Moist static energy budget diagnostics for monsoon research · 2015-06-14 · 4xCO. 2. minus. 20c3m...

Post on 13-Aug-2020

0 views 0 download

transcript

Moist static energy budget diagnostics for monsoon research

H. Annamalai

JJAS – Precipitation and SST Climatology

I II

III

• Multiple regional heat sources - • EIO and SPCZ – still experience high precipitation (thermal equator at 20oN) • Central India rainfall – dynamical effects; Rain-shadow regions • absolute ascent over a large domain

Observed Boreal Summer ISV

OLR anomalies W/m2

“one-phase of the ISV” Annamalai and Sperber (2005, JAS) Lau and Chan (1986, JAS); Krishnan (2000, JAS)

“internal dynamics”

JJAS rainfall anomalies (2002/04/09) - TRMM

“Boundary forcing”

4xCO2 minus 20c3m

Stowasser, Annamalai and Hafner (2009 J. Climate)

JJAS Precipitation response in CM_2.1

(mm/day) CMIP5 results – more rainfall over west Pacific – descent over EIO/India Radiative forcing effect – local air-sea

General Hypothesis Interaction between equatorial waves and moist physics needs to be understood for attributing the causes for precipitation anomalies over “mean ascent” regions

∂m∂t

= − V • ∇m − ω∂m∂p

+ LH + SH + LW + SW

m = CpT + gz + Lq

Representation of interaction between cumulus convection and circulation requires consideration of moisture and temperature that is represented by MSE, m, given by

The vertically integrated MSE tendency is approximately given by

WTG approximation – temperature advection is negligible

Horizontal advection

Cloud-radiative interaction

Charging/ discharging MSE export

Vertical adv

+ residuals

fluxes

“since 1994….in a given year monsoon rainfall over India has not exceeded 10% above normal” but such incidents have occurred in the past too -

30 years below normal

40 years above normal

40-50 years of below normal

2 La Nina years – below normal rainfall

Delaware product

PREC/Land (NOAA)

CRU product

“spatial coherency – amplitude differs”

SST averaged over the tropical Indian Ocean – West Pacific

Significant SST rise (above natural variability) since ~ 1950s

weak rise no change

Since 1950 – rate of rise is pronounced Its relevance to monsoon?

JJAS SST linear trend

SST rise over climatological low rainfall regions – any changes in evaporation – obs??

??????

SLP (shading) and 850 hPa wind – linear trend 1949-2005

monsoon trough over India weakens” SLP deepens over the tropical western Pacific – “Australian and Mascarene High – intensify – cross-equatorial flow over the western Pacific is strengthened - cyclonic vorticity (regional circulation changes) “ Despite SST rise, atmosphere over the tropical Indian Ocean has not yet responded”

(a) CRU rainfall (b) SLP / Wind 850hPa (c) Hadley Centre SST

(d) CM2.1 rainfall (e) CM2.1 SLP – Wind 850hPa (f) CM2.1 SST

SST rise – shifts the monsoon circulation more rainfall over tropical west Pacific less rainfall over South Asia

Presenter
Presentation Notes
(a) Linear trend in observed rainfall during boreal summer (June through September) from Climate Research Unit gridded data set for the period 1949-2000. Negative values are shown in blue and purple while positive values in red. Unit is (mm/month/52 years); (b) same as (a) but for sea level pressure from Hadley center data (shaded in blue for negative and red for positive values), and 850 hPa wind from NCEP-NCAR reanalysis. Unit for SLP is (hPa/52 years) and for wind is (m/sec/52years): (c) same as (a) but for sea surface temperature data from HadISST (shaded in blue for negative and red for positive values). Unit for SST is (oC/52years): (d) same as (a) but for CM2.1 coupled model; (e) same as (b) but from CM2.1 coupled model, and (f) same as (c) but from CM2.1 coupled model. In each of the panels, trend values above the interannual standard deviations are only shown. Note that in the model, the trend is estimated in each of the ensemble members for the period 1949-2000 separately and then a grand ensemble mean is

SST trend shifts the monsoon circulation – promotes more rainfall over the tropical western Pacific - subsequent descent through Rossby waves and dry air intrusion aid in the weakening of rainfall over South Asia

Working Hypothesis

Numerical experiments performed

Monthly observed SST trend (1949 – 2000) superimposed on clim. SST – 5 members 1.Tropical oceans (GFDL – AM2.1) 1.Tropical Indo-Pacific warm pool (GFDL - AM2.1) 1.Tropical west Pacific only (GFDL – AM2.1)

2. Linear barolcinic model (steady-state solutions) – to identify Rossby wave dynamics

Rainfall linear trend (AM21 simulated)

SLP (shaded) and 850 hPa wind

“Australian high – not consistent with reanalysis products”

Evaporation trend – simulated by AM2.1

“Evaporation decrease along the cross-equatorial flow is due to wind anomalies despite SST rise is prescribed in the model experiment”

(a) Precipitation (b) SLP and 850 hPa winds (c) MSE divergence

(d) Evaporation (e) Moisture advection (f) Temperature advection

Dry, cool air penetrates South Asia

Linear trend simulated by AM2.1

Presenter
Presentation Notes
52-year trend during boreal summer (June through September) simulated by the GFDL atmosphere model (AM2.1) forced by SST trend over the tropical Indo-Pacific warm pool region (EXP2). The trend is obtained by subtracting the EXP2 values from the Control run that was forced by climatological SST: (a) precipitation (W/m2); (b) SLP (hPa) and 850 hPa winds (m/s); (c) vertically integrated moist static energy divergence (W/m2); (d) surface evaporation (mm); (e) vertically integrated moisture advection (W/m2) and (f) vertically integrated temperature advection (W/m2). In (a) the region outlined in solid lines represents south Asia, and in (b), the region outlined in dotted lines represents tropical western Pacific. Note that precipitation unit conversion is 28 W/m2 = 1.0 mm/day.

(a) SLP and 850 hPa winds (b) Vertical velocity 400 hPa

Day 6

Day 9

(c) SLP and 850 hPa winds (d) Vertical velocity 400 hPa

(e) Precipitation and 850 hPa winds (f) Moisture advection

Day 20

Rossby wave interpretation

Presenter
Presentation Notes
Results from a linear atmosphere model forced by SST trend (52 years accumulate value of about 0.6oC) over the tropical western Pacific: (a) SLP (hPa) and 850 hPa winds (m/s) at day 3; (b) vertical velocity (hPa/s) at day 3; (c) same as (a) but for day 9; (d) same as (b) but for day 9; (e) precipitation (W/m2) and 850 hPa winds at day 20, and (f) vertically integrated moisture advection (W/m2) at day 20. Note that the model is integrated for 30 days with fixed forcing, and it attains steady-state at day 15 of the integration.

• MSE is a powerful diagnostic to identify leading moist and radiative processes deem responsible for rainfall anomalies over mean ascent regions • MSE budget residuals – observational constraints over Monsoon regions • Model improvement – need 3-d moisture and radiation observations • Monsoon Mission on Observations

Summary

May averaged CM2.1 composite of anomalous 850 hPa stream line and rainfall

Severe weak monsoons over south Asia co-occurred with developing phase of El Nino

Pillai and Annamalai (2012, J. Atmos. Sci. )

NIO – anticyclonic vorticity – within 2-3 days of SST forcing – rainfall after about 20 days Dry air advection from north is instrumental in initiating the dryness

May rainfall and 850 hPa wind response to El Nino SST forcing

W/m2

AM2.1 solutions – Forced with CM2.1 composite SST anomalies (El Nino)

Days from 15 March

Rainfall over S. Asia 850 hPa Vorticity (west of rainfall maximum)

“dry advection leads rainfall anomalies – long lead time – useful for prediction”

AM2.1 solutions

Forced AM2.1 with CM2.1 composite El Nino SST anomalies + 20c3m climatology (25 members; 01 March – 30 November)

(b)

Control simulation with a coupled model (no anthropogenic forcing included)

Blue (rainfall anomalies over India) Red (rainfall anomalies over tropical west Pacific)

Years

“no clear TREND of rainfall”

“breakdown of the decadal-multidecadal variability?”

anthropogenically induced?

Low variability

High variability High variability

6-8% decline

CanESM2 CCSM4

CNRM-CM5 GFDL-CM3