (St), Nimbostratus (Ns) Vertical: Cumulus (Cu) GPCP rainfall ( ): horizontal resolution: 1x1 deg., day band-pass filtered

Hovmöller diagram of GPCP precipitation (20-70-day filtered; o E) Time series of EEOF1 of 1-D 20-70d filtered GPCP rainfall (5 o S~25 o N, averaged over o E sector) for MJJAS, The EEOF1&2 basically captures northward propagation of the BSISO (mm/day)

-10day Time-latitude evolution (75-85 o E) Composite BSISV Evolution (7 events) GPCP rainfall (mm/day) Northward Propagation

Vertical Tilting in LWC Low-level LWC leading the convection center Composite Cloud LWC (85-95 o E average) (no time-filtering, seasonal mean removed) (mg/m 3) hPa (mm/day) Cloud LWC rainfall -5d 0d 5d 10d 15d 20d

IWC generally in phase with convection Composite Cloud IWC (mg/m3) (85-95 o E) -5d 0d 5d 10d 15d 20d hPa

LWC variation associated with BSISV mainly related to non-precipitating and drizzling mid-low clouds; Altocumulus cloud are crucial for mid-level LWC variation; Stratocumulus cloud important in the low-level with contribution from cumulus. LWC by Cloud Types S c +C u ACAC Total Non-precip Conditions Total Precipitating Conditions ACAC ScSc

80-90 E – Bay of Bengal CloudSat Application: MJO/ISV–driven Monsoon Onset & Breaks Convective Center Some Drizzling Sc Most Non-Precip Ac Some Non-Precip As Most Precip Deep Conv

Summary: III During the northward propagation of the BS MJO, the cloud ice water content (IWC) in upper troposphere tends to be in phase with convection. A marked vertical tilting is discerned in cloud liquid water content (LWC) with respect to the convection center. Increased LWC leads the convection, particularly in the lower troposphere. IWC variability is largely associated with deep convective clouds; while LWC is mainly linked to non-precipitating Altocumulus at mid-level and drizzling Stratocumulus cloud at low-level; with the latter two appearing to play a role in pre-conditioning for the northward propagation.

Washington DC USGS Map 13.5 km AIRS IR; AMSU & HSB m wave 6x7 km POLDER 5.3 x 8.5 km TES Cloud 0.5 km MODIS Band km CALIPSO 1. 4 km Cloudsat OCO 1x1.5 km Afternoon Constellation Instrument Footprints (Source: M. Schoeberl, 2003)

YOTC: A-Train Data Co-Location Possibilities for Studying & Modeling Cloud/Convection P (hpa) q i (p) q l (p) AMSRPrecipitationSST Prec Water LWP Surf. Wind Speed AIRSq(p)T(p) ECMWF  (p) u(p)du/dp(p) div H (p) q i (p) & IWP q l (p) & LWP Cloud Type (p) ~ Particle Size (p) Light Precip CloudSat Aerosol Opt Depth Cloud Top - Temperature Pressure, Particle Size, etc Pressure, Particle Size, etc MODIS Aerosol (p) Cloud (p) CALIPSO  < ~3 UTLS – T(p), q(p), q i (p), CO (p), O 3 (p), HNO 3 (p) MLS TOA and SFC radiative fluxes CERES