MesoScale Convective Systems over WA

8/7/2019 MesoScale Convective Systems over WA

1/16

MESOSCALE CONVECTIVE SYSTEMS (MCSs) OVER ELEVATED TERRAINS INWEST AFRICA

Cyprian U. Okoloye1, Wassila Thiaw2, Vadlamani Kumar2

1 Nigerian Meteorological Agency (NIMET), Central Forecast Office, Oshodi, LagosNigeria

2 African Desk, Climate Prediction Center, Washington DC. USA

ABSTRACT

This work uses a case study approach to describe synoptic circumstances associatedwith the occurrence of Mesoscale convective systems over West Africa during the heightof the 2005 Atlantic Hurricane season. The analysis which is limited to the months ofAugust through September 2005 is carried out using NCEP-NCAR Reanalysis ProjectDataset. The results show that the African Easterly Waves (AEWs) are preceded by alarge convective event composed of several mesoscale convective systems over

elevated terrains. The convection provides a forcing on the baroclinically andbarotropically unstable state that exist over tropical West Africa. It is observed that nearthe West African Coast, the baroclinic structure weakens, while convection ismaintained. Sometimes, organized convective systems are produced during theafternoon convection near the high elevation areas such as Air Mountains, Guineahighlands, Jos Plateau in north central Nigeria, in the vicinity of Lake Chad and Darfur inWestern Sudan. The method used in this study to identify and track the waves is that of(Hodges 1995). In the tracking method, a significant vorticity centre at 850 hPapropagates along the baroclinic zone. Therefore, there is a deliberate focus on thepositive relative vorticity centres that exist north and south of the Easterly Jet over WestAfrica and their relevance as potential precursors for Tropical Cyclones. Cloud clusterson satellite imagery relates well to synoptic features on observational analysis.

Preliminary result suggests that Atlantic Tropical Cyclones activity may be influenced bythe number of AEWs leaving the West African coast, and may have significant low-levelamplitudes, and not simply by the total number of AEWs.


2/16

1.0 Introduction

Over West Africa, almost all theprecipitable water is of convectivenature. It is therefore known that theeconomically useful rainfall comes from

vigorous organized convective systemssuch as Squall Lines (SLs), MesoscaleConvective Complexes (MCCs) andQuasi-Stationary Convection (QSCs).These systems are produced during theafternoon convection near the highelevation areas such as Air mountains,Guinea highlands, Jos plateau innorthcentral Nigeria, in the vicinity oflake Chad and Darfur in western Sudan.During their migration westward, theydevelop during the evening before

reaching their full mesoscale organizedcloud clusters (SL, MCC, QSC) andstrengthen during the night time (Ross,1985). Release of instability can betriggered by surface heating, orography,moisture sources, African WaveDisturbances (AWD) or othercirculations that encourage large-scaleconvergence. AEWs are an importantpart of the West African and tropicalclimate. They are known both tomodulate the daily rainfall over west

Africa (e.g., Reed et al. 1977) and toinitiate most Atlantic tropical cyclones(e.g., Landsea et al. 1998 andThorncroft et al., 2001). Bothobservations and models have shownthat characteristics of the AEJ(particularly horizontal and verticalshear) influence wave developmentcharacteristics (Burpee 1972, Thorncroft1995). However, the relationshipbetween these waves and convectiveactivity is complex.

Previous analyses of waves usingoperational analyses has mainly usedstatistical techniques to diagnose theAEW activity (e.g., Reed et al. 1988a)Ceron and Gueremy 1999). Thesepapers diagnosed different aspects ofthe AEW activity, including the differentregions of AEW activity north and south

of the African Easterly Jet (AEJ). Incontrast, a different perspective of thenature of AEW activity was obtained by(Reed et al. 1988b), who manuallytracked AEWs between August andSeptember in 1985. His result showed

the tracks of individual systems movingover Africa, into the Atlantic Ocean, andin some cases, recurving as tropicalcyclones. This paper considers atracking method, following the vorticitycenters within the AEWs. This trackingtechnique offers a different butcomplimentary view to results usingstatistical techniques. Therefore, thereis a focus on the positive relativevorticity centers that exist north andsouth of the easterly Jet over Africa and

their relevance as potential precursorsfor tropical cyclones. The trackingscheme adopted here, tracks onlysystems with closed vorticity contoursgreater than or equal to 0.5 X 10-5s-1.This may mean that weak AEWs earlierin their life cycle are not tracked usingthis scheme and may contribute to theweaker analyzed activity over land.

The present study therefore,investigates the synoptic circumstances

associated with the occurrence of AEWsover West Africa during the height of the2005 Atlantic Hurricane season, usinggridded NCEP-NCAR Reanalysisdataset and preliminary information on2005 Atlantic Hurricane report from theNOAAs Tropical Prediction Center.Therefore, a case-study approach isused to describe the evolution of anintense AEW from its initiation in Darfur,western Sudan (near (12oN, 30oE) on 15August until it crosses the West African

coast on 24 August 2005 (see Fig. 1).This system was the strongest AEW inthe summer of 2005. The analysistakes a much closer look at the synopticevolution during this period, focusingmore on the interactions of the synoptic-scale AEW and the mesoscaleconvective systems (MCSs), and


3/16

highlighting the relationship between theAEWs and the Tropical Cyclones.

2.0 Data and Methodology

This work is part of a broader study

which examines the African EasterlyWave (AEW) structures in relation towind fields during part of the 2005Atlantic Hurricane Season. In thisstudy, the results are limited to the useof NCEP-NCAR Reanalysis (RA2) andin particular, for the period August September 2005. The NCEPReanalysis dataset consists of daily andmonthly gridded data (u, v, Z etc) on2.5o X 2.5o horizontal resolution withusually at least 17 vertical levels. This

dataset is currently available from 1979to 2005, but at the time analysis arecarried out, only 2005 dataset are used.The dataset is derived as a combinationof observations and model results basedon the NCEP global operational model.The observational data consistsprimarily of global rawinsonde data, theConsolidated Ocean AtmosphereDataset (COADS) surface marine data,aircraft data, surface land synoptic data,SSM/I surface wind speeds, satellite

cloud drift winds, and satellite sounderdata. Details on the data and its qualitycontrol, the model data assimilationtechniques, and preliminary validationare given in Kalnay et al, (1996).Satellite Infra-red imagery fromMeteosat, on half hourly basis, analyzedthree hourly is also used.

2.1 The scale interaction is analyzedthrough:

(i) Hovmoller space-time diagram thattends to emphasize the synoptic-scale(Rossby wave type) signal of AEWs.

(ii) Analysis of the 700 hPa U-component (averaged), used to identifyand describe the characteristics of thelarge scale features such as the middlelevel African Easterly Jet (AEJ).

(iii) Analysis of 850 hPa potentialvorticity tracking of the nature of AEWs.The method used to identify and trackthe waves is that of Hodges (1995).This method identifies the waves as

maxima in the threshold relative vorticityfield (a threshold of +0.5 X 10-5s-1 isused) and tracks them using a methodthat minimizes a cost function for themotion coherence subject to constraintson the motion. It should be noted thatsystems that have closed vorticitycontours with values of at least +0.5 X10-5s-1 are being detected with thisapproach.

The first step is to examine the

Longitude-Time Section of meridionalwind averaged over the sector 5oN and15oN at 700 hPa, using the NCEPs RA2. Next is the Zonal wind fields,averaged over the sector 18oW and15oE. Characteristics of interest includethe general structure, intensity andlocation of the middle level jet streamand its seasonal development andassociated horizontal and vertical windshear.

Also, part of the cyclone statistics fromthe 2005 Atlantic Hurricane report of theNOAAs Tropical Prediction Center(TPC) is downloaded fromNCEP/NOAAs website, and evaluatedin an attempt to confirm some of themajor changes in the wind field assuggested by the datasets (see Tables1 and 2). Infra-red Meteosat Data isused to infer convection in deepconvective systems that supportdevelopment of AEWs. An analysis of

850 hPa potential vorticity fields inNCEP Reanalysis data completes thecharacterization of the basic state forthe AEWs. The basic-state variablesare computed for the period 16 August 16 September 2005, forming a 1-monthperiod centered on the passage of thisAEW.


4/16

Table 1.

SUMMARY TABLE

NAME DATES MAX WIND DEATHS U.S. DAMAGEMPH $MILLION

----------------------------------------------------------.TS HARVEY 2-8 AUG 65 0H IRENE 4-18 AUG 105 0TD TEN 13-14 AUG 35 0TS JOSE 22-23 AUG 50 1 **H KATRINA 23-31 AUG 175 ** **TS LEE 28 AUG - 40 0----------------------------------------------------------NOTE...DATES BASED ON COORDINATED UNIVERSAL TIME (UTC)* PRELIMINARY ESTIMATE** UNKNOWN AT THIS TIME

FORECASTER FRANKLIN/AVILA/BEVEN/KNABB/PASCH/STEWARTSource: http//www.nhc.noaa.gov/archive/2005/tws/miatwsat_sep.shtml?

Table 2.

SUMMARY TABLE

NAME DATES MAX WIND DEATHS U.S. DAMAGEMPH $MILLION

----------------------------------------------------------.H MARIA 1-10 SEP 115 0H NATE 5-10 SEP 90 0

H OPHELIA 6-18 SEP 85 1 50H PHILIPPE 17-24 SEP 80 0H RITA 18-26 SEP 175 ** **TD NINETEEN 30 SEP - 35 ** **----------------------------------------------------------NOTE...DATES BASED ONCOORDINATED UNIVERSAL TIME(UTC)* PRELIMINARY ESTIMATE** UNKNOWN AT THIS TIME

FORECASTER

PASCH/STEWART/BEVEN/FRANKLIN/KNABB/AVILASource:http//www.nhc.noaa.gov/archive/2005/tws/miatwsat_sep.shtml?

3.0 Results and Discussions

3.1 Time-longitude cross sections

Figure 1 shows a hovmoller space-timediagrams of unfiltered meridional winds(v) averaged between 5oN and 15oN at

700 hPa over West African region forAugust and September 2005. Fromfigure 1, it can be seen that the wavespropagate westward across the Africancontinent. The waves mostly beginbetween 15o and 25oE and occasionallyat 30oE, and continue into the Atlantic.The systems move across the continentevery 2-5 days, have wavelengths of


5/16

order 3000 km, and phase speeds in theregion of 10-15 ms-1, confirming earlierstudies of AEWs by (Reed et al. 1977).The NOAAs Tropical Prediction Center(TPC) identified many AEWs during the2005 Atlantic Hurricane season as

precursors to tropical cyclones.Features matching information in TPCreports can be singled out from Fig. 1.At least 10 out of the 2005 August andSeptember total of 12 tropical cyclonesare attributed to AEWs, including 7major Hurricanes in these two months.

The most outstanding feature in figure 1is an AEW that has a (positive)meridional wind signature that beginsnear 30oE on 16 August and crosses theWest African coast (near 15oW) on 23August. This system is well organized

and most distinct due to its strengthwhen compared to the other AEWs ofthe months of August throughSeptember which is the primary reasonfor this event forming the basis of thiscase study.

Fig. 1 Hovmoller diagram of 700 hPa unfiltered meridional wind, averagedbetween 5oN and 15oN for the period 16 Aug. 16 Sep. 2005

4.0 Zonal Winds

The basic state variables are computedfor the period 16 August 16September 2005, forming a 1 monthperiod centered on the passage of thisAEW. The mean zonal wind at 700 hPaon NCEP-NCAR (RA2) for this period isshown in Fig. 2. The African EasterlyJet (AEJ) is located around 13oN 17oN

in Fig. 2a over the continent and closeto 17oN over the Atlantic Ocean. TheJet is strongest on the West African

coast with a peak speed of 11ms-1 (Fig.2a). The easterly winds can be seen tostretch as far back as longitude 30oE.The detail analysis of the AEJ is ofinterest since it acts as a source ofenergy for the African wave disturbaces.Fig. 2b shows the zonal wind crosssection averaged between17.5oW and30oE, features an AEJ with a maximum


6/16

speed of 10 ms-1 at 600 hPa near 14oN.The figure also features a shallowwesterlies from the surface to about 950hPa level, and is located between 4.5oNand 11oN. The depth of the westerliesare seen to decrease with increasing

latitudes, but fizzle out at about 18oN.Above 950 hPa level, easterliesdominate the flow pattern. Resultspresented here further suggest that

equatorial westerlies significantlymodulate interannual variability. Theseserve to displace the jet and associateddisturbances poleward and to enhanceconvective activity. From this monthlymean structure, AEJ core can

reasonably differ from a few-day mean(map not shown).

Fig. 2a. Zonal wind at 700 hPa averaged between 00UTC16Aug. 18UTC16Sep.2005(RA2)


7/16

Fig. 2b: Mean zonal wind components (ms-1) fromRA2, averaged between 17.5oW

and 30oE for the period 00UTC16Aug. 18UTC16Sep.2005

5.0 Wind and Vorticity Fields

The meridional wind at 700 hPa servesto highlight the Rossby wavelike aspectsat the level of the AEJ, and relativevorticity at 850 hPa serves to highlightthe vorticity centers that are associatedwith AEWs. Figure 3 shows the 850hParelative vorticity compared with a 700hPa meridional wind (map not shown)over West Africa from 16 August to 23

August 2005. Some well definedfeatures of the AEW can be seenroughly from 15o and 20oE westward. Inthe 700 hPa meridional wind field, asynoptic-scale trough is first seen overWest Africa on 16 August at 0000 UTCand moves into the Atlantic on 23August. This is consistent with theinformation presented in the HovmollerDiagram (Fig. 1). A comparison of the700 hPa meridional wind and satelliteimagery (Fig. 4) indicates that over West

Africa, the center of the observedconvection is always ahead of thewestward moving trough, consistent withprevious composites studies (e.g, Reedet al 1977).

A significant vorticity center at 850 hPapropagates along the baroclinic zone.The vorticity center can first be seen on16 August at 1200 UTC (Fig. 3a) oversouthern Chad (approximately 11oN,17oE). The vorticity center can betracked over subsequent analysis timesto a position close to the West Africancoast (13oN, 13oW) on 20 August at0000 UTC (Fig. 3b) and (17oN, 14oW) on23 August at 1200UTC (Fig. 3c). From

wind field analysis, the 850 hPa vortexover Guinea stagnates and intensifiesfor the 24 hour as the upstream 700 hPatrough continues to move westward.With the passage of the 700 hPa troughand observed convection, (Fig. 4) thevortex over Guinea moves into theeastern Atlantic Ocean. This suggeststhe possibility of an interaction ormerger between the vortex over Guineaand the vorticity associated with the 700hPa trough. It is this vorticity center that

can be directly linked to HurricaneKatrina (not shown).


8/16

Fig. 3a

Fig. 3b


9/16

Fig. 3c

Fig. 3. 850 hPa Relative vorticity (contours) shown every 12 h from 1200 UTC on the16Aug. 2005 until 1200 UTC on 23 Aug. 2005

6.0 Satellite Observations

This study focuses on the mesoscaleconvective system (MCS) that istriggered over the Darfur area of Sudanon 16 August 2005. This event iscrucial for the subsequent developmentof Hurricane Katrina in the easternAtlantic Ocean. Figure 1 shows anintense AEW in the meridional windfield, that is associated with an

organized westward moving convectivesystem, shown by an Infra-red (IR)satellite imagery in Fig. 4. A verydistinct feature can be tracked in the IRimagery from an initial region ofconvection on 16 August (see Figs. 4(a-b)), evolving through several stages andmoving off the West African coast on 23August 2005. It should be noted that IRimagery before 15 August 2005 is notavailable during this study. The initialconvection occurs around 1500UTC on

15 August 2005 (map not shown) near12oN, 30oE. This is a mountainous areain the Darfur region of western Sudan,and a region favored for the initiation oflong-lived mesoscale convectiveweather systems (Hodges andThorncroft 1997). It is observed that theinitial convection is located east of andearlier than the first location of the AEW

shown in the Hovmoller diagram ofmeridional wind (Fig. 1). This suggeststhat the convection acts as theprecursor for the AEW that developsdownstream. Subsequent satelliteimages show that this convective regionrapidly develops another long-livedmesoscale convective system thatmoves across West Africa. Theconvection usually starts from Darfurregion around 1200 UTC. The

brightness of the pixels indicate that theamount of very deep convection beginsto decay 18 to 20 h after the initiation,by which time the center of theconvective region is close to 10oE (Fig.4). There is a reduced amount of deepconvection within the convective regionfor the following 12 h as it moves overwestern Chad, northern Cameroon andinto northeastern Nigeria. Deepconvection occurs on the leading edgeof the remnants of the first convective

burst along the Gulf of Guinea coast(Fig. 4b). This convection decays afterapproximately 12 h, in agreement withthe movement into a region of lowermean equivalent potential temperaturenear the coast. In addition to thiscontinuously evolving region ofconvection propagating westward, theIR imagery shows another interesting


10/16

feature. Persistent localized convectionoccurs over Guinea highlands,beginning at 1800 UTC and continuinguntil it is incorporated into the westwardpropagating region of convection. Fromthese series of analyses, it is likely that

interactions between convectionoccurring over the Guinea highlandsand the propagating convection mayhave played a role in the formation ofthe offshore tropical cyclone.

Fig. 4a


11/16

Fig. 4b


12/16

Fig. 4c


13/16

Fig. 4d

6.1 IR TemperatureFigure 5 shows a longitude-Time sectionof IR temperature of less than 235oK.This cross section is also used to inferconvection across the West Africanregion. It further indicates areas of

strong convection over places of highelevations, and mainly between west of15oE and 25oE. The convection here isalso associated with the westwardmovement of the AEWs up till the timethey enter eastern Atlantic Ocean.

Fig. 5. Longitude-Time diagram of IR Cloud Temperature


14/16

At 1800 UTC on 23 August, the TPCbest-track data places TropicalDepression 10 at 23.1oN, 75.1oW(Richard et al. 2005) within one of theseregions of convection. It is alsoobserved, twenty-four hours after

crossing the West African coast (around0600 UTC 24 August 2005) a singleorganized, cyclonically rotating cloudmass near 23oW (identified by the TPCas Tropical storm Katrina) dominatesthe cloud features in this region. In their2005 Tropical Cyclone Report, Knabb etal, has argued that Katrina involves thecomplex genesis of the interaction of a

tropical wave, the middle troposphericremnants of tropical Depression Ten,and an upper middle tropospherictrough. They conclude that this trough,located over the western Atlantic andthe Bahamas, produces strong westerly

shear across Tropical Depression Ten,causing it to degenerate on 24 Augustapproximately 825 n mi east ofBarbados. The best tract of the pathof the center of Katrina according to theNational Hurricane Center (NHC), isshown in Fig. 6.

20

25

30

35

40

-95 -90 -85 -80 -75 -70

Hurricane Katrina23-31 August

Hurricane

Tropical Storm

Tropical Dep.

Extratropical

Subtr. Storm

Subtr. Dep.

00 UTC Pos/Date

12 UTC Position

Low / Wave

PPP Min. press (mb)

25

24

31

30

29

2827

26

902 mb

920 mb

984 mb

928 mb

Fig. 6. Best track positions for Hurricane Katrina, 23-30 August 2005.Source: www.nhc.noaa.gov/pdf/TPC-AL122005_katrina.pdf


15/16

7. Summary and conclusions

A case-study approach has been usedto describe the evolution of an intenseAEW from its initiation in Darfur, westernSudan (near 12oN, 30oE) on 15 August

until it crosses the West African coaston 24 August 2005. This system wasthe strongest AEW in the summer of2005. The results show that the AEWsare preceded by a large convectiveevent composed of several mesoscaleconvective systems over elevated

terrains. The tracking method showsthat, a significant vorticity center at 850hPa propagates along the barocliniczone. Meteosat infrared cloud imageryfor the afternoons of August 16-27,2005, showed four major cloud clusters

making their way, in succession, acrossWest Africa from east to west. The firstand the fourth move in tandern with theAfrican wave disturbances that arediscernible on analysis of Hovmollerdiagram of 700 hPa meridional wind.

Acknowledgements

I wish to thank the Director of AfricanDesk of Climate Prediction Center, Dr.

Wassila Thiaw who conscienciously andconsistently monitored and supervisedthis project from the beginning to theend. Numerous discussions with Dr.Thiaw have also added insight into theresults. My special appreciation also

goes to Dr. Vadlamani Kumar whoprovided the much needed technicalassistance in programming and datahandling. Finally, but not the least, I

extend my gratitude to the Director ofClimate Prediction Center, Mr. JimLaver for his sweet words ofencouragement, on a day he found mebattling with this work at the AfricanDesk on a quiet public holiday.

References

Berry, G. J., and C. D. Thorncroft, 2005:Case study of an Intense AfricanEasterly Wave. Mon. Wea. Rev., 133,

752-766

Burpee, R. W., 1972: The origin andstructure of easterly waves in the lowertroposphere of north Africa. J. Atmos.Sci, 29, 77-90

Druyan, L., and T. Hall, 1996: Thesensitivity of the African wavedisturbances to remote forcing, J. Appl.Meteorol, 35, 1100-1110

Druyan, L., M., M. B. Fulakeza, and W.Thiaw, 2000: Regional model simulationof African wave disturbances. Jour.Geoph. Resea, vol. 105 No. D6, 7231-7255

Grist, J. P. and S. E., Nicholson, 2001:A study of the Dynamic Factors

Influencing the Rainfall Variability in theWest African Sahel.

Hodges, K. I., and C. D. Thorncroft,1997: Distribution and statistics of

African mesoscale convective weathersystems based on the ISCCP Meteosatimagery. Mon. Wea. Rev., 125, 2821-2837.

Hodges, K. I., 1995: Feature tracking onthe unit sphere. Mon. Wea. Rev., 123,3458-3465

Knabb, R. D., J. R. Rhome, and D. P.Brown, 2005: Tropical Cyclone Report,Hurricane Katrina, 23-30 August 2005.

National Hurricane Center, 20 Dec.2005. (Preliminary Report downloadedfromwebsite(www.nhc.noaa.gov/pdf/TCR-AL122005_Katrina.pdf)


16/16

Kalnay, E., and coauthors, 1996: TheNCEP/NCAR 40-year ReanalysisProject Bull. Amer. Meteo.soc., 77, 437-471 Ross, R. S., 1985: Diagnosticstudies of African easterly wavesobserved during GATE. Department of

earth sciences, Milersville University ofPennsylvania, Milersville, Pa. 374 PP

Landsea, G. D. Bell, W. M. Gray, and S.B. Goldenberg, 1998: The extremelyactive 1995 Atlantic Hurricane season:Environmental conditions andverification of seasonal forecasts. Mon.Wea. Rev., 126, 1174-1193.

Reed, R. J., D. C. Norquist, and E. E.Recker, 1977: The structure and

properties of African wave disturbancesas observed during phase III of GATE.Mon. Wea. Rev. 105, 317-333.

Reed, R. J., E. Klinker, and A.Hollingsworth, 1989: The structure andcharacteristics of African easterly wavedisturbance as determined from theECMWF Operational Analysis/ForecastSystem. Meteor. Atmos. Phys., 38,22-33

Ross, R. S., 1995: Diagnostic studies ofAfrican easterly waves observed duringGATE. Department of earth sciences,Milersville Univ. of Pennsylvatia,Milersville, PA. 374 pp.

Thiaw, W., and R. Scofield, Use ofsatellite imagery and NMC GDAS toanalyze easterly waves over Africa andthe Atlantic Ocean, in preprint vol of the20th Conference on Hurricanes andtropical Meteorology, pp. 259-262, Am.

Meteorol. Soc, Boston, Mass,

Thorncroft, C. D., 1995: An idealizedstudy of African easterly waves. Part III:More realistic basic states. Quart. J.Roy. Meteor. Soc., 123, 1589-1614

Thorncroft, C. D., and K. Hodges, 2001:African easterly wave variability and itsrelationship to Atlantic tropical cycloneactivity. J. Climate, 14, 1166-1179.

MesoScale Convective Systems over WA

Documents