Extended Analysis of the Eﬀects of the Sumatras Topography on Downstream Low-level Vortex Development over the Indian Ocean Pa.pcu.pcl.pc E. Ci.pce.pcs.pci.pce.pcl.pcs.pck.pci.pc a.pcn.pcd.pc Ri.pcc.pch.pca.pcr.pcd.pc H. Jo.pch.pcn.pcs.pco.pcn.pc

2025-05-06 0 0 3.19MB 15 页 10玖币

Extended Analysis of the Eﬀects of the Sumatra’s Topography on Downstream Low-level

Vortex Development over the Indian Ocean

Paul E. Ciesielski and Richard H. Johnson

Department of Atmospheric Science, 1371 Campus Delivery, Colorado State University, Fort Collins, CO 80523

ABSTRACT: Fine et al. (2016, hereafter F16) investigated the potential role of Sumatra Island, as well as the Malay Peninsula and

Java, in creating terrain-induced circulations over the Indian Ocean (IO) that subsequently develop into tropical cyclones (TCs). Applying

sophisticated vortex tracking software to 2.5 yrs of model analyses, F16 found four regions downstream of topographic features in the

Maritime Continent to be proliﬁc generators of low-level cyclonic vortices (123/yr). Vortices shed from these “hotspot” regions contributed

to 25% of all TCs occurring in the IO basin during that period. This present study extends the limited analyses of F16 by applying a similar

approach to 10 yrs (2008-2017) of ERA5 analyses. While the 2.5-yr period which F16 studied was slightly (∼8%) more active in terms

of vortex production than the 10-yr period, in general F16’s ﬁndings are representative of the longer-term record with ∼70% (∼80%) of

all vortices (shed vortices) occurring with easterly low-level (925 hPa) ﬂow over the “hotspot” regions. Additional analysis of the 10-yr

record found that vortex counts are highest near MJO phase 1 when low-level easterlies are strongest over the maritime continent region.

A secondary peak in vortex (and shed vortex) counts occurs during MJO phase 4 when low-level westerlies were present near the equator

west of Sumatra. This suggests that low-level westerly surges on the equator impinging on Sumatra associated with the MJO contribute

to an increase in wake vortex development. While the frequency of vortex genesis over the four “hotspot” regions is strongly tied to the

annual cycle of winds, periods with anomalous zonal ﬂow are shown to impact vortex counts. This is most apparent in the region oﬀ the

southern tip of Sumatra (SS) where vortex counts were 4 times higher during periods of anomalous easterlies compared to periods with

anomalous westerlies. While positive (negative) El-Nino-Southern Oscillation (ENSO) and Indian-Ocean Dipole (IOD) conditions drive

anomalous easterlies (westerlies), the relationship between vortex count and these large-scale indices is weak (r < 0.1) in most regions.

Only in region SS did the large-scale conditions associated with these indices appear to impact vortex formation frequency in the sense that

positive ENSO and IOD conditions drive anomalous easterlies which result in higher vortex formation rates.

1. Introduction

Under certain atmospheric conditions, stratiﬁed ﬂow can

be blocked by topographic features resulting in downstream

formation of wake vortices (Smolarkiewicz and Rotunno

1989; Rotunno and Smolarkiewicz 1991). Kuettner (1967,

1989) was the ﬁrst to propose that Sumatra may serve

as a generator of wake vortices that subsequently develop

into tropical cyclones (TCs). Easterly ﬂow blocked by

Sumatra’s narrow mountain range, which exceeds 3000m

in places and straddles the equator from 6◦N to 6◦S, can

result in the formation of wake cyclonic vortices over the

India Ocean (IO). Due to the paucity of observations over

the IO, Kuettner’s observations of these wake vortices have

received little attention until recently.

As a result of limitations in our basic understanding of,

and ability to predict the intraseasonal oscillation, partic-

ularly its initiation, a ﬁeld campaign known as the Dy-

namics of the Madden-Julian Oscillation (DYNAMO) was

conducted in the Indian Ocean (IO) region during the pe-

riod October 2011–March 2012 (Yoneyama et al. 2013).

This experiment provided researchers with unprecedented

atmospheric and oceanic datasets over a data-sparse re-

gion of our planet including observations from upper-air

soundings from ships and islands (Ciesielski et al. 2014),

radars, aircraft, and satellites. During DYNAMO a wake

Corresponding author: Paul E. Ciesielski, ciesiels@colostate.edu

vortex initiated oﬀ the northern tip of Sumatra in November

2011 as low-level easterlies associated with an approaching

MJO convective envelope (Johnson and Ciesielski 2013;

Gottschalck et al. 2013) ﬂow impinged upon the island.

This vortex drifted slowly westward for four days develop-

ing into a TC which caused signiﬁcant damage and deaths

in Sri Lanka (http:// www.webcitation.org/63Vnhm01u).

This storm and other TCs, which formed from wake vor-

tices during DYNAMO, resulted in a renewed an interest

in this topic.

Fine et al. (2016) (hereafter F16) investigated the poten-

tial role of Sumatra Island, as well as the adjacent topogra-

phy of the Malay Peninsula and Java (Fig. 1), in producing

terrain-induced circulations over the Indian Ocean that sub-

sequently develop into TCs. Applying sophisticated vortex

tracking software (Hodges 1995, 1999) to high-resolution

(0.25°) model analyses, F16 found that these topographic

features are proliﬁc generators of low-level cyclonic vor-

tices which contributed to 25% of the TCs occurring in the

IO basin during their study period. F16’s ﬁndings were

based on a limited period (2.5 yr) in which high-resolution

ECMWF analyses were available (YOTC analyses from

May 2008 to April 2010) and 6 months of a special DY-

NAMO operational analyses (from October 2011 to March

2012).

With the recent availability of the ECMWF Reanalysis

5th Generation (or ERA5) high-resolution dataset for the

arXiv:2210.14370v1 [physics.ao-ph] 25 Oct 2022

years 1979 to present, the opportunity to extend the F16

analyses to additional years became possible. The goal of

this study is to determine the representativeness of F16’s

ﬁndings compared to a longer period (2008-2017) and

a diﬀerent, and presumably improved, reanalysis dataset.

Utilizing this longer analysis period, we investigate vari-

ability of vortex formation and its relationship to various

large-scale signals such as the MJO, IOD, and ENSO sig-

nals. Finally, we consider some long-term IO TC statistics

to examine the relationship of TCs to wind regimes and

large-scale indices.

2. Data and Methodology

a. Datasets

ECMWF Reanalysis 5th Generation (or ERA5), which

replaces the ERA-Interim reanalysis, is based on 4D-Var

data assimilation using Cycle 41r2 of the Integrated Fore-

casting System (IFS), which was operational at ECMWF in

2016. A detailed description of the ERA5 conﬁguration,

which beneﬁts from a decade of developments in model

physics, core dynamics, and data assimilation relative to

ERA-Interim, can be found in Hersbach et al. (2020). For

this study we used both vorticity and zonal winds from

ERA5.

The Madden-Julian Oscillation (MJO) is the major ﬂuc-

tuation in tropical weather on weekly to monthly timescales

(Madden and Julian 1971). The MJO can be character-

ized as an eastward moving ’pulse’ of cloud and rainfall

near the equator that typically recurs every 30 to 60 days.

The location and strength of the MJO are given by the

Real-time Multivariate MJO (RMM) index developed by

Wheeler and Hendon (2004). This index is based on a pair

of empirical orthogonal functions (EOFs) of the combined

ﬁelds of near-equatorially averaged 850-hPa zonal wind,

200-hPa zonal wind, and satellite-observed outgoing long-

wave radiation (OLR) data. Daily values of the RMM

index provide the amplitude and phase of the MJO. For

this study, classiﬁcation of the vortices by MJO phase was

only done when the MJO signal had a signiﬁcantly robust

amplitude deﬁned here as when the RMM amplitude was

> 1.

To represent the variability of the El Niño/Southern Os-

cillation (ENSO) we use the Multivariate ENSO Index

Version 2 (MEI.v2) which combines both oceanic and at-

mospheric variables into a single assessment of the state of

ENSO (Wolter and Timlin 2011; Zhang et al. 2019). Pos-

itive (negative) values of this index imply warm, El Niño

(cool, La Niña) conditions across the east-central equato-

rial Paciﬁc. The MEI.v2 is available on a monthly basis.

Variability in the IO region is often characterized by

changes in the east-west sea surface gradient across the IO

basin referred as the Indian Ocean Dipole (IOD) mode (Saji

et al. 1999). The IOD is commonly measured by the Dipole

Mode Index (DMI), that is, the diﬀerence between sea

surface temperature (SST) anomalies between the western

(50◦-70◦E) and eastern (90◦-110◦E) tropical (10◦S-10◦N)

IO. A positive (negative) IOD period is characterized by

cooler (warmer) than average water in the tropical eastern

Indian Ocean and warmer (cooler) than average water in

the tropical western Indian Ocean.

Information on IO tropical cyclone tracks and inten-

sity was obtained from the Joint Typhoon Warning Center

(JTWC) best-track archive which goes back to 1945 (Chu

et al. 2002). For this study TC data were used from 1980

to 2017, although Chu et al. (2002) state that the years

after 1984 have the best data quality. The best-track data

contains the storm center locations and intensities (i.e., the

maximum 1-minute mean sustained 10-meter wind speed)

at six-hour intervals. Storms are only considered here that

(1) reach TC intensity (i.e., maximum sustained winds of

35 knots or 39 mph) and (2) formed over the IO (i.e., ex-

tending from Africa to 105◦E). Storms with missing wind

speeds, which are most prevalent in the earlier years of the

archive, were not considered.

b. Vortex tracking

As in F16, identiﬁcation and tracking of low-level lee

vortices was carried out based on the relative vorticity ﬁeld

using the objective feature tracking code of Hodges (1995,

1999). To facilitate the use of this software, relative vor-

ticity at 6 h and 0.25◦horizontal resolution from ERA5

analyses were vertically averaged using data at 850, 875,

900, 925 hPa from 50◦-110◦E, 20◦N-20◦S. The F16 anal-

yses had only three vertical levels in the 850 to 925 hPa

layer. After smoothing the vorticity ﬁeld to retain spatial

scales greater than 450 km, cyclonic vortex features were

tracked if they maintained an amplitude greater than 1.0 ×

10−5s−1for longer than 2 days.

To focus on wake vortices with the potential to develop

into TCs, we restrict our analyses to cases where the vortex

moved westward over the Indian Ocean. As in F16 these

shed vortices are deﬁned as those with 1) a ﬁnal location

over the Indian Ocean that is > 500 km from Sumatra, and

either 2) a ﬁnal minus initial displacement from Sumatra

> 250 km, or 3) their average speed away from Sumatra

is > 0.5 ms−1. The ﬁrst condition ensures that the shed

vortex at the end of its track is some critical distance from

Sumatra while conditions 2 or 3 guarantee that the vortex

is moving away from its generating landmass.

Application of the tracking code to ten years of ERA5

data yields information on both the tracks and locations

of cyclonic lee vortices. Because of seasonal changes in

the ﬂow regime, genesis locations are shown in Fig. 2 for

the boreal cold season (November to April) and the boreal

warm season (May to October). Overall, these analyses

based on 10 years of statistics are remarkably similar to

those shown in F16. For example, during the boreal win-

ter season (Fig. 2a), northeasterly ﬂow across the Malay

Fig. 1. Analysis domain for study and topography of the region (elevation scale to right in km). Blue lines at: 3◦-7◦N,100◦E, 2◦N-2◦S, 97◦E,

and 3◦-7◦S, 105◦E represent averaging areas for zonal winds corresponding to these line segments (𝑈𝑁,𝑈𝐸,and 𝑈𝑆, respectively) to delineate

the various ﬂow regimes in the area which aﬀect lee vortex production.

Peninsula and northern tip of Sumatra result in a high fre-

quency of lee vortices being formed downstream of these

topographic features. Two boxes shown in this ﬁgure (cov-

ering the same areas as in F16) are deﬁned to help quantify

the relationship between the winds and vortex generation

over these regions in subsequent analyses. These two re-

gions are referred to a SN (Sumatra North) and MP (Malay

Peninsula). Other hotspots for lee vortex generation are

also seen over Sri Lanka and the west coast of India but are

not considered further in this study. Between the equator

and 10◦S, westerly ﬂow predominates such that production

of lee vortices over the IO west of the any topographic

features is much less common in this area.

During the boreal summer monsoon (Fig. 2b), south-

easterly ﬂow prevails across the southern tip of Sumatra

and Java resulting in a high frequency of vortex production

in the lee of these topographical features. As in F16, analy-

sis boxes are deﬁned as SS (Sumatra South) and JV (Java)

to highlight these lee vortex hotspots for further analy-

sis. Westerly ﬂow dominates the NH IO region during

this period resulting in several lee vortex hotspots. How-

ever, these areas are not considered further as any vortices

shed from these areas would move eastward and likely not

contribute to any TCs.

To better understand the relationship of diﬀerent ﬂow

regimes on the production of lee vortices, three compu-

tational lines shown in Fig. 1 are deﬁned: (1) 3◦-7◦N,

100◦E, (2) 2◦N-2◦S, 97◦E, and (3) 3◦-7◦S, 105◦E. Aver-

ages of zonal winds in these 3 areas, referred to as 𝑈𝑁,

𝑈𝐸, and 𝑈𝑆, respectively, are considered representative of

the ﬂow in the adjacent analysis boxes and over the equa-

tor west of Sumatra. In regressing relative vorticity in the

analysis boxes onto zonal winds, 𝑈𝑁and 𝑈𝑆, F16 found

that correlations maximize around 0.8 in the 950-850 hPa

layer in the north and around 0.6 in the 950-1000 hPa layer

to the south. Such correlations are expected considering

ﬂow blocking eﬀects of topography and the average height

of 1-2 km in northern Sumatra and slightly lower in the

south. Thus, for future analyses involving the winds in this

study, data from the 925 hPa level are used.

3. Results

a. Comparison to Fine et al. 2016

Vortex statistics for the four vortex production hotspots

identiﬁed in Fig. 2 are listed in Table 1 for the YOTC/DYN

(YD) period. For each region, we compare the number

of total and shed vortices from F16 to that from ERA5

(referred hereafter as EYD). In general, the statistics for

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

ExtendedAnalysisoftheEectsoftheSumatra'sTopographyonDownstreamLow-levelVortexDevelopmentovertheIndianOceanPaulE.CiesielskiandRichardH.JohnsonDepartmentofAtmosphericScience,1371CampusDelivery,ColoradoStateUniversity,FortCollins,CO80523ABSTRACT:Fineetal.(2016,hereafterF16)investigatedthepotentialrole...

展开>> 收起<<

Extended Analysis of the Eﬀects of the Sumatras Topography on Downstream Low-level Vortex Development over the Indian Ocean Pa.pcu.pcl.pc E. Ci.pce.pcs.pci.pce.pcl.pcs.pck.pci.pc a.pcn.pcd.pc Ri.pcc.pch.pca.pcr.pcd.pc H. Jo.pch.pcn.pcs.pco.pcn.pc.pdf

共15页,预览3页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Extended Analysis of the Eﬀects of the Sumatras Topography on Downstream Low-level Vortex Development over the Indian Ocean Pa.pcu.pcl.pc E. Ci.pce.pcs.pci.pce.pcl.pcs.pck.pci.pc a.pcn.pcd.pc Ri.pcc.pch.pca.pcr.pcd.pc H. Jo.pch.pcn.pcs.pco.pcn.pc

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: