Tropical weather systems within a global data assimilation and forecasting framework

Tropical weather systems Tropical weather systems within a global data within a global data

assimilation and forecasting assimilation and forecasting frameworkframeworkOreste RealeOreste Reale

NASA Goddard Laboratory for AtmospheresNASA Goddard Laboratory for Atmospheresandand

GEST/UMBCGEST/UMBC

MotivationMotivation The deadliest natural events are related to tropical The deadliest natural events are related to tropical

weather systems (500,000 people died because of weather systems (500,000 people died because of 1970 cyclone in Bangla Desh) 1970 cyclone in Bangla Desh)

Almost 40 years later (2008): Tropical Cyclone Almost 40 years later (2008): Tropical Cyclone Nargis killed Nargis killed at leastat least 140,000 people 140,000 people

Numerical weather forecasts in the tropics have Numerical weather forecasts in the tropics have improved at a slower pace than mid-latitude improved at a slower pace than mid-latitude weather forecastsweather forecasts

Acknowledging intrinsic predictability limitations, Acknowledging intrinsic predictability limitations, there is vast room for improvement there is vast room for improvement

OutlineOutline Part I:Part I: the global analysis in the tropics the global analysis in the tropics Part II:Part II: the representation of tropical cyclones in the representation of tropical cyclones in

global modelsglobal models Part III:Part III: improvements stemming from the improvements stemming from the

assimilation of AIRS-derived products assimilation of AIRS-derived products Part IV:Part IV: other improvements other improvements ConclusionsConclusions and futureand future

Outline of Part IOutline of Part I

AEJ representation in state-of-the-art AEJ representation in state-of-the-art reanalysesreanalyses AEJ representation on weather-time-scales in AEJ representation on weather-time-scales in

operational analysesoperational analyses during SOP-3 NAMMA during SOP-3 NAMMA (2006)(2006)

Vertical soundings during SOP-3 NAMMA Vertical soundings during SOP-3 NAMMA (2006)(2006) Mid-tropospheric flow over the entire tropical Mid-tropospheric flow over the entire tropical

Pacific in Pacific in August 2010August 2010 in in NCEP operational, NCEP operational, ECMWF operational, ECMWF operational, andand MERRA MERRA

AEJ representation in AEJ representation in state-of-the-art reanalysesstate-of-the-art reanalyses

Previously published work (Wu et al., 2009) shows Previously published work (Wu et al., 2009) shows substantial differencessubstantial differences between reanalyses in the between reanalyses in the monthly meanmonthly mean representation of the representation of the African African Easterly Jet (AEJ)Easterly Jet (AEJ)

ERA-40, NCEP-R2, JRA-25ERA-40, NCEP-R2, JRA-25 provide provide very different very different descriptions descriptions of theof the AEJ structure AEJ structure, and of the, and of the horizontal shearhorizontal shear in thein the cyclonically-sheared cyclonically-sheared portion portion of the AEJ of the AEJ

M.-L. C. Wu, Reale, O., S. Schubert, M. Suarez, M.-L. C. Wu, Reale, O., S. Schubert, M. Suarez, R. Koster, P. Pegion, 2009: R. Koster, P. Pegion, 2009:

African Easterly Jet: Structure and Maintenance. African Easterly Jet: Structure and Maintenance. J. Climate, J. Climate, 22,22, 4459-4480. 4459-4480.

From From Wu et al. (2009)Wu et al. (2009)

Fig 2Fig 2

July zonal windJuly zonal wind (m s−1, contours every 1 m s−1, 0 omitted, solid: positive, (m s−1, contours every 1 m s−1, 0 omitted, solid: positive, dashed: negative) climatology dashed: negative) climatology (1980–2001)(1980–2001) based on (top to bottom) based on (top to bottom) ERA-40, NCEP R2, ERA-40, NCEP R2, and JRA-25and JRA-25 data: (left) meridional horizontal shear of the zonal wind at 600 hPa data: (left) meridional horizontal shear of the zonal wind at 600 hPa and (right) meridional cross section at 0° longitude. (Wu et al., 2009, and (right) meridional cross section at 0° longitude. (Wu et al., 2009, J.Climate)J.Climate)

Large differencesLarge differences in inAEJ AEJ SHAPESHAPE,, INTENSITY INTENSITYVERTICAL STRUCTUREVERTICAL STRUCTUREand and distribution of thedistribution of thehorizontal shearhorizontal shearin a in a 22-year average22-year averageperformed on performed on ERA-40,ERA-40,NCEP-R2,NCEP-R2, and and JRA-25JRA-25. .

AEJ and its instability properties in AEJ and its instability properties in state-of-the-art reanalysesstate-of-the-art reanalyses

Work recently submitted (Wu et al., 2011) shows Work recently submitted (Wu et al., 2011) shows differencesdifferences in the representation of the in the representation of the African Easterly Jet African Easterly Jet (AEJ)(AEJ) seasonalseasonal instability propertiesinstability properties between reanalyses between reanalyses across a 22-year averageacross a 22-year average

Despite revealing some instability property of the AEJ that Despite revealing some instability property of the AEJ that appear data-independent, ERA-40, NCEP-R2, JRA-25 and appear data-independent, ERA-40, NCEP-R2, JRA-25 and MERRA provide very different descriptions of the AEJ MERRA provide very different descriptions of the AEJ horizontal structure, intensity, and of some properties that horizontal structure, intensity, and of some properties that control wave instability on a control wave instability on a seasonal scale (JAS)seasonal scale (JAS)..

M.-L. C. Wu, Reale, O., S. Schubert, M.-L. C. Wu, Reale, O., S. Schubert, M. Suarez, C. Thorncroft, 2011: M. Suarez, C. Thorncroft, 2011:

African Easterly Jet: barotropic instability, waves and African Easterly Jet: barotropic instability, waves and cyclogenesis.cyclogenesis.

Submitted to: J. Climate.Submitted to: J. Climate.

From From Wu et al. (2010)Wu et al. (2010)

Fig 2Fig 2

The analyses differ in terms ofThe analyses differ in terms ofstrengthstrength and and intensity intensity of the low-levelof the low-level

monsoonal flowmonsoonal flow, , slopeslope of the of the barotropically unstablebarotropically unstable part of the part of theAEJ, AEJ, horizontal shear distributionhorizontal shear distribution. .

All Figures show a All Figures show a 22-year JAS average22-year JAS average

Unexpected discrepancies between snapshots of Unexpected discrepancies between snapshots of analyzed representation of the African Monsoon-analyzed representation of the African Monsoon-

Eastern Tropical Atlantic regionsEastern Tropical Atlantic regions

TheThe African Easterly Jet African Easterly Jet at about 600hPa,at about 600hPa, thethe low- low-level monsoonal flow level monsoonal flow (predominantly southwesterly (predominantly southwesterly between 1000 and 800 hPa)between 1000 and 800 hPa) and theand the Tropical Easterly Tropical Easterly Jet Jet (between 200 and 100 hPa)(between 200 and 100 hPa) are theare the critical players critical players inin Atlantic tropical development. Atlantic tropical development.

Comparison between operational NCEP analyses and Comparison between operational NCEP analyses and GEOS-5-produced analyses reveal GEOS-5-produced analyses reveal serious serious discrepanciesdiscrepancies

Validation agains the only vertical sounding in the area Validation agains the only vertical sounding in the area at Cape Verde (15N, 23.5W) during the 2006 NAMMA at Cape Verde (15N, 23.5W) during the 2006 NAMMA campaign,campaign, show that both analyses have large errors show that both analyses have large errors

Huge discrepancies between Huge discrepancies between GEOS-5 and NCEP operational analysesGEOS-5 and NCEP operational analyses

Wind at 5-15N, Wind at 5-15N, 500-600 hPa, has500-600 hPa, hasopposite opposite direction!direction!

Only in the tropicsOnly in the tropicsthe two analysesthe two analysesdiffer substantiallydiffer substantially

Section at 23.5WSection at 23.5W

Largest differences between reanalyses are in the tropics, Largest differences between reanalyses are in the tropics, at about 15N (at about 15N (on the order of 12m/son the order of 12m/s))

larger even thanlarger even thandiscrepancies in the southern hemisphere jet streamdiscrepancies in the southern hemisphere jet stream

NCEPNCEP

GEOS-5GEOS-5

Huge differences in the entire tropical zonal flow Huge differences in the entire tropical zonal flow from 20S to 20N at all levels from 20S to 20N at all levels

Largest Largest mid-troposphericmid-tropospheric wind difference is in wind difference is in the tropics, at 0-10Nthe tropics, at 0-10N

GEOS-5 analysesGEOS-5 analysesproduce a produce a weakerweaker easterly floweasterly flowthan NCEPthan NCEP

GEOS-5GEOS-5

NCEPNCEP

Largest Largest low-tropospheric low-tropospheric wind difference is in wind difference is in the tropics, between 10S and Equatorthe tropics, between 10S and Equator

Opposite-sign Opposite-sign discrepancydiscrepancywith respect to with respect to previous slide: previous slide: GEOS-5 analysesGEOS-5 analysesproduce produce stronger stronger easterly flow than easterly flow than NCEP)NCEP)

NCEPNCEP

GEOS-5GEOS-5

Additional vertical soundings at Cape Verde during Additional vertical soundings at Cape Verde during SOP-3 provided the chance to validate operational SOP-3 provided the chance to validate operational

analyses in 2006analyses in 2006

One of the rare cases in which One of the rare cases in which NCEPNCEP and and GEOS-5GEOS-5 differ less than 5 m/s) differ less than 5 m/s)

Both NCEP and GEOS-5 miss theBoth NCEP and GEOS-5 miss theAEJ maximum at 600hPa. AEJ maximum at 600hPa. ErrorErrorlarger than 10 m/s at AJE level!!!larger than 10 m/s at AJE level!!!

NCEPNCEP vsvs

GEOS-5GEOS-5

obsobs

obsobs

Catastrophic Catastrophic non-systematicnon-systematic differences differences NCEPNCEP provides a good representation of provides a good representation oflow-level and upper-level flows but misses low-level and upper-level flows but misses the AEJ. the AEJ. GEOS-5 GEOS-5 has huge errors at all levels has huge errors at all levels except at 600hPa.except at 600hPa.

NCEP NCEP and and GEOS-5 GEOS-5 both missboth missthe low-level flow, with the low-level flow, with NCEPNCEP having havinglarger errors.larger errors.

NCEPNCEP vsvs

GEOS-5GEOS-5

obsobs

obsobs

Catastrophic Catastrophic non-systematicnon-systematic differences differences

GEOS-5GEOS-5 produces a stronger AEJ. produces a stronger AEJ. NCEPNCEP produces a stronger AEJ. produces a stronger AEJ.

NCEPNCEP vsvs

GEOS-5GEOS-5

Huge differences between operational ECMWF, Huge differences between operational ECMWF, NCEP and MERRA over the entire tropical Pacific NCEP and MERRA over the entire tropical Pacific

during strong La Nina conditions (Aug 2010)during strong La Nina conditions (Aug 2010)

Weather prediction over the tropical Pacific is Weather prediction over the tropical Pacific is controlled by a good representation of the controlled by a good representation of the predominantly easterly flow and periodic westerly predominantly easterly flow and periodic westerly bursts along the Equatorbursts along the Equator

Large errors in the equatorial flowLarge errors in the equatorial flow propagate propagate away from the Equatoraway from the Equator affecting affecting TC genesisTC genesis prediction, and prediction, and TC track forecastTC track forecast as far as 30N/S as far as 30N/S

HugeHuge 600hPa zonal wind difference 600hPa zonal wind difference affects the affects the entireentire tropical Pacific in tropical Pacific in 20102010

Speeds are very Speeds are very comparable away comparable away from the tropics.from the tropics.

Difference ofDifference ofabout about 10m/s10m/sover Eq.Pacificover Eq.Pacific


50% speed50% speedDifference Difference Over Eq.Over Eq.PacificPacific

Opposite sign Opposite sign wind over, wind over, and NE of, and NE of, HawaiiHawaii


involving all 3 data setsinvolving all 3 data sets

The largest 600hPa wind difference at 165W occurs The largest 600hPa wind difference at 165W occurs in the tropics, between in the tropics, between 20S and 10N20S and 10N

MERRAMERRA

ECMWFECMWF

NCEPNCEP

Part IIPart II The representation of tropical cyclones The representation of tropical cyclones

in global modelsin global models

The overall forecast quality is a The overall forecast quality is a blend blend of the impacts of of the impacts of initial conditionsinitial conditions produced produced by the Data Assimilation System -and- by the Data Assimilation System -and- the forecast model capabilitythe forecast model capability

It is important to It is important to separateseparate the the intrinsic model capabilityintrinsic model capability from the from the impact of the impact of the analysisanalysis

Less-than-optimal model performanceLess-than-optimal model performance with respect to TCs with respect to TCs can be somewhat can be somewhat improved with very good TC initializationimproved with very good TC initialization

BUT BUT less-than-optimal TC initializationless-than-optimal TC initialization can be somewhat compensated by a very can be somewhat compensated by a very good model representation of the large scale forcing good model representation of the large scale forcing

What What past and currentpast and current modelsmodels can produce in can produce in `free-running mode’`free-running mode’ or inor in `̀weather-forecasting mode’weather-forecasting mode’ concerningconcerning Tropical Cyclone Tropical Cyclone verticalvertical structure,structure, scale,scale, intensity,intensity, track realism,track realism, genesis process, large-scale forcinggenesis process, large-scale forcing

Model comparison in Model comparison in forecast modeforecast mode:: NMCNMC MRF (1998), MRF (1998), NASANASA GEOS-4 (2004), GEOS-4 (2004), NCEP NCEP GFS (2004);GFS (2004); NASA NASA GEOS-5 v2 (2009) GEOS-5 v2 (2009)

Long simulations (Long simulations (so as to free the model from the memory of the ICs, can be so as to free the model from the memory of the ICs, can be performed to assess the performed to assess the intrinsic modelintrinsic model capabilities with respect to TC structure and capabilities with respect to TC structure and realism) realism) :: ECMWF ECMWF T511 Nature Run, T511 Nature Run, NASANASA GEOS-5 (2009) GEOS-5 (2009)

The problem of missing TCs in the The problem of missing TCs in the operational ANALYSIS can deteriorate the operational ANALYSIS can deteriorate the forecast of any good modelforecast of any good model

TCs in high-resolution global modelsTCs in high-resolution global models

It has been empirically noted in the operational wx forec. community that at hor. res. of 1 degree one can start seeing vertically aligned structures and an eye-like feature, at 0.5 degree the maximum winds begin to develop in the lower levels (instead of the mid-troposphere, as observed in lower resolution global models), at resolutions of few tens of kilometers global models start displaying realistic radii of maximum wind (e.g., Atlas et al., 2005; Shen et al., 2006; Reale et al., 2007),

But it takes cloud-resolving models at resolution of few kilometers to detect eye-wall replacement cycles

Accepting the limitation imposed by global models, it is interesting to follow the representation of tropical cyclones in global forecast models over the last 12 years.

Is high resolution always exploited?Is high resolution always exploited?

At any resolution, a wind speed vertical cross-section of a mature tropical cyclone should present two approximately symmetric maxima around a wind minimum.

The compactness of this eye-like feature increases with resolution but often high resolution models display structures that are much broader and more diluted than what could be expected at that resolution.

Unrealistically large eye-like features (on the order of hundreds of km, encompassing several gridpoints) are common in GCMs even when horizontal resolution is of a quarter of a degree.

The optimal, theoretical representation that should be possible at a given resolution, is NOT always reached.

It is important to perform proper diagnostics that allow to assess the quality of the representation of a TC at any given resolution

TCs in Global Operational forecasting modelsTCs in Global Operational forecasting models

tracktrack versus versus intensityintensity forecast forecast Forecast track failuresForecast track failures in in earlier global operational modelsearlier global operational models were were

generally assessed only from the point of view ofgenerally assessed only from the point of view of large-scale forcings, large-scale forcings, irrespective of how the TC structure was represented irrespective of how the TC structure was represented

In the In the latest latest global operational modelsglobal operational models forecast, forecast, structure realismstructure realism and and good forecastgood forecast tracktrack appear to be appear to be connected connected (unlike the past, where (unlike the past, where track and intensitytrack and intensity were treated as were treated as completely separatecompletely separate problems) problems)

The quality of the representation of some large-scale forcings (i.e. The quality of the representation of some large-scale forcings (i.e. ITCZ position) appear to control part of the weather forecasting scales ITCZ position) appear to control part of the weather forecasting scales involved with TC motioninvolved with TC motion

In the past (>10 years ago), In the past (>10 years ago), TC representation in global operational TC representation in global operational models was sporadic and very poor models was sporadic and very poor

Bogusing Bogusing was a necessity (now replaced by was a necessity (now replaced by vortex relocationvortex relocation))

13 years ago: Bonnie (1998) 13 years ago: Bonnie (1998)

as seen by the MRF (ancestor of NCEP GFS)as seen by the MRF (ancestor of NCEP GFS)

NMC state-of-the-art representation of TCs in 1998: no more than 25 m/s, excessively large scale (~1000km); center pressures above 1000 hPa(despite containing Hurricane Hunters flight data). TCs away from operational HH flights were often absent from analyses and forecasts.

850 hPa windSea level pressure

Bonnie (1998) cont.Bonnie (1998) cont.

MRF (NMC-now NCEP) state-of-the-art representation of TCs in 1998: no more than 25 m/s, unrealistically wide eye-like feature (r~100km); very weak warm core

TC Structure: TC Structure: NASA GEOS-4 in NASA GEOS-4 in 20042004

wind speed,temp, vort

Isidore (2002)Modeled with GEOS-4 in 2004Realistic deepening (center down to 960 hPa, unseen in any un-bogused GCMs). The NCEP Analyses confirmthe position but are not as deep withrespect to observations.

Atlas, R., O. Reale, B.-W. Shen, S.-J. Lin, J.-D. Chern, W. Putman, T. Lee, K.-S. Yeh, M. Bosilovich, and J. Radakovich, 2005: Hurricane forecasting with the high-resolution NASA finite-volume general circulation model. Geophysical Research Letters, 32, L03807, doi:10.1029/2004GL021513.

Example of a very realistic NASA GEOS4 simulationExample of a very realistic NASA GEOS4 simulationin which in which track and intensity forecasttrack and intensity forecast go side-by-side go side-by-side

Ivan (2004)Ivan (2004)

Example of hurricane vertical structure as Example of hurricane vertical structure as modeled by the GEOS-4 (2004): Ivanmodeled by the GEOS-4 (2004): Ivan

The GEOS4 could produce a The GEOS4 could produce a very compact eye-like feature very compact eye-like feature throughout the troposphere, a throughout the troposphere, a prominent warm core; wind prominent warm core; wind maxima located at about 850-maxima located at about 850-900 hPa and a radius of 900 hPa and a radius of maximum wind of about 50-100 maximum wind of about 50-100 km km

In this 66 hour forecast of In this 66 hour forecast of hurricane Ivan for hurricane Ivan for 18z12Sep2004, initalized at 18z12Sep2004, initalized at 00z10September, the 900 hPa 00z10September, the 900 hPa wind is higher than wind is higher than 55 m/s55 m/s

Wind speed, temp

– Frances (2004): early Frances (2004): early phasephase

Example of rapid Example of rapid deepening deepening andand good good forecast track forecast track despite poor despite poor analyzed intensityanalyzed intensity

One run reaches the One run reaches the correct intensity (IC: correct intensity (IC: 00z30Aug) and produces 00z30Aug) and produces the best forecast track as the best forecast track as wellwell

It takes four days for the It takes four days for the model to compensate the model to compensate the deficient initializationdeficient initialization

Realistic Cyclogenesis in a global operational model (NASA GEOS-4, 2004) WITHOUT BOGUSING

In In free-running modefree-running mode, the TCs are , the TCs are spontaneously spontaneously produced by the produced by the model without memory of ICs.model without memory of ICs.

Seasonal runs or long runs exceed forecast capability but Seasonal runs or long runs exceed forecast capability but statistical behaviorstatistical behavior of of TC activityTC activity over over months/seasons/years, and months/seasons/years, and realism of TC structurerealism of TC structure can be can be inferredinferred

TC activity is controlled only by TC activity is controlled only by global forcingsglobal forcings (SST) (SST) TC cyclogenesis and structureTC cyclogenesis and structure are produced by the model are produced by the model

alone alone without any contribution of Initial Conditions without any contribution of Initial Conditions These areThese are good tests to assess the intrinsic model good tests to assess the intrinsic model

capability with respect to TC processescapability with respect to TC processes

T511 ECMWF Nature Run (2007)T511 ECMWF Nature Run (2007) Free running model – Free running model – no memory of initial no memory of initial

conditions – no additional dataconditions – no additional data A long simulation is the only way to assess the A long simulation is the only way to assess the

capability of a forecasting model – as opposed as capability of a forecasting model – as opposed as a DAS+forecasting model. a DAS+forecasting model. No bogusing,No bogusing, vortex vortex relocationrelocation, , targeted obs, targeted obs, can be added.can be added.

13-month run, initialized May 200513-month run, initialized May 2005 Only SST (2005) and Sea-Ice as boundary Only SST (2005) and Sea-Ice as boundary

forcingsforcings Analysis published in Reale et al. (2007)Analysis published in Reale et al. (2007)

Reale, O., J. Terry, M. Masutani, E. Andersson, L. P. Riishojgaard, J. C. Jusem,2007: Preliminary evaluation of the European Centre for Medium-Range Weather Forecasts (ECMWF) Nature Run over the Tropical Atlantic and African Monsoon region.Geophysical Research Letters, 34, L22810, doi:10.1029/2007GL31640.

EC T511 NR: realistic activity (9 strong TCs)EC T511 NR: realistic activity (9 strong TCs)From Reale et al. (2007) GRLFrom Reale et al. (2007) GRL

EC T511: Realistic Variability of Atl. TC tracksEC T511: Realistic Variability of Atl. TC tracks

Looping andBinary vortex interaction

4 systems:Looping,Binary vortex Interaction,ExtratropicalTransitionsand Extra-tropicalRe-intensification

Singuarities, binary vortexInteractions, Intensity fluctuationsDue to large-scale forcing fluctuations

A long simulation must produce complex tracks

GEOS-5 with Stocastic Tokioka (2009)GEOS-5 with Stocastic Tokioka (2009)Simulations by Myong-In Lee, PI S. Schubert (NASA): Simulations by Myong-In Lee, PI S. Schubert (NASA):

Same experiment settings of ECMWF Nature Run Same experiment settings of ECMWF Nature Run Behavior comparable to the EC T511Behavior comparable to the EC T511

Control Run GEOS-5 0.25(with rel. Arakawa-Schubert) GEOS-5 (with Tokioka)

No cyclone reaches 1000hPa in the Control during September. At least 7 cyclones below 1000 hPa in the GEOS-5 w.Tok. One hurricane

goes below 960hPa. Very realistic track variability, scale. Even non-developing waves are well captured.

EC T511: Multiple simultaneous tropical cyclones can be EC T511: Multiple simultaneous tropical cyclones can be present in the Atlantic in very active seasonspresent in the Atlantic in very active seasons

Another important –realistic- capability of the ECMWF NRAnother important –realistic- capability of the ECMWF NR

500 hPa geop (m) and 900 hPa rel vort (s-1)

3 TCs simultaneously present in the 3 TCs simultaneously present in the GEOS-5 w. Tokioka 11SepGEOS-5 w. Tokioka 11Sep

Simulations by Myong-In Lee, PI S. Schubert (NASA): Simulations by Myong-In Lee, PI S. Schubert (NASA):

Slp (hPa) and 925 hPa wind (m/s)

Intensity Intensity In the operational forecasting environment, In the operational forecasting environment, 10m observed 10m observed

wind and center pressure are currently usedwind and center pressure are currently used PROBLEMPROBLEM: excessively : excessively high draghigh drag in the marine boundary in the marine boundary

layer seems to occur in global models when winds exceed layer seems to occur in global models when winds exceed 30m/s: 10m wind often about 60% of the 850hPa wind 30m/s: 10m wind often about 60% of the 850hPa wind (unlike 90% in real world)(unlike 90% in real world)

Possibly due to Possibly due to unrealistically high roughness lengthunrealistically high roughness length over over oceans with wind speeds exceeding 30m/soceans with wind speeds exceeding 30m/s

As a consequence, it may better to use As a consequence, it may better to use 850hPa or 900hPa 850hPa or 900hPa windwind as as intensityintensity diagnostics in global modelsdiagnostics in global models

One simple way of assessing comprehensively the TC One simple way of assessing comprehensively the TC intensity reached in a simulation is to produce the max intensity reached in a simulation is to produce the max wind at 850hPa throughout the system’s lifespanwind at 850hPa throughout the system’s lifespan

Example of Intensity inferred from 850hPa Example of Intensity inferred from 850hPa wind max (Isabel, 2003)wind max (Isabel, 2003)

Operational GFS and GEOS-4 have comparable intensity Different degree of compactness

A possible metrics to assess how well the A possible metrics to assess how well the horizontal scale is representedhorizontal scale is represented

Horizontal Compactness, ratio of radius of maximum wind (rmw) over radius of wind greater than the environmental wind of a given threshold, which we can consider the radius of the tropical cyclone (TC) in the model (rtc).

The wind magnitude of a modeled tropical cyclone decreases from the center and is not distinguishable from the large-scale wind at a certain distance. This distance could be considered the tc-influenced domain in the model and can be compared with the rmw. The smaller rmw with respect to the rtc the more realistic the modeled cyclone is. In low-resolution global models, the radius of maximum wind occupies a large fraction of the domain affected by the cyclone.

Example from older GEOS-5 v.2: Example from older GEOS-5 v.2: how how compactcompact is this 0.5 simulation of Helene? is this 0.5 simulation of Helene?

Compactness evaluated in GEOS-5 simulation at .5 Compactness evaluated in GEOS-5 simulation at .5

for for Helene (2006)Helene (2006)

[RMW(l)+RMW(r)] / [RTC(l)+RTC(r)]=0.27

Despite being a relativelyweak simulation, the representation of thesystem is quite compact in the above sense

850 hPa wind at 18.5N

Compactness in the GEOS-5 w. Tokioka at 0.25Compactness in the GEOS-5 w. Tokioka at 0.25a much better rpresentation of a TCa much better rpresentation of a TC

[RMW(l)+RMW(r)] / [RTC(l)+RTC(r)]=0.07

RTC(l) RTC(r)

RMW

At ~60W, a RAINBAND

Very clear evidence of a rainband at 61WVery clear evidence of a rainband at 61W

RAINBAND

Warm Core StructureWarm Core Structure

One immediate, effective way of assessing if a model produces a vertically aligned and symmetric system, is to measure the strenght of its warm core. One way is simply to subtract a standardized zonal mean intersecting the center of the storm. Examples: GEOS-5 versus NCEP GFS

Examples of warm core (Helene, 2006)Examples of warm core (Helene, 2006)

GEOS-5 (0.25)

GFS GFS

GEOS-5 (0.25)

48-h Fc

48-h Fc

72-h Fc

72-h Fc

Ms. M. Fuentes, Ph.D. Thesis

Vertical Structure inferred through zonal Vertical Structure inferred through zonal and meridional vertical cross-sections of and meridional vertical cross-sections of wind speed and temperature of mature wind speed and temperature of mature

TCs in the deep tropics TCs in the deep tropics Desirable features:Desirable features:

Wind maximum at 900hPa or lowerWind maximum at 900hPa or lower Small radius of maximum wind Small radius of maximum wind Perfectly vertically aligned low-speed columnPerfectly vertically aligned low-speed column Vorticity column with maximum in the lower levelsVorticity column with maximum in the lower levels Low-level convergence confined below 800 hPaLow-level convergence confined below 800 hPa Upper-level divergence confined above 200 hPaUpper-level divergence confined above 200 hPa

Side by side comparison Side by side comparison ECT511 vs GEOS5 with TokiokaECT511 vs GEOS5 with Tokioka

GEOS runs by Myong-In Lee, PI S. Schubert (NASA): GEOS runs by Myong-In Lee, PI S. Schubert (NASA):

EC T511 (2007) GEOS-5 with Tokioka (2009)

GEOS-5 has slightly sharper warm core, better-defined eye, max wind at lowerelevation, slightly smaller radius of max wind. Intensity is about the same.

Zonal MeridionalMeridional Zonal

Hurricane in the AtlanticHurricane in the Atlantic (GEOS-5 long simulation w. Tok (GEOS-5 long simulation w. Tok

by Myong-In Lee)by Myong-In Lee)

Warm core up to 10C!

Winds up to 60m/s

Vorticity up to 3x10-3s-1

Summary of TC features that can be seen in Summary of TC features that can be seen in operational global models operational global models

((GEOS-4, GFS, ECMWF T511GEOS-4, GFS, ECMWF T511)) observations,observations, 0.250.25 GEOS-5 with TokiokaGEOS-5 with Tokioka

Horizontal scale ~wind speed comparable to the large-Horizontal scale ~wind speed comparable to the large-scale environment (300-1000 km;scale environment (300-1000 km; 250-1000 km; 250-1000 km; 300-300-1000km1000km))

Radius of max. wind (50-300 km; Radius of max. wind (50-300 km; 20-100km; 20-100km; 50-100km50-100km) ) Low-level vorticity (10^-3Low-level vorticity (10^-3 s-1, s-1, 3x10^-3 s-13x10^-3 s-1 )) 850 hPa wind: above 60 m/s, 850 hPa wind: above 60 m/s, above 100m/sabove 100m/s, , aboveabove 60m/s60m/s Warm core: 2-10 C; Warm core: 2-10 C; 6-14 C; 6-14 C; 4-12C4-12C Horizontal compactness: 0.07-0.35; Horizontal compactness: 0.07-0.35; 0.05-0.15; 0.05-0.15; 0.07-0.200.07-0.20

However, despite the current capability of global However, despite the current capability of global models, state-of-the-art operational analyses can models, state-of-the-art operational analyses can

completely misscompletely miss existingexisting Tropical Cyclones. Tropical Cyclones.

Analyses are particularly deficient in the depiction Analyses are particularly deficient in the depiction of of developing, deepeningdeveloping, deepening and and transitioning transitioning tropical cyclonestropical cyclones

Analyses are deficient in representing Analyses are deficient in representing cyclogenesis and existing deepening cyclones in cyclogenesis and existing deepening cyclones in the the Eastern AtlanticEastern Atlantic

Analyses are Analyses are particularly deficientparticularly deficient in representing in representing even even fully-developed fully-developed TCs over the TCs over the Indian OceanIndian Ocean

TS Debby (2006) at 06z 24 Aug 2006TS Debby (2006) at 06z 24 Aug 2006Obs center slp 999 hPa; Max wind 22 m/sObs center slp 999 hPa; Max wind 22 m/s

NCEP analysesNCEP analysesdo do notnot produce produce

a a closed closed circulationcirculation

GEOS-5 An.GEOS-5 An.200km200km

displacement displacement error for centererror for center(obs. center X) (obs. center X)

Wind speed m/sWind speed m/s

Conclusions (Part I and II)Conclusions (Part I and II) State of the art reanalyses State of the art reanalyses (ERA-40, JRA-25, NCEP-R2 and (ERA-40, JRA-25, NCEP-R2 and

MERRA)MERRA) show susbtantial differences in the show susbtantial differences in the seasonally-seasonally-averaged averaged representation of the representation of the African Easterly JetAfrican Easterly Jet and more and more generally of the circulation in the African Monsoon and tropical generally of the circulation in the African Monsoon and tropical Atlantic regionsAtlantic regions

Operational analyses or reanalyses differ also at Operational analyses or reanalyses differ also at instantaneous timesinstantaneous times in the tropical region. On the contrary, in the tropical region. On the contrary, away from the tropics, different analyses provide almost away from the tropics, different analyses provide almost identical representations of the wind fieldidentical representations of the wind field

Despite changes in models and assimilation systems, and Despite changes in models and assimilation systems, and increase in resolution, increase in resolution, the representation of wind in the tropicsthe representation of wind in the tropics does not show much improvementdoes not show much improvement from from 2006 to 20102006 to 2010

Major deficiencies appear on Major deficiencies appear on all 3 basins:all 3 basins: Atlantic, Indian and Atlantic, Indian and Pacific OceansPacific Oceans on scales spanning from on scales spanning from storm-scalestorm-scale to to planetaryplanetary, from , from weatherweather to to seasonalseasonal

Outline Part IIIOutline Part IIIImprovements stemming Improvements stemming

out of use of AIRSout of use of AIRS AIRS impactAIRS impact on on midlatitude winter dynamicsmidlatitude winter dynamics Global AIRS impactsGlobal AIRS impacts in boreal in boreal winterwinter, , springspring, , summersummer andand

fall conditionsfall conditions in in fivefive different years different years TC analysis.TC analysis. Improvement in tropical cyclone position and Improvement in tropical cyclone position and

structurestructure, , leading to improved leading to improved forecast track, forecast track, overover all all basinsbasins

AIRS impactAIRS impact on on tropical cyclone Nargistropical cyclone Nargis AIRS impactAIRS impact on on tropical cyclones in the Atlantic tropical cyclones in the Atlantic AIRS impact AIRS impact onon cyclogenesis cyclogenesis AIRS impact AIRS impact onon extra tropical transitions extra tropical transitions AIRS ImpactAIRS Impact on on Precipitation Analyses and ForecastsPrecipitation Analyses and Forecasts

Understanding and improving the impact of AIRS in the Understanding and improving the impact of AIRS in the GEOS-5 Data Assimilation and Forecasting SystemGEOS-5 Data Assimilation and Forecasting System

Previously published work (Reale et al., 2008) has shown Previously published work (Reale et al., 2008) has shown substantial improvement in analysis and forecasts over the substantial improvement in analysis and forecasts over the northern hemisphere northern hemisphere extratropicsextratropics in boreal winter in boreal winter conditionsconditions, due to an improved representation of the , due to an improved representation of the lower-mid tropospheric thermal structure in the high lower-mid tropospheric thermal structure in the high latitudes, and consequently anlatitudes, and consequently an improved improved polar vortexpolar vortex..

The improvement comes from the assimilation of The improvement comes from the assimilation of quality-quality-controlled AIRS retrievals obtained under partially cloudy controlled AIRS retrievals obtained under partially cloudy conditionsconditions

Reale, O., J. Susskind, R. Rosenberg, E. Brin, E. Liu, L.P. Riishojgaard, Reale, O., J. Susskind, R. Rosenberg, E. Brin, E. Liu, L.P. Riishojgaard, J. Terrry, J.C. Jusem, 2008: Improving forecast skill by assimilation of J. Terrry, J.C. Jusem, 2008: Improving forecast skill by assimilation of quality-controlled AIRS temperature retrievals under partially cloudy conditions. quality-controlled AIRS temperature retrievals under partially cloudy conditions. Geophys. Res. Lett., 35, L08809, doi: 10.1029/2007GL033002Geophys. Res. Lett., 35, L08809, doi: 10.1029/2007GL033002

3 sets of 27 5-day forecasts3 sets of 27 5-day forecasts, initialized at 00Z each day, , initialized at 00Z each day, from 1/5/03 through 1/31/03:from 1/5/03 through 1/31/03:

– ““CNTRLCNTRL” set initialized from the control assimilation, ” set initialized from the control assimilation,

where all operational data (conventional and satellite) where all operational data (conventional and satellite) are ingested except for AIRSare ingested except for AIRS

– ““AIRSAIRS” set initialized from the assimilation in which ” set initialized from the assimilation in which AIRS cloudy retrievals are ingested in addition to all AIRS cloudy retrievals are ingested in addition to all data used by the CNTRLdata used by the CNTRL

– ““CUTFCUTF” set initialized from the assimilation in which ” set initialized from the assimilation in which AIRS retrievals are ingested only above 200hPaAIRS retrievals are ingested only above 200hPa

All three sets verified against NCEP analysisAll three sets verified against NCEP analysis

RESULTS:RESULTS:

500hPa geopotential height 500hPa geopotential height anomaly correlationanomaly correlation (AC) in the (AC) in the Northern hemisphere extra-Northern hemisphere extra-tropics of the average of all 27 tropics of the average of all 27 forecasts verif. against NCEP forecasts verif. against NCEP analysis as a function of analysis as a function of forecast time. forecast time.

The day-5The day-5 500mb geopotential 500mb geopotential height anomaly correlation height anomaly correlation (AC5) for each of the 27 (AC5) for each of the 27 forecastsforecasts

AIRS forecasts demonstrated AIRS forecasts demonstrated superior skill over both the superior skill over both the CNTRL and CUTF in most of CNTRL and CUTF in most of the cases. Case 21 (init. 25 the cases. Case 21 (init. 25 Jan) in which the CNTRL AC5 Jan) in which the CNTRL AC5 is high (.85) and the AIRS is high (.85) and the AIRS forecast produces further forecast produces further significant improvement significant improvement

800hPa temperature anomaly 800hPa temperature anomaly (AIRS minus CNTRL) analysis(AIRS minus CNTRL) analysis (00Z 25 Jan). Notice the (00Z 25 Jan). Notice the large large area of negative temperature area of negative temperature impact impact over over northeastern Siberia, northeastern Siberia, Alaska and the Arctic regionAlaska and the Arctic region. .

Analyzed emperature profiles, area-Analyzed emperature profiles, area-averagedaveraged over the entire Arctic region over the entire Arctic region (70-90N(70-90N), from 1000 to 100mb at the ), from 1000 to 100mb at the initial forecast time (00Z 25 Jan) of:initial forecast time (00Z 25 Jan) of:

CNTRL = blackCNTRL = blackAIRS = greenAIRS = greenCUTF = redCUTF = red

and the and the AIRS minus CNTRLAIRS minus CNTRL temperature temperature difference profile (orange).difference profile (orange).

The assimilation of AIRS The assimilation of AIRS cloudy cloudy retrievals in the lower-mid troposphere retrievals in the lower-mid troposphere results in results in significantly colder significantly colder temperatures between 950 and 700mbtemperatures between 950 and 700mb, , with a peak at about 875mb.with a peak at about 875mb.

From Reale et al. (2008)From Reale et al. (2008)

500hPa geopotential height 500hPa geopotential height anomaly (AIRS minus CNTRL) anomaly (AIRS minus CNTRL) at 00Z 25 Jan (analyses). at 00Z 25 Jan (analyses).

The hydrostatic adjustment induced by lower temperatures causes the 500mb geopotential in the AIRS case to drop substantially, modifying drastically the structure of the polar vortex.

From Reale et al. (2008)

The initial negative anomaly The initial negative anomaly over Siberia and Alaska, over Siberia and Alaska, appears as a appears as a wave packet wave packet undergoing dispersion,undergoing dispersion, amplifying and propagating amplifying and propagating eastwardeastward . The . The AIRS minus AIRS minus CNTRLCNTRL anomaly observed at anomaly observed at day 5 over Canada and the day 5 over Canada and the north Atlantic corresponds north Atlantic corresponds well with the well with the NCEP AN. NCEP AN. minus CNTRLminus CNTRL in the same in the same region. region.

From Reale et al. (2008)From Reale et al. (2008)

Latitudinally averaged (40-80N) 500 hPa geopotential height Latitudinally averaged (40-80N) 500 hPa geopotential height anomaly (AIRS minus CNTRL, shaded, and NCEP minus CNTRL, anomaly (AIRS minus CNTRL, shaded, and NCEP minus CNTRL, solid black line) as a function of forecast time.solid black line) as a function of forecast time.

Global Impact of Global Impact of Clear-sky RadiancesClear-sky Radiances versus versus

Quality Controlled cloudy RetrievalsQuality Controlled cloudy Retrievals

A small fraction of AIRS data is still retained in operational A small fraction of AIRS data is still retained in operational weather systems, where the only AIRS data assimilated are weather systems, where the only AIRS data assimilated are radiance observations of channels unaffected by clouds. radiance observations of channels unaffected by clouds. This This imposes a severe limitation on the horizontal distribution of the imposes a severe limitation on the horizontal distribution of the data.data.

Susskind (2007, 2010) strategy, based upon previous work by Susskind (2007, 2010) strategy, based upon previous work by Chahine, allows improvement of soundings in partly-cloudy Chahine, allows improvement of soundings in partly-cloudy conditions: a key element is the ability conditions: a key element is the ability to generate case-by-to generate case-by-case and level-by-level error estimates and use them for case and level-by-level error estimates and use them for quality controlquality control

A very large number of experiments were produced, comparing A very large number of experiments were produced, comparing AIRS retrievals and radiances in AIRS retrievals and radiances in all seasons,all seasons, five different five different yearsyears, with , with different quality controlsdifferent quality controls, looking at both , looking at both global global impactsimpacts and and individual individual high-impacthigh-impact weather systems weather systems

AIRS Experiments settingsAIRS Experiments settings

– GEOS-5 DAS: versions GEOS-5 DAS: versions 2.0.2, 2.1.2, 2.1.42.0.2, 2.1.2, 2.1.4– Control assimilation (Control assimilation (CNTRLCNTRL): assimilating all conventional ): assimilating all conventional

and satellite data, but no AIRS retrievals, from and satellite data, but no AIRS retrievals, from 8/10/06 to 8/10/06 to 9/15/20069/15/2006 (NAMMA), (NAMMA), 10/15/2005 to 11/15/2005 10/15/2005 to 11/15/2005 (Active TC (Active TC Atlantic season), Atlantic season), 4/15/2008 to 5/15/20084/15/2008 to 5/15/2008 (Nargis), (Nargis), 7/1/2010 7/1/2010 to 8/15/2010to 8/15/2010 (Pakistan floods) (Pakistan floods)

– AIRS ``standard’’ QC RETAIRS ``standard’’ QC RET: Same data as control plus AIRS : Same data as control plus AIRS version 5 retrievals with “standard” quality control added as version 5 retrievals with “standard” quality control added as rawinsonde temperature profiles.rawinsonde temperature profiles.

– AIRS ``medium’’ QC RETAIRS ``medium’’ QC RET: More restrictive QC for AIRS ret: More restrictive QC for AIRS ret– AIRS ``tight’’ QC RETAIRS ``tight’’ QC RET: Most restrictive QC for AIRS ret: Most restrictive QC for AIRS ret– AIRSAIRS RADRAD: AIRS clear-sky radiances from NESDIS: AIRS clear-sky radiances from NESDIS– Forecasts at 0.25 and/or 0.5 degreesForecasts at 0.25 and/or 0.5 degrees

GEOS-5 2.0.2 GEOS-5 2.0.2 Boreal SpringBoreal Spring Conditions: Conditions:global impactglobal impact of cloudy retrievals (tight QC) vs. of cloudy retrievals (tight QC) vs.

clear-sky radiancesclear-sky radiancesPositive Positive globalglobal impact of impact of AIRS retrievals (red).AIRS retrievals (red).

Negative impact of AIRS Negative impact of AIRS clear-sky radiances (green).clear-sky radiances (green).

In addition, representation of In addition, representation of individual weatherindividual weather systemssystems in the tropics are strongly in the tropics are strongly impacted by AIRS. impacted by AIRS.

Anomaly Correlations computed from 90S to 90N

GEOS-5 2.0.2 GEOS-5 2.0.2 Boreal SummerBoreal Summer Conditions: Conditions:global impactglobal impact of cloudy retrievals (tight QC) vs. of cloudy retrievals (tight QC) vs.

clear-sky radiancesclear-sky radiancesStrong Strong globalglobal impact of impact of AIRS retrievals (red).AIRS retrievals (red).

Smaller impact of AIRS Smaller impact of AIRS clear-sky radiances (green).clear-sky radiances (green).



GEOS-5 2.0.2 GEOS-5 2.0.2 Boreal FallBoreal Fall Conditions: Conditions:global impactglobal impact of cloudy retrievals of cloudy retrievals (tight QC)(tight QC) vs. vs.

clear-sky radiancesclear-sky radiancesStrong Positive Strong Positive globalglobal impact of impact of AIRS retrievals (red).AIRS retrievals (red).

Smaller positive impact of AIRS Smaller positive impact of AIRS clear-sky radiances (green).clear-sky radiances (green).



In addition to global skill, AIRS affects In addition to global skill, AIRS affects the the depiction of tropical weather systemsdepiction of tropical weather systems

AIRS cloudy retrievals change particularly the depiction of AIRS cloudy retrievals change particularly the depiction of developingdeveloping and and transitioning transitioning tropical cyclonestropical cyclones

AIRS impact on Tropical Cyclones in the GEOS-5 has AIRS impact on Tropical Cyclones in the GEOS-5 has been studied over the been studied over the Atlantic, Indian and Pacific OceansAtlantic, Indian and Pacific Oceans

AIRS improves the Tropical Cyclone ANALYSIS in AIRS improves the Tropical Cyclone ANALYSIS in GEOS5-DAS in terms of GEOS5-DAS in terms of intensityintensity, , confinementconfinement and and positionposition

The cause of the improvement arises from tight, strongThe cause of the improvement arises from tight, strong upper-tropospheric positive thermal anomalies upper-tropospheric positive thermal anomalies detected detected overover organized convection organized convection

No or minimal improvementNo or minimal improvement derives from the assimilation derives from the assimilation of of clear-sky radiancesclear-sky radiances

Published study on the impact of AIRS, Published study on the impact of AIRS, focused on a focused on a particulaly difficultparticulaly difficult tropical cyclone: tropical cyclone:

Nargis (2008)Nargis (2008) Work published in 2009 shows improvements in analysis over the Work published in 2009 shows improvements in analysis over the tropicstropics

in in the GEOS-5 DAS and forecasting model consequent to assimilation in in the GEOS-5 DAS and forecasting model consequent to assimilation of AIRS-derived information in of AIRS-derived information in CLOUDY CLOUDY areas. Case chosen: areas. Case chosen: catastrophic cyclone Nargis which hit Burma causing devastating loss of catastrophic cyclone Nargis which hit Burma causing devastating loss of lifelife

Tropical Cyclones in the Northern Indian Oceans are extremely difficult Tropical Cyclones in the Northern Indian Oceans are extremely difficult to predict because of to predict because of shorter lifespanshorter lifespan and and erratic trackserratic tracks

Operational global analyses often do not represent these cyclones’ Operational global analyses often do not represent these cyclones’ position position (or (or even the TCs’ very existenceeven the TCs’ very existence) accurately) accurately partly because of partly because of strongly asymmetric data distribution geometrystrongly asymmetric data distribution geometry

Forecasts are particularly penalized by analysis errors.Forecasts are particularly penalized by analysis errors.

Reale, O., W. K. Lau, J. Susskind, R. Rosenberg, E. Brin, E. Liu, L.P. Riishojgaard, M. Reale, O., W. K. Lau, J. Susskind, R. Rosenberg, E. Brin, E. Liu, L.P. Riishojgaard, M. Fuentes, R. Rosenberg, 2009: AIRS impact on the analysis and forecast track of tropical Fuentes, R. Rosenberg, 2009: AIRS impact on the analysis and forecast track of tropical cyclone Nargis in a global data assimilation and forecasting system. cyclone Nargis in a global data assimilation and forecasting system. Geophys. Res. Lett., 36, L06812, doi: 10.1029/2008GL037122Geophys. Res. Lett., 36, L06812, doi: 10.1029/2008GL037122

Complete Complete miss of TC Nargismiss of TC Nargis (2008) in both (2008) in both operational NCEPoperational NCEP and and MERRAMERRA analyses at a analyses at a time when is declared having time when is declared having hurricane-level hurricane-level

windswinds by the JTPC and IMC by the JTPC and IMC

800x600km800x600kmContours Contours every 1hPaevery 1hPa

WINDS DOWINDS DONOT FORMNOT FORMA CLOSEDA CLOSEDCIRCULATIONCIRCULATION

800x600km800x600kmContours Contours every 1hPaevery 1hPa

X observed X observed cyclone’s center cyclone’s center

COMPLETELYCOMPLETELYFLAT PRESSUREFLAT PRESSURE

FIELDFIELD

WINDS DO NOT REACH 12m/sWINDS DO NOT REACH 12m/sWINDS DO NOT FORM A CLOSED CIRCULATIONWINDS DO NOT FORM A CLOSED CIRCULATION

Spectacular forecast track improvement for Spectacular forecast track improvement for tropical cyclone Nargis (2008) consequent to tropical cyclone Nargis (2008) consequent to

qc-ed AIRS qc-ed AIRS cloudycloudy retrieval assimilation retrieval assimilation

Control AIRS clear-sky radiances AIRS cloudy retrievals

5 out of 7 forecasts initialized from the improved analyses 5 out of 7 forecasts initialized from the improved analyses have a displacement error at landfall ofhave a displacement error at landfall of about 50km about 50km

(Reale et al., 2009, (Reale et al., 2009, Geophys. Res. LettGeophys. Res. Lett.).)Assimilation of clear-sky radiances produce minimal improvement

Improvement with AIRS Improvement with AIRS cloudycloudy retrievals retrievals

Analysis obtained Analysis obtained assimilating AIRS assimilating AIRS cloudycloudy retrievals retrievalsWell-definedWell-definedCycloneCycloneGreen:Green:Observed Observed TrackTrack

108-hour108-hourforecast (slp)forecast (slp)initialized from initialized from improved improved analysesanalyses

Green:Green:Observed Observed TrackTrack

CNTRL Analysis (above)CNTRL Analysis (above)And forecast (below): And forecast (below): No CycloneNo Cyclone

Accurate landfall is produced in the forecasts initialized Accurate landfall is produced in the forecasts initialized with AIRS: (Reale et al., 2009, with AIRS: (Reale et al., 2009, Geophys. Res. LettGeophys. Res. Lett.).)

Why AIRS radiances do Why AIRS radiances do not not impact impact the forecast for NARGIS?the forecast for NARGIS?

There are simply NO DATA accepted by the DAS in the area where NARGISdevelopes, because the measurements are in cloudy areas.

USED REJECTED

QC-ed AIRS cloudy retrievals provide QC-ed AIRS cloudy retrievals provide substantial coverage over the areasubstantial coverage over the area

The temperature information provided by cloudy AIRS retrievals whereThe temperature information provided by cloudy AIRS retrievals wherethe storm is developing leads to improved analyses and forecaststhe storm is developing leads to improved analyses and forecasts

How AIRS retrievals improve the analysis of a TC?How AIRS retrievals improve the analysis of a TC?

Shaded: 200 hPa AIRS minus CNTRL temp anomalyShaded: 200 hPa AIRS minus CNTRL temp anomalyContour: AIRS minus CNTRL slp anomaly Contour: AIRS minus CNTRL slp anomaly (Reale et al., 2009)(Reale et al., 2009)

The localized, intenseThe localized, intenseUpper-Level heatingUpper-Level heatinginduced by AIRS datainduced by AIRS datain correspondence to in correspondence to organized convection organized convection deepensdeepens the the low-level low-level cyclonic circulationcyclonic circulation of of TC NargisTC Nargis

AIRS impact study in AIRS impact study in boreal summerboreal summer conditionsconditions

AIRS improves the representation of the thermal structure AIRS improves the representation of the thermal structure of the atmosphere in the tropicsof the atmosphere in the tropics

In particular, developing tropical lows are better defined In particular, developing tropical lows are better defined and confined with the ingestion of AIRS temperature and confined with the ingestion of AIRS temperature retrievals under partly cloudy conditionsretrievals under partly cloudy conditions

The improvement consists of a) more confined and tight The improvement consists of a) more confined and tight circulations b) more accurate center locationscirculations b) more accurate center locations

Experiments covering the Experiments covering the NAMMA SOP-3 periodNAMMA SOP-3 period (15Aug-(15Aug-15Sep 2006)15Sep 2006) to investigate TC representation in the to investigate TC representation in the Atlantic in response to AIRS data ingestionAtlantic in response to AIRS data ingestion

AIRS TIGHTAIRS TIGHT QC CLOUDY RET improves TC QC CLOUDY RET improves TC position in the position in the AnalysisAnalysis of Helene (2006) of Helene (2006)

Slp RAD analysis (contour)RAD slp impact (shaded)X: observedHelene position

300 hPa TempImpact (ret minus Control, shaded)And slp impact(contour)

Slp RET analysis (contour)RETRIEVALS slp impact (shaded)

AIRS TIGHT RET produces a PERFECTposition for Helene and a deeper storm

Large improvement in the Large improvement in the forecastforecast of Hurricane of Hurricane Helene’s Helene’s genesisgenesis with `tight QC’ cloudy with `tight QC’ cloudy

retrievalsretrievals

Forecasts from Analysis in which AIRS TIGHT RET are assimilated improve Helene’sFormation as a hurricane (12z 16Sep). Improvement is minimal in RAD case

ComparisonOf 36-hForecastsof AIRS TIGHT RET(lower left)with AIRS RAD(upper right)

AIRS impact on AIRS impact on extra-tropical extra-tropical transitionstransitions: a difficult problem.: a difficult problem.

Rapid changes in dynamics from tropical to baroclinicRapid changes in dynamics from tropical to baroclinic Very strong asymmetric vertical shear Very strong asymmetric vertical shear Rapid acceleration of the systemsRapid acceleration of the systems Small errors in TC locationSmall errors in TC location before transition lead to before transition lead to large large

forecast track errorsforecast track errors Small errors in the thermal structureSmall errors in the thermal structure of the atmosphere of the atmosphere

before transition lead to before transition lead to large misrepresentations of storm large misrepresentations of storm intensityintensity

AIRS quality-controlled cloudy retrievals positively impact AIRS quality-controlled cloudy retrievals positively impact EXTRA TROPICAL TRANSITIONSEXTRA TROPICAL TRANSITIONS

Extra-tropical transition of Hurricane Florence Extra-tropical transition of Hurricane Florence (2006): (2006): improving the analysisimproving the analysis beforebefore

transition transition 06z10Sep200606z10Sep2006

400 hPa AIRS RET-inducedTemp anomaly (shaded)And impact on slp (contour)

RAD slp analysis (contour)And difference from CNTRL.. Negative impact From assimilation of clear-skyRadiances.

AIRS RET slp analysis (contour)and difference from CNTRL (shaded)

AIRS RET improves location and intensity also at subsequent times; FORECASTS from the improved analyses are much superior.

Predicting the Predicting the ET intensificationET intensification of Florence from of Florence from improved (before ET) analysesimproved (before ET) analyses

The 60h forecastInitialized fromAnalyses in whichAIRS retrievalsare assimilated produce a deeperafter ET-cyclone with respect tothe RAD case,in agreement with observations.

Improvement in Improvement in cloud structurecloud structure caused by caused by AIRS cloudy retrievalsAIRS cloudy retrievals

TS Helene Analysis at 06z 15Sep2006TS Helene Analysis at 06z 15Sep200630 hours before becoming a hurricane30 hours before becoming a hurricane

800 hpa relative humidity, sea level pressure (hPa)

CNTRL RADIANCE

RETRIEVALSDisplay an

Eye-like feature

NCEP OperationalAnalyses,Very poor

The The 36-hour 36-hour forecastforecast initialized from analyses in which AIRS initialized from analyses in which AIRS retrievals are assimilated is the only one that produce an retrievals are assimilated is the only one that produce an eye,eye,

a a closed circulationclosed circulation, and a , and a reasonable scalereasonable scale Helene at 12z 16Sep2006, upgraded to hurricaneHelene at 12z 16Sep2006, upgraded to hurricane

CNTRL RAD

RETRIEVALSClear eye-likefeature

NCEP analysesToo broad wrt to obs

850 hpa relative humidity, sea level pressure (hPa)

AIRS Impact on precipitation analysis and AIRS Impact on precipitation analysis and forecastforecast

Weather-produced precipitation is generally poorly Weather-produced precipitation is generally poorly predicted by global models in the tropics.predicted by global models in the tropics.

In addition to the problems of convective parametrizations, In addition to the problems of convective parametrizations, model resolution and physics improvements, model resolution and physics improvements, a more a more accurate thermal representation of the tropical atmosphereaccurate thermal representation of the tropical atmosphere can produce better precipitationcan produce better precipitation

While the While the next-generation improvement must come from next-generation improvement must come from direct precipitation assimilation,direct precipitation assimilation, some benefit also arise some benefit also arise from AIRS temperature and moisture profile assimilation from AIRS temperature and moisture profile assimilation obtained under cloudy conditions.obtained under cloudy conditions.

In the GEOS-5 experiments here described, the In the GEOS-5 experiments here described, the `precipitation analysis’`precipitation analysis’ described does described does notnot come from come from precipitation assimilation but from a set of very short term precipitation assimilation but from a set of very short term forecasts strongly constrained by observations (corrector forecasts strongly constrained by observations (corrector sequence)sequence)

Precipitation `ànalysis’’ for Precipitation `ànalysis’’ for Helene (2006)Helene (2006)

Not a true precipitation analysis since no Not a true precipitation analysis since no precip data are assimilated. Precip comes precip data are assimilated. Precip comes

from the `corrector sequence’ and is from the `corrector sequence’ and is essentially a set of very short term essentially a set of very short term forecasts strongly constrained by forecasts strongly constrained by

observations.observations.The assimilation containing AIRS The assimilation containing AIRS

retrievals, besides improving Helene’s retrievals, besides improving Helene’s structure, also produces thestructure, also produces the best best

accumulated precipitationaccumulated precipitation

Zhou, Y., W. K. Lau, O. Reale, R. Zhou, Y., W. K. Lau, O. Reale, R. Rosenberg, 2010: AIRS Impact on Rosenberg, 2010: AIRS Impact on

precipitation analysis and forecast of precipitation analysis and forecast of tropical cyclone in a global data tropical cyclone in a global data

assimilation and forecasting system. assimilation and forecasting system. Geophys. Res. Lett.,Geophys. Res. Lett., 37,37, L02806, L02806,

doi.1029/2009GL041494doi.1029/2009GL041494

OBS

CNTRL

RAD

RET (st)

RET (t)

Precipitation Forecast for Precipitation Forecast for HeleneHelene

Precipitation forecast computed along Precipitation forecast computed along track and validated with SSM/I data. track and validated with SSM/I data. Ingestion of AIRS retrievals cause the Ingestion of AIRS retrievals cause the

GEOS-5 to have best skillGEOS-5 to have best skill. Improvement . Improvement with respect of CNTRL caused by AIRS with respect of CNTRL caused by AIRS

retrievals is about 30%,retrievals is about 30%,radiances only 15% for 1-day forecasts. radiances only 15% for 1-day forecasts.

Overall skill is Overall skill is very good in the 1-dayvery good in the 1-day forecasts, forecasts, reasonable at day 2reasonable at day 2, but drops , but drops

at day 3. at day 3.

Zhou et al., (2010) Zhou et al., (2010)

AIRS cloudy retrievals impact the forecast of AIRS cloudy retrievals impact the forecast of Nargis Nargis structurestructure moremore than clear radiances than clear radiances

RadiancesVery weakSystem, lowVorticity

RetrievalsMuch higher (100%) Vorticity

Retrievals: Realistic 2-bandstructure comparing well with satellite

Radiances: very poor structure:Two unconnected convective systems without a deep circulation

Precipitation Precipitation `Ànalysis’’`Ànalysis’’for Nargisfor Nargis

Not a true precipitation analysis Not a true precipitation analysis since no precip data are since no precip data are

assimilated. Precip comes from assimilated. Precip comes from the `corrector sequence’ and is the `corrector sequence’ and is essentially a set of very short essentially a set of very short

term forecasts strongly term forecasts strongly constrained by observations.constrained by observations.

The assimilation containing AIRS The assimilation containing AIRS retrievals –which improves retrievals –which improves

Nargis structure- also producesNargis structure- also produces the the best precipitationbest precipitation ànalysis’.ànalysis’.

Validation is made against Validation is made against SSM/I, AMSU and TMI dataSSM/I, AMSU and TMI data

Zhou et al., (2010)Zhou et al., (2010)

OBS

CNTRL

AIRS RAD

AIRS RET

Precipitation Forecast for Precipitation Forecast for NargisNargis

Forecasts computed along track and Forecasts computed along track and validated with SSM/I data. validated with SSM/I data.

Ingestion of AIRS retrievals cause the Ingestion of AIRS retrievals cause the GEOS-5 to have better skillGEOS-5 to have better skill. Improvement . Improvement with respect of CNTRL caused by with respect of CNTRL caused by AIRS AIRS

cloudy retrievalscloudy retrievals (tight QC) is (tight QC) is about 20%.about 20%.The impact of radiances is negligibleThe impact of radiances is negligible. . Overall skill is very good in the 1-day Overall skill is very good in the 1-day

forecasts. Skill forecasts. Skill still reasonable at day 3.still reasonable at day 3.

Since the largest amount of Since the largest amount of casualties caused by Nargis were casualties caused by Nargis were

due to FLOODs,due to FLOODs,this result has prominent this result has prominent

implicationsimplications

Zhou et al., (2010)Zhou et al., (2010)also show also show consistentconsistent AIRS impact AIRS impact

on on Wilma (2005)Wilma (2005)

Extreme precipitation over Indus River Valley Extreme precipitation over Indus River Valley

(July-Aug 2010 floods)(July-Aug 2010 floods)

However, However, AIRS cloudy retriev. AIRS cloudy retriev. improve acc. prec.improve acc. prec.along the along the central central partpart of the Indus of the Indus Valley with respect Valley with respect to radiance assim.to radiance assim.

All operational All operational models missedmodels missedthe precip maxthe precip maxover the Upper over the Upper Indus ValleyIndus Valley

Conclusions (Part III)Conclusions (Part III)

Sets of data assimilation experiments without AIRS, with AIRS Sets of data assimilation experiments without AIRS, with AIRS cloudy retrievals (at two different quality controls) and with AIRS cloudy retrievals (at two different quality controls) and with AIRS clear-sky radiances were produced for clear-sky radiances were produced for boreal winterboreal winter,, spring, two spring, two summers summers and and fall fall conditions, for a total of conditions, for a total of about 600 days; about 600 days; 5- or 5- or 7-day forecasts are produced from all sets of analyses, for a 7-day forecasts are produced from all sets of analyses, for a total of about total of about 600 forecasts600 forecasts

The overall skill of forecasts initialized from analyses in which The overall skill of forecasts initialized from analyses in which retrievals are assimilated is higher in every seasonretrievals are assimilated is higher in every season

Consistent improvements in the Consistent improvements in the analysis of Tropical Cyclonesanalysis of Tropical Cyclones are noted as a consequence of AIRS retrievals ingestionare noted as a consequence of AIRS retrievals ingestion

The improvements affectThe improvements affect FORECAST TRACK, TC structureFORECAST TRACK, TC structure andand EXTREME PRECIPITATION FORECASTS EXTREME PRECIPITATION FORECASTS

The importance of not rejecting AIRS-derived information The importance of not rejecting AIRS-derived information from cloudy areas becomes even more evidentfrom cloudy areas becomes even more evident

Part IV: Other improvements.Part IV: Other improvements.One of One of many possiblemany possible model improvements: model improvements:

interactive aerosol in the GEOS-5interactive aerosol in the GEOS-5 An interactive aerosol capability based on Colarco et al An interactive aerosol capability based on Colarco et al

(2009, JGR) implemented in the NASA GEOS-5 by Arlindo (2009, JGR) implemented in the NASA GEOS-5 by Arlindo da Silva and collaboratorsda Silva and collaborators

Experiments to cover the NAMMA period (15Aug-15Sep Experiments to cover the NAMMA period (15Aug-15Sep 2006)2006)

Five sets of 30 5-day forecasts with a) no aerosol, b) Five sets of 30 5-day forecasts with a) no aerosol, b) climatologically varyiing aerosol, c) interactive aerosol, at climatologically varyiing aerosol, c) interactive aerosol, at two different global resolutions (0.5 and 0.25 deg)two different global resolutions (0.5 and 0.25 deg)

Results in: Results in: Reale, O. K. M. Lau, and A. da Silva (2011): Reale, O. K. M. Lau, and A. da Silva (2011): Impact of interactive aerosol on the African Easterly Jet in Impact of interactive aerosol on the African Easterly Jet in the NASA GEOS-5 global forecasting system. the NASA GEOS-5 global forecasting system. Weather and Weather and Forecasting,Forecasting, in press. in press.

Interactive aerosol in the NASA GEOS-5Interactive aerosol in the NASA GEOS-5

Temperature impact (shaded) induced by Temperature impact (shaded) induced by interactive aerosol (conc. solid black). interactive aerosol (conc. solid black). Vertical section at 10WVertical section at 10W

From Reale et al. (2011) From Reale et al. (2011)

Good forecast of aerosol transportGood forecast of aerosol transport

Wind forecast improvement Wind forecast improvement up to 108-hoursup to 108-hours for the for the same soundings in Cape Verde taken during same soundings in Cape Verde taken during

NAMMA SOP-3 (2006)NAMMA SOP-3 (2006)LEFT:LEFT: Temper (obs, Temper (obs, verif analysis, contrl verif analysis, contrl fcst); induced temp. fcst); induced temp. anomaly (clim aer minus anomaly (clim aer minus cntrl, and inter. aer. cntrl, and inter. aer. minus cntrl 108 hour minus cntrl 108 hour forecasts)forecasts)

RIGHT:RIGHT:Wind profiles:Wind profiles:Obs, verif analysis,Obs, verif analysis,interactive aerosol,interactive aerosol,climatological aer.climatological aer.

Reale et al. (2011)Reale et al. (2011)

Conclusions and FutureConclusions and FutureHow can we improve How can we improve

weather forecasts in the tropics?weather forecasts in the tropics? The analysis of the global atmosphere is still The analysis of the global atmosphere is still very deficientvery deficient

in the tropicsin the tropics The largest number of victims by any natural catastrophe The largest number of victims by any natural catastrophe

are caused by extreme weather events in the tropics (firstly are caused by extreme weather events in the tropics (firstly freshwater floods caused by TCs).freshwater floods caused by TCs).

Current Current high-resolution global modelshigh-resolution global models start to resolvestart to resolve some some of the of the features of weather systemsfeatures of weather systems in the tropics in the tropics

Bad initializationBad initialization hinders model performance hinders model performance Improvements inImprovements in modelsmodels (resolution, more sophisticated (resolution, more sophisticated

treatments of processes, parametrizations) are importanttreatments of processes, parametrizations) are important Improvements inImprovements in analysesanalyses stemming from a more efficient stemming from a more efficient

use of use of existing sensors existing sensors (e.g. (e.g. AIRSAIRS) are ) are equally importantequally important Improvements deriving from Improvements deriving from future sensorsfuture sensors (GPM, ISSWL, (GPM, ISSWL,

etc) can be evaluated with OSSEs and could drive our etc) can be evaluated with OSSEs and could drive our next-generation weather forecasting ability.next-generation weather forecasting ability.

AcknowledgmentsAcknowledgments

Donald Anderson Donald Anderson (NASA HQ) for past support to proposal (NASA HQ) for past support to proposal `Òbserving `Òbserving System Experiments ’(OSE and OSSE) to evaluate and enhance the System Experiments ’(OSE and OSSE) to evaluate and enhance the impact of current and future satellite observations’ impact of current and future satellite observations’ (PI: Oreste Reale, (PI: Oreste Reale, 2006-2009)2006-2009)

Ramesh KakarRamesh Kakar (NASA HQ) for current support to proposal (NASA HQ) for current support to proposal ``Relationships among precipitation characteristics, atmospheric water ``Relationships among precipitation characteristics, atmospheric water cycle, climate variability and change’’ cycle, climate variability and change’’ (PI: W. K. Lau, 2009-2011)(PI: W. K. Lau, 2009-2011)

Ramesh KakarRamesh Kakar (NASA HQ) for new support to proposal (NASA HQ) for new support to proposal `Ùsing AIRS `Ùsing AIRS data to understand processes affecting tropical cyclone structure in a data to understand processes affecting tropical cyclone structure in a global data assimilation and forecasting framework’’ global data assimilation and forecasting framework’’ (PI: Oreste Reale, (PI: Oreste Reale, 2011-2013)2011-2013)

Tsengdar LeeTsengdar Lee (NASA HQ) for generous allocations of NASA High End (NASA HQ) for generous allocations of NASA High End Computer resourcesComputer resources

AIRS teamAIRS team at at JPLJPL and the and the Sounder Research TeamSounder Research Team at at NASA GSFCNASA GSFC

Reale, O., J. Terry, M. Masutani, E. Andersson, L. P. Riishojgaard, J. C. Jusem, 2007: Preliminary evaluation Reale, O., J. Terry, M. Masutani, E. Andersson, L. P. Riishojgaard, J. C. Jusem, 2007: Preliminary evaluation of the European Centre for Medium-Range Weather Forecasts (ECMWF) Nature Run over the Tropical of the European Centre for Medium-Range Weather Forecasts (ECMWF) Nature Run over the Tropical Atlantic and African Monsoon region. Geophysical Research Letters, 34, L22810, Atlantic and African Monsoon region. Geophysical Research Letters, 34, L22810, doi:10.1029/2007GL31640.doi:10.1029/2007GL31640.

Reale, O., J. Susskind, R. Rosenberg, E. Brin, E. Liu, L. P. Riishojgaard, J. Terry, J. C. Jusem, 2008: Improving Reale, O., J. Susskind, R. Rosenberg, E. Brin, E. Liu, L. P. Riishojgaard, J. Terry, J. C. Jusem, 2008: Improving forecast skill by assimilation of quality-controlled AIRS temperature retrievals under partially cloudy forecast skill by assimilation of quality-controlled AIRS temperature retrievals under partially cloudy conditions. Geophysical Research Letters, 35, L08809, doi:10.1029/2007GL033002.conditions. Geophysical Research Letters, 35, L08809, doi:10.1029/2007GL033002.

Reale, O., W. K. Lau, J. Susskind, E. Brin, E. Liu, L. P. Riishojgaard, M. Fuentes, R. Rosenberg, 2009: AIRS Reale, O., W. K. Lau, J. Susskind, E. Brin, E. Liu, L. P. Riishojgaard, M. Fuentes, R. Rosenberg, 2009: AIRS Impact on the Analysis and Forecast Track of Tropical Cyclone Nargis in a global data assimilation and Impact on the Analysis and Forecast Track of Tropical Cyclone Nargis in a global data assimilation and forecasting system. Geophysical Research Letters, 36, L06812, doi:10.1029/2008GL037122.forecasting system. Geophysical Research Letters, 36, L06812, doi:10.1029/2008GL037122.

Wu, M.-L, O. Reale, S. Schubert, M. J. Suarez, R. Koster, P. Pegion, 2009: African Easterly Jet: Structure and Wu, M.-L, O. Reale, S. Schubert, M. J. Suarez, R. Koster, P. Pegion, 2009: African Easterly Jet: Structure and Maintenance. Journal of Climate, 22, 4459-4480.Maintenance. Journal of Climate, 22, 4459-4480.

Reale, O., W. K. Lau, K.-M. Kim, E. Brin, 2009: Atlantic tropical cyclogenetic processes during SOP-3 NAMMA Reale, O., W. K. Lau, K.-M. Kim, E. Brin, 2009: Atlantic tropical cyclogenetic processes during SOP-3 NAMMA in the GEOS-5 global data assimilation and forecast system. Journal of the Atmospheric Sciences, 66, in the GEOS-5 global data assimilation and forecast system. Journal of the Atmospheric Sciences, 66, 3563-3578.3563-3578.

Zhou, Y., W. K. Lau, O. Reale, R. Rosenberg, 2010: AIRS Impact on precipitation analysis and forecast of Zhou, Y., W. K. Lau, O. Reale, R. Rosenberg, 2010: AIRS Impact on precipitation analysis and forecast of tropical cyclones in a global data assimilation and forecasting system. Geophysical Research Letters, 37, tropical cyclones in a global data assimilation and forecasting system. Geophysical Research Letters, 37, L02806, doi.1029/2009GL041494.L02806, doi.1029/2009GL041494.

Reale, O., and W. K. Lau, 2010: Reply to Comment on: Àtlantic tropical cyclogenetic processes during SOP-3 Reale, O., and W. K. Lau, 2010: Reply to Comment on: Àtlantic tropical cyclogenetic processes during SOP-3 NAMMA in the GEOS-5 global data assimilation and forecast system.‘ Journal of the Atmospheric NAMMA in the GEOS-5 global data assimilation and forecast system.‘ Journal of the Atmospheric Sciences, 67, 2411-2415.Sciences, 67, 2411-2415.

Reale, O., W. K. Lau, and A. da Silva, 2011: Impact of interactive aerosol on the African Easterly Jet in the Reale, O., W. K. Lau, and A. da Silva, 2011: Impact of interactive aerosol on the African Easterly Jet in the NASA GEOS-5 global forecasting system. In press on Weather and Forecasting.NASA GEOS-5 global forecasting system. In press on Weather and Forecasting.

Wu, M.-L, O. Reale, S. Schubert, M. J. Suarez, C. Thorncroft, 2011: African Easterly Jet: barotropic instability, Wu, M.-L, O. Reale, S. Schubert, M. J. Suarez, C. Thorncroft, 2011: African Easterly Jet: barotropic instability, waves and cyclogenesis. Conditionally accepted on Journal of Climate.waves and cyclogenesis. Conditionally accepted on Journal of Climate.

Documents

Tropical weather systems within a global data assimilation and forecasting framework