EARLY ONLINE RELEASE

This is a PDF of a manuscript that has been peer-reviewed

and accepted for publication. As the article has not yet been

formatted, copy edited or proofread, the final published

version may be different from the early online release.

This pre-publication manuscript may be downloaded,

distributed and used under the provisions of the Creative

Commons Attribution 4.0 International (CC BY 4.0) license.

It may be cited using the DOI below.

The DOI for this manuscript is

DOI:10.2151/jmsj.2019-046

J-STAGE Advance published date: April 22nd, 2019

The final manuscript after publication will replace the

preliminary version at the above DOI once it is available.

1

1

Numerical Analysis of the Mesoscale Dynamics of an 2

Extreme Rainfall and Flood Event in Sri Lanka in May 3

2016 4

5

Suranjith Bandara KORALEGEDARA 6

Earth System Science Program, Taiwan International Graduate Programme, Academia 7 Sinica, Taipei, Taiwan. 8

Research Center for Environmental Changes, Academia Sinica, Taipei, Taiwan. 9 Institute of Atmospheric Physics, College of Earth Science, National Central University, 10

Zhongli, Taiwan. 11 12

Chuan-Yao LIN1 13

Earth System Science Program, Taiwan International Graduate Programme, Academia 14

Sinica, Taipei, Taiwan. 15 Research Center for Environmental Changes, Academia Sinica, Taipei, Taiwan. 16

and 17

Yang-Fan SHENG 18

Research Center for Environmental Changes, Academia Sinica, Taipei, Taiwan. 19 20

21

4th Revised 4 April, 2019 22

23

24

------------------------------------ 25

1) Corresponding author: Chuan-Yao Lin, Research Fellow, Research Center for 26 Environmental Changes, Academia Sinica, No.128, Section 2, Academia Road, Taipei, 27 Taiwan. 28

Email: yao435@rcec.sinica.edu.tw 29 Tel: +886-226539885 Ext. 435 30 Fax: +886-227833584 31

2

Abstract 32

This study analyzed the synoptic and mesoscale dynamics and underlying mechanism 33

of an extreme rainfall and flood event that occurred in Sri Lanka from 14–17 May 2016 34

using the Weather Research and Forecasting Model simulations with a horizontal grid size 35

of 3 km and observational data. This extreme rainfall event was associated with a low-36

pressure system (LPS) that originated over the Bay of Bengal in the Indian Ocean and 37

passed over Sri Lanka. The observed maximum accumulation of rainfall during the event 38

exceeded 300 mm at several weather stations during 15–16 May and resulted in severe 39

flooding and landslides, particularly in the western part of the island. The model closely 40

simulated the timing of the initiation of the LPS and its development along the east coast 41

of Sri Lanka. The model could capture the overall rainfall tendency and pattern of this 42

event. Synoptic and mesoscale analyses indicated that this extreme rainfall event 43

occurred as the cumulative effect of a sustained low-level convergence zone generated by 44

an enhanced westerly monsoon flow and the circulation of the LPS alongside a continuous 45

supply of high-magnitude moisture, strong vertical motion, and orographic effects of the 46

Central Mountains of Sri Lanka. Model sensitivity experiments indicated that rainfall over 47

the western slope area of the mountains was enhanced by mountain lifting, whereas 48

western coastal rainfall was reduced because the mountains blocked the northeasterly 49

flow of the LPS. 50

51

Keywords: Heavy Rainfall, Weather Research and Forecasting Model, Extreme event, 52

Low-pressure system 53

54

55

56

57

3

1. Introduction 58

The Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment Report 59

documents the increasing frequency of extreme weather events such as extreme heat and 60

heavy rainfall events in a number of regions worldwide (IPCC 2014b). Temperatures of the 61

Indian Ocean have exhibited an increasing trend (Goswami et al., 2006; Rajeevan et al., 62

2008), which may be related to the increase of extreme weather events in the region. 63

Particularly the South Asian summer monsoon rainfall is projected to increase in a future 64

warmer world, both in terms of long-term mean rainfall and inter-annual variability of 65

rainfall leading to extreme rainfall events (Kitoh 2017; Ogata et al., 2014). The IPCC 66

further predicts increased summer monsoon rainfall over the Indian subcontinent and 67

increased extreme rainfall over the coasts of the Bay of Bengal (BoB) caused by 68

cyclogenesis in the BoB and Arabian Sea (IPCC 2014a). Extreme rainfall events can 69

cause dramatic social effects such as severe socioeconomic losses due to flooding, 70

reservoir replenishment, agricultural damage or loss, and damaged infrastructure 71

(Easterling et al., 2000; Groisman et al., 1999). Damage is particularly extensive when 72

these events occur over metropolitan areas consisting of dense populations and 73

concentrated economic activity (Oldenborgh et al., 2016; Promchote et al., 2016). 74

To prevent or mitigate severe losses caused by extreme rainfall, it is vital to understand 75

the atmospheric processes leading to heavy rainfall events within recognized weather 76

patterns. Disturbances such as tropical cyclone (TC), low pressure system (LPS), 77

monsoonal flow, and mesoscale convective systems are mostly thermally driven and 78

powered by the latent heat of condensed water vapor carried by trade winds. Most of 79

these types of disturbances form and develop over the warmer parts of the oceans where 80

there is an almost unlimited supply of heat and moisture (Saha 2010; Trenberth et al., 81

2003). Usually, heavy rainfall occurs because of moisture availability, moisture transport, 82

storm propagation, and rainfall efficiency. Interaction between low-level winds and the 83

4

local topography also plays important role in heavy rainfall events. The presence of an 84

orographic obstacle across the moisture flow further contributes to extreme rainfalls (Chen 85

et al., 2005; Juneng et al., 2007; Lin and Chen 2002). During TC or LPS, the higher terrain 86

elevations tend to enhance the rainfall mostly on the windward side of the mountains, 87

because the steep terrain tend to lift the airflow and produce strong upward motions (Tang 88

et al., 2012; Yang et al., 2008). 89

Among these disturbances, low pressure systems (LPSs) are renowned for their 90

frequent contribution to extreme rainfall events, particularly over the BoB region 91

(Krishnamurthy 2012). Tropical cyclones (TCs) and LPSs occur during two peak periods in 92

the BoB region. The first peak period is the premonsoon/spring season (March–May), 93

which is known to produce low-intensity disturbances/LPSs, and the second peak period is 94

the postmonsoon/fall season (October–December), which is known to produce severely 95

intense cyclonic disturbances (Chand and Singh 2017; Pattanaik and Mohapatra 2016). 96

Deep convection that occur over South Asia and the Indian subcontinent (Zipser et al., 97

2006) are also known to be associated with the premonsoon and monsoon seasons (Saha 98

2010; Webster et al., 1998). Generally, depressions and LPSs form over the moisture-rich 99

BoB and involve mesoscale convective systems (Houze and Churchill 1987). 100

The various thermodynamic and dynamic effects of a local area further aid in the 101

development of weather systems. Sri Lanka, being an island nation in the BoB region, is 102

vulnerable to TCs, particularly those generated over the southern part of the BoB. 103

According to the Disaster Management Center (DMC)’s ranking of natural disasters in Sri 104

Lanka, cyclone events are ranked second and flood events are ranked fourth in terms of 105

the number of human lives lost. In terms of the frequency of natural disaster events, flood 106

events are ranked first, and cyclone events are ranked fourth [Disaster Management 107

Centre (DMC) and the United Nations Development Program (UNDP), 2012]. 108

5

A LPS originated in the BoB southeast of Sri Lanka developed into a tropical cyclone 109

named Roanu in May 2016. This system prevailed over Sri Lanka from 14-17 May 2016, 110

and over 300 mm of maximum accumulation rainfall was recorded at several weather 111

stations during 15-16 May 2016 (Fig. 1a) , particularly in the western and northern parts of 112

Sri Lanka (Alahacoon et al., 2016). Initially, widely scattered rainfall fell across the country, 113

but more intense rainfalls were observed in western and northeastern parts of the island, 114

which resulted in prolonged flooding in those areas that persisted until 22 May. The 115

intense rainfall moved northward along with the movement of the LPS. The LPS triggered 116

extensive flooding and landslides that affected thousands of lives and livelihoods (United 117

Nations Office for the Coordination of Humanitarian Affairs (UN-OCHA), 2016) (Figs. 1a 118

and 1b). This torrential rainfall event caused the worst floods of the last 25 years in the 119

densely populated Western Province of Sri Lanka (UN-OCHA 2016). Three large-scale 120

landslides were also reported during the event over the western slopes of the Central 121

Mountains. A total of 22 administrative districts out of the 25 in Sri Lanka were affected by 122

the event (Fig. 1b). According to the DMC, the whole event resulted in over 200 fatalities, 123

with 100,000 homes and other property being damaged and more than 450,000 people 124

being displaced (DesInventar 2016; UN-OCHA 2016). The estimated economic loss 125

calculated by total damage exceeded US$ two billion throughout the country (Gunathilaka 126

2018). Overall, the event is considered to be one of the biggest natural disasters in the 127

recent history of Sri Lanka. 128

Sri Lanka frequently experiences extreme rainfalls under the influences of two monsoon 129

systems and atmospheric disturbances from the BoB. However, detailed analyses on both 130

the synoptic and mesoscale dynamics in extreme events in Sri Lanka using numerical 131

model are rare. Further, while a number of studies stressed the importance of topography 132

on heavy rainfall events，the effects of Sri Lanka's topography on heavy rainfalls have not 133

been investigated. Thus, this research sought to understand the dynamics of the extreme 134

6

rainfall event from 14-17 May 2016 using the Advanced Research Weather Research and 135

Forecasting Model (WRF-ARW, hereafter WRF). Through the analysis of model results, we 136

seek to fill the scale gap between synoptic and mesoscale, and clarify the role of 137

topography in Sri Lanka on the extreme event. Improving knowledge concerning the 138

mechanisms of this extreme event can help meteorological services effectively identify and 139

forecast similar events in the future. The present study is one of the first literature to 140

examine this heavy rainfall and flood event. Thus, we hope that the findings of this study can serve 141

as a reference for future numerical simulations. In Section 2, a detailed description of the data, 142

model settings, and the overall methodology is provided. In Section 3, the results are illustrated 143

and discussed. Last, a summary and conclusion are presented in Section 4. 144

2. Data sources and methodology 145

2.1. Data sources 146

The 6-hourly National Center for Environment Prediction Climate Forecast System 147

version 2 data (NCEP-CFSv2) (Saha et al., 2011) at 0.5 degrees and 37 vertical levels 148

were used for initial and lateral boundary conditions in this study. The data are available 149

online at the research data archive website of the National Center for Atmospheric 150

Research (http://rda.ucar.edu/datasets/ds094.0/). 151

Rainfall in the simulation results were validated against the observation data from the 152

ground weather stations of the Department of Meteorology of Sri Lanka. Because of the 153

limited availability of continuous high temporal resolution quality–controlled data, most of 154

the analysis of observation data was conducted using daily rainfall data; the hourly rainfall 155

data from the Colombo station data were the only hourly data analyzed. The European 156

Organization for the Exploitation of Meteorological Satellites (EUMETSAT)’s Meteosat-7 157

visible and infrared spin-scan radiometer (VISSR) Indian Ocean data coverage infrared 158

cloud imageries available at Dundee Satellite Receiving Station 159

7

(http://www.sat.dundee.ac.uk), were used to justify the LPS development and rainfall 160

structure of the simulated model results. 161

2.2. Model and experimental setup 162

The simulations constructed in this study were based on version 3.6.1 of the WRF 163

model, which was designed to serve and support both atmospheric research and 164

operational forecasting requirements (Skamarock and Klemp 2008). The WRF model 165

system is composed of a new-generation mesoscale forecasting model featuring multiple 166

dynamical cores, which are nonhydrostatic, fully compressible, and have a terrain-167

following hydrostatic-pressure vertical coordinate system. More details concerning the 168

WRF model can be found in Skamarock et al., (2008) and Skamarock and Klemp (2008). 169

All experiments were conducted over a single domain (Fig. 1c). To provide the model 170

with enough area for the weather system development and initiation, the domain extended 171

beyond the island of Sri Lanka. Model resolution is an important aspect in simulating the 172

spatial distribution of rainfall (Gao et al., 2006; Mass et al., 2002). The horizontal grid size 173

was set to be 3 km throughout the model domain. The model simulation was initialized at 174

0000 Coordinated Universal Time (UTC) 14 May 2016, one day before the LPS began 175

affecting Sri Lanka, and ended at 0000 UTC 21 May 2016. To control for the accumulation 176

of errors in the model simulation, a four-dimensional data assimilation scheme was applied 177

that incorporated the boundary conditions every 6 h from the NCEP-CFSv2 dataset. First 178

12 hours of the simulation was treated as the spin up time for the model and excluded 179

from the analysis. The model results from 1200 UTC 14 May – 0200 UTC 17 May were 180

used for detailed analysis because that period represents the peak period of the event 181

over the heavily affected western part of Sri Lanka. 182

Depending on the strength of the synoptic-scale forcing and season, a proper cumulus 183

parameterization scheme (CPS) is required when simulating certain atmospheric settings 184

(Gilliland and Rowe 2007). Deep convection in the gray zone resolutions (2–10 km) are 185

8

partly resolved and partly subgrid. Thus, such cases stress the necessity of a CPS for 186

numerical predictions of convective weather, even at high resolutions (Gerard 2007). In 187

the current study, we opted to use a CPS at the fairly high resolution of 3 km. The details 188

of the model configuration and physics and dynamics employed are summarized in Table 189

1. The output simulation results from the described model configuration were further 190

analyzed and discussed. 191

In addition to the control simulation as described above, a series of simulations were 192

performed to check the rainfall sensitivity with respect to the topographic height of the 193

model. Three sensitivity studies were conducted where all model configures are similar to 194

the control simulation except for the terrain height being reduced to zero (case Topo-Z), 195

25% (case Topo-Q) and 50% (case Topo-H) of the control simulation. Conclusions of 196

these sensitivity studies are given in section 3.3.c. 197

3. Results and discussion 198

3.1. Synoptic conditions and Rainfall 199

The NCEP-CFSv2 data indicated a weak cyclonic circulation originating over the 200

southwest of the BoB off the southeast coast of Sri Lanka on 14 May 2016, which was 201

attributable to a LPS. The disturbance lingered over the southeast of Sri Lanka and then 202

drifted offshore of the east coast of Sri Lanka early on 15 May. Sea surface pressure data 203

from the origin of the LPS to the time it passed along Sri Lanka indicated that it was a well-204

defined but weak system that would only be categorized as a low-pressure area or 205

depression (Figs. 2a–c). From 16 May, the LPS moved from north to northwest, intensified 206

into a deep depression, and then became the cyclonic storm Roanu (India Meteorological 207

Department, 2016). The cyclonic flow of the LPS was maintained from the surface to 208

upper air. The mid-troposphere steering-flow patterns up to 500 hPa and beyond also 209

indicated the presence of the closed LPS on a broader scale (Figs. 2a–c). NCEP-CFSv2 210

9

data clearly demonstrated the origin, intensification, and development of the LPS in the 211

BoB and along the east coast of Sri Lanka (Figs. 2a–c). It is evident from the 500-hPa-212

level geopotential height (Fig. 2a) that a northwest- to southeast-oriented trough was over 213

the Indian Ocean next to the LPS origin area. In general, these monsoon troughs lead to 214

cyclonic system formation, particularly during the pre-monsoon period in the BoB (Wang et 215

al., 2013; Wu et al., 2010). The thermal infrared images from the EUMETSAT Meteosat-7 216

satellite displayed in Figs. 2d and 2f clearly depict the transformation of the cloud pattern 217

associated with the LPS during its time over Sri Lanka. The images also help to identify 218

the development of the LPS and the well-established cloud cover over most of Sri Lanka 219

during this period. These infrared satellite images depict the development of the cyclonic 220

cloud bands during this event and the cloud system persistently covering the entire of Sri 221

Lanka (Fig. 2e). These cloud images also correspond to the continuous moisture transport 222

from the westerlies and strengthening of the wind (Fig. 2f). Most of the thick cloud cover 223

and patterns in the satellite imagery matched the observations of heavy rainfall and LPS 224

development. 225

The NCEP-CFSv2 data revealed persistently strong westerly winds at sea level with 226

maximum values of 15–20 m s−1 over the west coast of Sri Lanka and the southwest 227

quadrant of the LPS (Figs. 2a–c). These westerly winds were extending from the Somali 228

jet, a strong cross-equatorial flow (next to the East African coast, pronounced between 5° 229

S and 10° N; Figs. 2a–c). This Somali jet stream acts as one of the main suppliers of 230

moisture for South Asian rainfall (Cadet and Desbois 1981; Saha 2010). Our results also 231

indicated a continuous supply of moisture from the westerlies to Sri Lanka and the BoB 232

region during the study period. The westerlies and the wind from the LPS collided over the 233

west coast of Sri Lanka to create a convergence zone over the western slopes of the 234

Central Mountains resulting in heavy rainfall in the adjacent areas (see Section 3.3.1). 235

10

Regions of high water vapor and rainfall moved toward Sri Lanka from the southeast 236

and western directions, crossed the island from 14 to 16 May, and left Sri Lanka heading 237

toward the northeast. Rainfall continued during this period, following the direction of the 238

LPS and mainly centered over the western slopes of the Central Mountains. The relatively 239

limited number of ground weather stations in Sri Lanka recorded this particular event, and 240

several stations reported over 300 mm of accumulation rainfall between 15 and 16 May 241

(Fig. 1a). One of the highest daily rainfalls recorded for 15 May was 350 mm at 242

Deraniyagala (Fig. 3a). Unfortunately, data was only available on 15 May for this station. 243

The western slopes of the Central Mountains of Sri Lanka received an excessive amount 244

of rainfall that triggered large-scale landslides, particularly in Kegalle District. The Maha 245

Illuppallama weather station located in Anuradhapura District in north-central Sri Lanka 246

recorded the maximum rainfall of 267.8 mm for 16 May (Fig. 3b). The highest rainfall totals 247

were recorded on 15 and 16 May when the center of the LPS was located over the east 248

coast of Sri Lanka. These observations highlight that the locations of maximum rainfall 249

followed the movement of the LPS. Overall rainfall on the west coast was relatively higher 250

than on the east coast during this event. Compared with the rest of the island, the 251

southeastern coastal area of the island experienced comparably light rainfall throughout 252

this period. 253

3.2. Model validation 254

Model validation was performed by comparing the observations and model simulations 255

in terms of rainfall and circulation during the peak period of the event. Successful 256

simulation of the rainfall was a key objective of this study. Figures 3a–b illustrate the 257

observed daily rainfall and can be compared with Figs. 3c-d, which present the model-258

simulated rainfall for 15–16 May 2016. The LPS moved northwestwards and lingered as a 259

pronounced LPS over the east coast of Sri Lanka on 15 May. The highest rainfall on 15 260

May over the southwestern part of Sri Lanka was captured by the model, particularly the 261

11

rainfall over the western slopes of the Central Mountains. The model-derived 24-h results 262

of rainfall accumulation indicated that areas receiving over 150 mm of rainfall stretched 263

from the west coast eastward to the Central Mountains (Fig. 3c), which was similar to the 264

real observations (Fig. 3a). However, the simulation showed lower rainfall accumulation 265

over the northern part of Sri Lanka (Fig. 3c) compared with observations (Fig. 3a). Heavy 266

rainfall on the west coast and northwest side of the Central Mountains on 16 May was 267

captured by the model, but overestimation occurred for the middle of the Central 268

Mountains region. The spatial patterns of the model-derived daily rainfall amounts for 15 269

and 16 May were closely related to the observations. During 16 May, the system moved 270

along the east coast of Sri Lanka, and by the end of 17 May the LPS had already passed 271

the northern territory of Sri Lanka. 272

A more detailed comparison was conducted with the hourly rainfall variations at the 273

Colombo weather station (the starred station in Fig. 1a). Colombo and the surrounding 274

suburban areas experienced record-breaking rainfall, and it was the area worst affected by 275

the subsequent floods. Figure 4 depicts the hourly development of the observed and 276

model-simulated rainfall for the Colombo weather station during the period from 0000 277

Local Standard Time (LST) 15 May to 0800 LST 17 May 2016. This period coincides with 278

the daily rainfall of 15–16 May (daily rainfall is calculated by Sri Lanka’s Department of 279

Meteorology from 0830 LST to 0830 LST the following day) and also with the peak period 280

of the event over this area. Rainfall data from the Colombo station indicated that the 281

extreme rainfall event started at approximately 0300 LST 15 May and lasted continuously 282

until the middle of 16 May, exhibiting several peak rainfall spells in between. The model-283

derived hourly rainfall variation followed the observed trend of the Colombo weather 284

station. Although the model could not capture the exact magnitude of the initial peak 285

rainfall spell that occurred early on 15 May, the subsequent rainfall spells of late 15 May 286

and 16 May were reasonably modeled (Fig. 4). 287

12

To further validate the model, we also computed the following statistical indicators: 288

coefficient of determination (R-squared), standard error of the regression (S), mean 289

absolute error (MAE), and mean fractional bias (MFB). All the indices were calculated for 290

the Colombo weather station in Sri Lanka using the nearest model grid point rainfall and 291

the hourly rainfall data from ground observations using Eqs. 1–4, respectively. 292

293

294

295

296

297

298

where SSRegression is the sum squared regression error and SSTotal is the sum squared total 299

error. Model predictions are Pi (i = 1, 2 . . . n) and observations are Oi (i = 1, 2 . . . n). All 300

these indices are indicators of the goodness of fit of the model results with the 301

observations. R-squared is a statistical index used to determine how close the data are to 302

the fitted regression line. The statistical indicator S is the average distance that the values 303

fall from the regression line, given in natural units of the variable. The MAE gives the 304

average of the magnitudes of difference, and MFB indicates both the direction and 305

probable magnitude of the error. Figure 5 illustrates the scatter plot of hourly rainfall 306

derived from the model results versus the observed hourly rainfall for the Colombo 307

weather station. Although the R-squared value for this regression was 0.6, the other 308

13

statistical indices reflect that the model-derived rainfall was reasonably close to the 309

observations with lower bias. We found it challenging to validate the model results for 310

some parts of Sri Lanka because of the limited availability of high spatial and temporal 311

resolution quality–controlled weather data from Sri Lanka. Thus, we only used the 312

Colombo weather station data for this detailed analysis, because it represented the 313

primary area affected by this extreme rainfall event. 314

The model results were validated by comparing the 10 m wind vectors (Figs. 6 a, b, and 315

f) against those derived from the NCEP-CFSv2 data. Because the LPS was still a 316

depression and had no clear LPS center when it passed along Sri Lanka, comparing the 317

exact path of the system was challenging; yet, the model simulations captured the LPS 318

center and wind field well as it developed. The simulation data indicated the strengthening 319

of the cyclonic winds at the LPS center (Figs. 6 e and f). Overall, the simulated vortex 320

managed to maintain an LPS center similar to the NCEP-CFSv2 data. The model also 321

accurately simulated wind speed which increased from 5 to 20 m s−1 as the LPS 322

developed. 323

As evident from the comparison, the analysis of rainfall and the circulation system, and 324

the supporting satellite images, the simulation reconstructed the circulation and the rainfall 325

system for this particular extreme event with reasonably accuracy. Thus, the modeling 326

result can be subjected to further diagnostic analysis. 327

3.3. Mechanisms responsible for extreme rainfall 328

a. Moisture transport 329

The airflow over the west coast of Sri Lanka mainly came from the moisture-rich 330

westerlies over the Indian Ocean and Arabian Sea. The BoB area also exhibited high 331

water vapor content. The continuous moisture supplies from each of these areas played a 332

notable role in the development of the LPS. To analyze the moisture level and distribution, 333

the water vapor flux (WVF) was calculated using Eq. 5: 334

14

335

where q is specific humidity, and and are the zonal wind component and meridional 336

wind component, respectively. The WVF was relatively weak prior to 15 May over the 337

northwest coast of Sri Lanka, the Palk Strait (the Strait between India and Sri Lanka), and 338

the east coast of India (Fig. 6a). The highest WVF values for 15 and 16 May (Figs. 6 b–f) 339

occurred on the west coast and around the center of the LPS when the system passed 340

along the east coast of the island. In particular, the atmospheric moisture content over the 341

west coast and surrounding areas of Sri Lanka gradually increased and then leveled off, 342

as reflected in Figs. 6d–f, because the northerly flow of the LPS and the westerly monsoon 343

flow interacted with the Central Mountains. Figure 7 illustrates the simulated low-level (925 344

hPa) wind convergence (negative), divergence (positive), and zonal and meridional 345

divergent wind vectors that occurred during the study period. The distribution of the low-346

level convergence zone throughout this period indicated that the heavy rainfall areas (Fig. 347

3b) were located next to the area of significant convergence (Figs. 7c and 7d). 348

Concurrently, the model results indicated heavy rainfall occurred over the southwest to 349

west coasts (Fig. 3e). The low rainfall in the southeast coastal region and at the eastern 350

slopes of the Central Mountains on 16 May (Fig. 3f) was mainly attributable to the LPS 351

tracking north (Figs. 6e and 6f); divergence clearly dominated (Figs. 7e and 7f) in this 352

region. Reduced moisture content was evident near the center of the LPS, and a slowing 353

of the development of the LPS from 1200 UTC 16 May through 1200 UTC 17 May was 354

also evident (not illustrated here; see Supplementary Figs.S1 and S2, a-e). These 355

phenomena were possibly attributable to the blocking of the structure of the LPS by Sri 356

Lanka’s landmass and discontinued moisture supply from the westerlies. However, after 357

0600 UTC on the afternoon of 17 May the LPS again intensified and developed into a TC 358

when moisture support from the BoB resumed and favorable conditions prevailed (not 359

illustrated here; see Supplementary Figs. S1 and S2, c-f). 360

15

b. Locally enhanced convection and potential instability 361

Heavy rainfall is usually associated with strong convection. The synoptic environmental 362

conditions such as the moisture-rich lower atmospheric flow were favorable in generating 363

convective rainfall in this event. Convective available potential energy (CAPE) is a 364

measure of the potential updraft strength of thunderstorms and has been used in the 365

analyses of various weather systems (Bhat et al.,1996; Murugavel et al., 2012). High 366

CAPE values larger than 2500 J kg-1 are usually associated with severe weather (Storm 367

Prediction Center, 2017). Figure 8 shows model simulated CAPE and equivalent potential 368

temperature (EPT) at the lowest model level (～160m above surface). During the rainfall 369

event, high CAPE values over 3000J kg-1 persisted on the west coast of Sri Lanka and 370

west coast of southern India (Figs. 8 a-c). On the other hand, the CAPE and EPT over 371

land were much lower than those over the ocean (Figs. 8b and 8c). The low-level 372

moistening of the synoptic environment near Sri Lanka’s coastal areas (particularly on the 373

west coast) increased the EPT of the boundary layer air (Fig.8b). High EPT together with 374

high WVF facilitated the increase of instability and convective development in this area. 375

Favorable conditions for the creation of a deep and moist convection include high CAPE 376

values and a strong ascending air current that can reach the free convection level (Schultz 377

et al., 2000). Generally, the results also highlight that even with high CAPE values (>2500 378

J kg−1), the resulting rainfall would not be strong or even occur at all if the available 379

moisture was low and the vertical motion was weak (as evident on the eastern side of the 380

island). On the other hand, high CAPE values with a high moisture supply and strong 381

vertical motion (as evident on the western side of the island) generate strong rainfall. 382

Similar conditions were seen during typhoon Morakot in Taiwan (Lin et al., 2011). 383

The strong upward motion and deep convection were further examined by the cross-384

section plot of EPT and vertical velocity (Figs. 9a–d). Steep vertical gradients in the EPT 385

cross-section to the west of the heavy rainfall area on the west coast of Sri Lanka 386

16

suggested that low-level moist air flows with an EPT greater than 350 K were the main 387

energy source driving the weather system (Figs. 9a–c). These relatively high EPT (EPT 388

ridges) in the low-level are often the burst points of thermodynamically induced events like 389

this (Figs. 9a–d). Lifting of moisture rich air parcel from its original pressure level to the 390

upper levels of the troposphere will release the more latent heat of condensation in that 391

parcel. This also could be confirmed with the follow up heavy radar reflectivity in these 392

areas (Figs. 9f-h). At 0600 UTC 15 May, dry air with EPT values less than 351 K was 393

prevalent in the middle levels over Sri Lanka (Fig. 9a). Relatively low echo top height at 394

this time (Fig. 9e) suggests that the vertical development of convection was firstly 395

suppressed by the dry air capping. Later, this resulted in explosive deep convection and 396

strong upward motion (Fig. 9d). The simulated radar reflectivity can also be used as a 397

proxy to indicate the deep cloud convection and rainfall in the area. Figures 9e–h illustrate 398

the cross-sections of the radar reflectivity that can be used to identify the vertical 399

distribution of convective cores. The strongest upward motions were associated with the 400

largest reflectivity up to 60 dBZ. The overall effect during the period resulted in a strong 401

uplift of moisture, leading to a heavy rainfall event in the area. 402

c. Role of topography 403

The dynamical forcing of the local topography was also a critical aspect of this event. To 404

further examine the effects of topography on precipitation, sensitivity studies with the 405

terrain height changed to zero (case Topo-Z), a quarter (case Topo-Q), and half (case 406

Topo-H) of the control run were conducted. Figures 10 a–c show the differences between 407

the sensitivity cases (Topo-Z, Topo-Q, and Topo-H) and the control run for the 48-h 408

accumulated rainfall (15–16 May) over Sri Lanka. To further analyze the difference, two 409

rectangular areas over the western part of the island were selected. Rectangle A in Figs. 410

10a–c is the Colombo metropolitan region, which represents the area in which the most 411

flooding over the coastal and plain areas occurred, as indicated in Fig. 1a. Rectangle B in 412

17

Figs. 10a–c is the western slope of the Central Mountains where another heavy 413

accumulation of rainfall was observed, as displayed in Fig. 1a. The differences of rainfall 414

amounts over rectangle A indicate that coastal precipitation is enhanced as the 415

topographic height is reduced. The difference in 48-h accumulated rainfall between case 416

Topo-Z and the control could be as high as 400 mm (Fig. 10a). The northeasterly airflow 417

associated with the LPS easily penetrated the island in case Topo-Z because of the 418

absence of the mountains’ blocking effect, and then the flow interacted with the westerly 419

monsoon flow over the western coastal area. However, the different rainfall amounts over 420

rectangle B indicate that the precipitation amount is reduced as the topographic height is 421

reduced (Figs.10a–c) because the effect of mountain lifting on the westerly airflow is 422

reduced. The 48-h accumulated rainfall over rectangle B in Topo-Z was approximately 400 423

mm less than that in the control run. Figure 11 depicts the percentage changes of 424

accumulated total rainfall in both rectangles. Over the Colombo municipal region 425

(rectangle A), both the maximum and average accumulated rainfall increased as the 426

topographic height was reduced. In the Topo-Z case, the maximum and average 427

accumulated rainfall over rectangle A increased by 60% and 13% relative to the control, 428

respectively. However, over the western mountain slope area (rectangle B), the maximum 429

and area-average accumulated rainfall decreased as the topographic height was reduced 430

(Fig.11a–b). In case Topo-Z, the maximum and average accumulated rainfall decreased 431

by 33% and 17% relative to the control, respectively. These results over rectangle B 432

demonstrated that the mountains again critically affected air mass lifting and then 433

enhanced the upward motion over the western slope area. 434

The detailed mechanism of this event is summarized and illustrated in Fig. 12. 435

According to the results, we suggest that the interaction of westerlies with the land mass 436

and the northward-propagating LPS modified the low-level moisture convergence structure. 437

This low-level convergence zone was sustained for a relatively long period of 438

18

approximately 20 h, with a constant moisture supply. This persistent low-level 439

convergence enabled a deep convection to occur, which was supported by a continuous 440

moisture supply in this area. This was the primary mechanism that caused the enhanced 441

rainfall over the western coastal belt. The Central Mountains blocked the northeasterly 442

flow of the LPS, whereas the western slopes of the mountains uplift the westerly monsoon. 443

Consequently, the rainfall was particularly enhanced over the western slopes of the 444

Central Mountains of the island. 445

446

4. Summary and conclusion 447

This study numerically simulated and examined the mechanism behind the heavy 448

rainfall event that occurred in Sri Lanka from 14–17 May 2016. The WRF was used to 449

simulate the event and the physical conditions that led to it, and the model performance 450

was also assessed. The extreme rainfall was examined based on the analysis of NCEP-451

CFSv2 data, satellite imagery, ground weather station data, and model simulation results. 452

A single domain with a fixed horizontal resolution of 3 km was used with the best-selected 453

model configuration in the simulations for detailed analysis of the event. 454

Model simulation captured the synoptic-scale features of atmospheric circulation 455

involved in the heavy rainfall event. The model results satisfactorily estimated the path of 456

the LPS and its timing, and captured the overall rainfall trend of the heavy rainfall event 457

over Sri Lanka. However, some spatial and temporal deviations with respect to the total 458

rainfall were apparent at certain locations. Overall, the continuous low-level moisture 459

supplies from the BoB and from the Arabian Sea/Indian Ocean, the sustaining low-level 460

convergence zone that resulted from the interaction of the LPS and the westerlies, and the 461

locally enhanced deep convection together with strong vertical motion appeared to be the 462

prominent features of this event that resulted in the unprecedented heavy rainfall. 463

Sensitivity studies indicated that rainfall over the western slope area of the mountains was 464

19

enhanced by mountain lifting, whereas western coastal rainfall was reduced because the 465

mountains blocked the northeasterly flow of the LPS. The results indicated that the 466

observed extreme rainfall event over the western part of the island was the result of 467

interactions among the following phenomena: 1). the LPS in the BoB, 2). moisture-rich 468

westerlies from the Indian Ocean and the Arabian Sea, 3) persistent low-level 469

convergence and 4). the orographic effect from the Central Mountains of Sri Lanka. 470

The findings of this study verified the usefulness of numerical mesoscale models such 471

as the WRF for predicting and studying the dynamics of extreme events such as the heavy 472

rainfall event in May 2016 over Sri Lanka. The WRF model demonstrated satisfactory 473

ability for predicting the major features of this extreme event. Longer lead time warnings 474

for extreme rainfall and flood events through the use of a single unified modeling system 475

such as the WRF model could have considerable value in disaster preparedness and 476

management in Sri Lanka. 477

Supplement 478

The supplementary material contains figure S1 and S2 which present the model 479

simulated water vapor fluxes and wind convergence/divergence areas at 925 hPa during 480

16-17 May 2016. The purpose of these plots is to provide supporting information for the 481

detail processes of the water vapor flux and the locations of convergence/divergence 482

areas in Figs. 6 and 7. 483

Acknowledgments 484

The authors acknowledge two anonymous reviewers for their valuable comments. The 485

authors are grateful to the Department of Meteorology of Sri Lanka for the ground 486

observation data. We also acknowledge Ms. Anusha R. Warnasooriya at the Department 487

of Meteorology of Sri Lanka for some insights of the event and observations and Mr. R. B. 488

Koralegedara and Mr. Susantha Udagedara for the initial coordination with the data 489

purchase. Further, we would like to thank NCEP, National Center for Atmospheric 490

20

Research for the initial/boundary conditions data. This research was produced with the 491

financial support of the Ministry of Science and Technology, Taiwan (MOST 105-2111-M-492

001-003) and the Cuomo Foundation, Monaco. The contents of this document are solely 493

the liability of its authors and under no circumstances may be considered as a reflection of 494

the position of the Cuomo Foundation and/or the IPCC. 495

References 496

Alahacoon, N., P. Pani, K. Matheswaran, S. Samansiri, and G. Amarnath, 2016: Rapid 497

emergency response mapping for the 2016 floods in Kelani river basin, Sri Lanka. 498

Proc. 37th Asian Conference on Remote Sensing, Colombo, Sri Lanka, Asian 499

Association on Remote Sensing (AARS). 500

Beljaars, A. C. M., 1995: The parametrization of surface fluxes in large-scale models 501

under free convection. Q. J. Roy. Meteor. Soc., 121, 255-270. 502

Bhat, G. S., J. Srinivasan, and S. Gadgil, 1996: Tropical Deep Convection, Convective 503

Available Potential Energy and Sea Surface Temperature. J. Meteor. Soc. Japan, 74, 504

155-166. 505

Cadet, D., and M. Desbois, 1981: A Case-Study of a Fluctuation of the Somali Jet during 506

the Indian-Summer Monsoon. Mon. Wea. Rev.,, 109, 182-187. 507

Chand, R., and C. Singh, 2017: Relation of Frequency of Tropical Cyclones Over North 508

Indian Ocean and North West Pacific Ocean with Sea Surface Temperature Anomaly 509

Over Nino 3.4 Region and Indian Ocean Dipole. Tropical Cyclone Activity over the 510

North Indian Ocean, Mohapatra, M., B. K. Bandyopadhyay, and L. S. Rathore, (Eds.), 511

Springer International Publishing, Cham, Switzerland, 233-243. 512

21

Chen, C. S., W. C. Chen, Y. L. Chen, P. L. Lin, and H. C. Lai, 2005: Investigation of 513

orographic effects on two heavy rainfall events over southwestern Taiwan during the 514

Mei-yu season. Atmos. Res., 73, 101-130. 515

DesInventar, 2016: SriLanka- Flood and Landslide, May 2016. United Nations Office for 516

Disaster Risk Reduction, Colombo, Sri Lanka, (Available online at 517

http://www.desinventar.lk/.) 518

Disaster Management Centre (DMC), 2016. Flood Inundation Maps [Online]. Disaster 519

Management Centre of Sri Lanka, Colombo, Sri Lanka, (Available online at 520

http://www.dmc.gov.lk/IndexInnundationMaps.htm.) 521

Disaster Management Centre (DMC) and the United Nations Development Program 522

(UNDP), 2012: Hazard Profiles of Sri Lanka. Disaster Management Centre, 232 pp. 523

Dudhia, J., 1989: Numerical Study of Convection Observed during the Winter Monsoon 524

Experiment Using a Mesoscale Two-Dimensional Model. J. Atmos. Sci., 46, 3077-525

3107. 526

Easterling, D. R., G. A. Meehl, C. Parmesan, S. A. Changnon, T. R. Karl, and L. O. 527

Mearns, 2000: Climate extremes - observations, modeling, and impacts. Science, 289, 528

2068-2074. 529

Gao, X. J., Y. Xu, Z. C. Zhao, J. S. Pal, and F. Giorgi, 2006: On the role of resolution and 530

topography in the simulation of East Asia precipitation. Theor. Appl. Climatol., 86, 173-531

185. 532

Gerard, L., 2007: An integrated package for subgrid convection, clouds and precipitation 533

compatible with meso-gamma scales. Q. J. Roy. Meteor. Soc., 133, 711-730. 534

22

Gilliland, E. K., and C. M. Rowe, 2007: A comparison of cumulus parameterization 535

schemes in the WRF model. Proc. 87th AMS Annual Meeting and 21th Conference on 536

Hydrology, San Antonio, TX, USA, American Meteorological Society. 537

Goswami, B. N., V. Venugopal, D. Sengupta, M. S. Madhusoodanan, and P. K. Xavier, 538

2006: Increasing trend of extreme rain events over India in a warming environment. 539

Science, 314, 1442-1445. 540

Groisman, P. Y., T. R. Karl, D. R. Easterling, R. W. Knight, P. F. Jamason, K. J. Hennessy, 541

R. Suppiah, C. M. Page, J. Wibig, K. Fortuniak, V. N. Razuvaev, A. Douglas, E. 542

Forland, and P. M. Zhai, 1999: Changes in the probability of heavy precipitation: 543

Important indicators of climatic change. Clim. Change, 42, 243-283. 544

Gunathilaka, S., 2018: The Impact of Natural Disasters on Micro, Small and Medium 545

Enterprises (MSMEs): A Case Study on 2016 Flood Event in Western Sri Lanka. Proc. 546

Eng., 212, 744-751. 547

Hong, S.-Y., Y. Noh, and J. Dudhia, 2006: A New Vertical Diffusion Package with an 548

Explicit Treatment of Entrainment Processes. Mon. Wea. Rev., 134, 2318-2341. 549

Houze, R. A., and D. D. Churchill, 1987: Mesoscale Organization and Cloud Microphysics 550

in a Bay of Bengal Depression. J. Atmos. Sci., 44, 1845-1867. 551

India Meteorological Department, 2016: Cyclonic Storm,’Roanu’ over the Bay of Bengal 552

(17-22 May 2016) - A Report. Cyclone Warning Division, India Meteorological 553

Department, New Delhi, India, 58 pp. 554

IPCC, 2014a: Climate Change 2013 – The Physical Science Basis: Working Group I 555

Contribution to the Fifth Assessment Report of the Intergovernmental Panel on 556

23

Climate Change. Stocker, T. F., and Coauthors (Eds.), 557

doi:10.1017/CBO9781107415324. 558

IPCC, 2014b: Climate Change 2014: Synthesis Report. Contribution of Working Groups I, 559

II and III to the fifth assessment report of the Intergovernmental Panel on Climate 560

Change. Core Writing Team, R. K. Pachauri, and L. A. Meyer (Eds.), IPCC, 151 pp. 561

Juneng, L., F. T. Tangang, and C. J. C. Reason, 2007: Numerical case study of an 562

extreme rainfall event during 9-11 December 2004 over the east coast of Peninsular 563

Malaysia. Meteor. Atmos. Phys., 98, 81-98. 564

Kain, J. S., 2004: The Kain-Fritsch convective parameterization: An update. J. Appl. 565

Meteor., 43, 170-181. 566

Kitoh, A., 2017: The Asian Monsoon and its Future Change in Climate Models: A Review. 567

J. Meteor. Soc. Japan, 95, 7-33. 568

Krishnamurthy, V., 2012: Extreme Events and Trends in the Indian Summer Monsoon. 569

Extreme Events and Natural Hazards - The Complexity Perspective, Sharma, A. S., A. 570

Bunde, V. P. Dimri, and D. N. Baker, (Eds.), American Geophysical Union, 571

Washington, DC, USA, 153-168. 572

Lin, C. Y., and C. S. Chen, 2002: A study of orographic effects on mountain-generated 573

precipitation systems under weak synoptic forcing. Meteor. Atmos. Phys., 81, 1-25. 574

Lin, C. Y., H. M. Hsu, Y. F. Sheng, C. H. Kuo, and Y. A. Liou, 2011: Mesoscale processes 575

for super heavy rainfall of Typhoon Morakot (2009) over Southern Taiwan. Atmos. 576

Chem. Phys., 11, 345-361. 577

24

Lin, Y. L., R. D. Farley, and H. D. Orville, 1983: Bulk Parameterization of the Snow Field in 578

a Cloud Model. J. Climate Appl. Meteor., 22, 1065-1092. 579

Mass, C. F., D. Ovens, K. Westrick, and B. A. Colle, 2002: Does increasing horizontal 580

resolution produce more skillful forecasts? Bull. Amer. Meteor. Soc., 83, 407-430. 581

Mlawer, E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. Clough, 1997: 582

Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k 583

model for the longwave. J. Geophys. Res.- Atmos., 102, 16663-16682. 584

Murugavel, P., S. D. Pawar, and V. Gopalakrishnan, 2012: Trends of Convective Available 585

Potential Energy over the Indian region and its effect on rainfall. Int. J. Climatol., 32, 586

1362-1372. 587

Ogata, T., H. Ueda, T. Inoue, M. Hayasaki, A. Yoshida, S. Watanabe, M. Kira, M. Ooshiro, 588

and A. Kumai, 2014: Projected Future Changes in the Asian Monsoon: A Comparison 589

of CMIP3 and CMIP5 Model Results. J. Meteor. Soc. Japan, 92, 207-225. 590

Oldenborgh, G. J. V., F. E. L. Otto, K. Haustein, and K. AchutaRao, 2016: The Heavy 591

Precipitation Event of December 2015 in Chennai, India. Bull. Amer. Meteor. Soc., 97, 592

S87-S91. 593

Pattanaik, D. R., and M. Mohapatra, 2016: Seasonal forecasting of tropical cyclogenesis 594

over the North Indian Ocean. J. Earth Syst. Sci., 125, 231-250. 595

Promchote, P., S. Y. S. Wang, and P. G. Johnson, 2016: The 2011 Great Flood in 596

Thailand: Climate Diagnostics and Implications from Climate Change. J. Climate, 29, 597

367-379. 598

25

Rajeevan, M., J. Bhate, and A. K. Jaswal, 2008: Analysis of variability and trends of 599

extreme rainfall events over India using 104 years of gridded daily rainfall data. 600

Geophys. Res. Lett., 35. 601

Saha, K., 2010: Tropical Circulation Systems and Monsoons. Springer-Verlag, 329 pp. 602

Saha, S., S. Moorthi, X. Wu, J. Wang, S. Nadiga, P. Tripp, D. Behringer, Y.-T. Hou, H.-y. 603

Chuang, M. Iredell, M. Ek, J. Meng, R. Yang, M. P. Mendez, H. van den Dool, Q. 604

Zhang, W. Wang, M. Chen, and E. Becker, 2011: NCEP Climate Forecast System 605

Version 2 (CFSv2) 6-hourly Products. (Available online at 606

http://dx.doi.org/10.5065/D61C1TXF.) 607

Schultz, D. M., P. N. Schumacher, and C. A. Doswell, 2000: The intricacies of instabilities. 608

Mon. Wea. Rev., 128, 4143-4148. 609

Skamarock, W. C., and J. B. Klemp, 2008: A time-split nonhydrostatic atmospheric model 610

for weather research and forecasting applications. J. Comp. Phys., 227, 3465-3485. 611

Skamarock, W. C., J. B. Klemp, J. Dudhia, D. O. Gill, D. M. Barker, M. G. Duda, X. Y. 612

Huang, W. Wang, and J. G. Powers, 2008: A description of the Advanced Research 613

WRF Version 3. Mesoscale and Microscale Meteorology Division, National Center for 614

Atmospheric Research, Boulder, Colorado, USA, 88 pp. (Available online at doi: 615

http://dx.doi.org/10.5065/D6DZ069) 616

Storm Prediction Center, 2017: CAPE for Instability Descriptors. Storm Prediction Center, 617

National Centers for Environmental Prediction, NOAA / National Weather Service, 618

Norman, OK, U.S.A., (Available online at 619

http://www.spc.noaa.gov/misc/tables/capetext.htm.) 620

26

Tang, X.-D., M.-J. Yang, and Z.-M. Tan, 2012: A modeling study of orographic convection 621

and mountain waves in the landfalling typhoon Nari (2001). Q. J. Royal Meteor. Soc., 622

138, 419-438. 623

Tewari, M., F. Chen, W. Wang, J. Dudhia, M. A. LeMone, K. Mitchell, M. Ek, G. Gayno, J. 624

Wegiel, and R. H. Cuenca, 2004: Implementation and verification of the unified NOAH 625

land surface model in the WRF model. Proc. 20th conference on weather analysis and 626

forecasting /16th conference on numerical weather prediction, Seattle, Washington, 627

USA, 11–15. 628

Trenberth, K. E., A. Dai, R. M. Rasmussen, and D. B. Parsons, 2003: The changing 629

character of precipitation. Bull. Amer. Meteor. Soc., 84, 1205-1217. 630

United Nations Office for the Coordination of Humanitarian Affairs (UN-OCHA), 2016: Sri 631

Lanka - Floods and landslides Situation Report No. 2. UN-OCHA, Colombo, Sri 632

Lanka, 5pp. (Available online at 633

https://reliefweb.int/sites/reliefweb.int/files/resources/OCHA%20ROAP_SL_Sitrep2.pdf634

) 635

Wang, S. Y., B. M. Buckley, J. H. Yoon, and B. Fosu, 2013: Intensification of premonsoon 636

tropical cyclones in the Bay of Bengal and its impacts on Myanmar. J. Geophys. Res.- 637

Atmos., 118, 4373-4384. 638

Webster, P. J., V. O. Magana, T. N. Palmer, J. Shukla, R. A. Tomas, M. Yanai, and T. 639

Yasunari, 1998: Monsoons: Processes, predictability, and the prospects for prediction. 640

J. Geophys. Res.- Oceans., 103, 14451-14510. 641

27

Wu, G., Y. Guan, T. Wang, Y. Liu, J. Yan, and J. Mao, 2010: Vortex genesis over the Bay 642

of Bengal in spring and its role in the onset of the Asian Summer Monsoon. Sci. China, 643

Ser. D Earth Sci., 54, 1-9. 644

Yang, M.J., D.L. Zhang, and H.L. Huang, 2008: A modeling study of Typhoon Nari (2001) 645

at landfall. Part I: Topographic effects. J. Atmos. Sci., 65, 3095-3115. 646

Zipser, E. J., C. Liu, D. J. Cecil, S. W. Nesbitt, and D. P. Yorty, 2006: Where Are the Most 647

Intense Thunderstorms on Earth? Bull. Amer. Meteor. Soc., 87, 1057-1071. 648

649

650

List of Figures 651

Fig. 1: (a) Accumulated rainfall distribution of the event from 15–16 May 2016, observed by 652

Department of Meteorology weather stations. (Elevation is shaded with 250m contours, showing 653

areas above 250m of the central mountains) (b) Distribution map of the number of people affected 654

by the floods and landslides due to the extreme event from 14–20 May 2016 (Source: DMC 655

(2016)). (c) Map of the study domain with topography. Shaded point circle refers to the location of 656

Colombo weather station (6.90oN; 79.87oE) in the western coast of Sri Lanka. A-B transect used in 657

vertical cross sections. 658

659

Fig. 2: The Sea level pressure (hPa) (in blue colored contours in 2 hPa interval), Surface winds (10m) 660

(m s-1) (in vector arrows colored based on magnitude) and 500 hPa level geopotential height (in 661

green colored contours in 10m interval in the range of 5700 – 6000m) from 6-hourly National 662

Centers for Environmental Prediction Climate Forecast System version 2 (NCEP-CFSv2) data 663

valid at (a) 0600 UTC 14 May (b) 0600 UTC 15 May and (c) 0600 UTC 16 May 2016. Meteosat-7 664

VISSR (visible and infrared spin-scan radiometer) Indian Ocean data coverage thermal infrared 665

cloud imageries (Dundee 2016) 2(d, e, f) valid at corresponding time as in 2(a, b, c). 666

667

Fig. 3: Accumulation daily rainfall for 15 and 16 May 2016 (a-b) observed by the Department of 668

Meteorology, and corresponding accumulated rainfall from the model simulation (c-d). Pointed 669

locations correspond to the stations with maximum daily rainfall amount. Elevation shows above 670

250m of mean sea level only. 671

28

672

Fig. 4: Time series of observed (red line), and model simulated (blue line) rainfall at Colombo weather 673

station (6.90oN; 79.87oE) from 0000 LST 15 May to 0000 LST 17 May 2016. 674

675

Fig. 5: Scatter plot of hourly rainfall derived from the model results verses the observed hourly rainfall 676

for Colombo weather station location from 0000 LST 15 May to 0000 LST 17 May 2016. 677

678

Fig. 6: Model simulated water vapor fluxes at 925hPa and 10m wind vectors at (a) 1200 UTC 14 May, 679

(b) 0600 UTC, (c) 1200 UTC, (d) 1500 UTC 15 May and (e) 0000 UTC, (f) 0600 UTC 16 May 2016. 680

Elevation shows above 250m of mean sea level only (shaded/line terrain contour interval 250m). 681

682

Fig. 7: Model simulated Low level (925 hPa) wind convergence (negative), divergence (positive) and 683

zonal and meridional divergent wind vectors at (a) 1200 UTC 14 May, (b) 0600 UTC, (c) 1200 684

UTC, (d) 1500 UTC 15 May and (e) 0000 UTC, (f) 0600 UTC 16 May 2016. Elevation shows above 685

250m of mean sea level only (terrain contour (line) interval 250m). 686

Fig. 8: Model simulated convective available potential energy (CAPE) (colored) (units: J Kg-1) and 687

equivalent potential temperature (contours at an interval of 4 K) at lowest model level (~160m 688

above surface) valid at (a) 1200 UTC 14 May, (b) 1800 UTC 14 May, and (c) 0000 UTC 15 May 689

2016. Blank areas denote areas with CAPE below 500 J Kg-1. 690

691

Fig. 9: Model simulated equivalent potential temperature (in line contours at an interval of 3 K), 692

vertical velocity (shaded contours) in vertical cross section of the A-B transect (given in Fig. 1c) 693

valid at (a) 0600 UTC, (b) 1200 UTC, (c) 1500 UTC and (d) 2100 UTC 15 May 2016. Simulated 694

Radar reflectivity and wind vectors (e-h) in corresponding times respectively as in Fig.9 (a, b, c, d). 695

696

Fig. 10: The 48h (15-16 May, 2016) accumulated rainfall difference between the sensitivity cases and 697

control run. (a) Topo-Z (No Topography), (b) Topo-Q (topographic height 25% of the control run), 698

(c) Topo-H (topographic height 50% of the control run). Elevation shows above 250m of mean sea 699

level only (terrain contour interval in 100m) for the respective cases. 700

701

Fig. 11: Percentage change (%) of 48h (15-16 May, 2016) accumulated total rainfall (average / 702

maximum) for cases Topo-Z (No Topography), Topo-Q (topographic height 25% of the control run) 703

and Topo-H (topographic height 50% of the control run) cases with respect to the control 704

simulation for (a) Colombo municipal region (rectangle A in Fig. 10) and (b) western slope of the 705

central mountains (rectangle B in Fig. 10) 706

707

29

Fig. 12: Schematic presentation of the possible mechanism for the observed extreme rainfall in the 708

western coast of Sri Lanka, showing interactions of westerlies and LPS, and effects of central 709

mountains of Sri Lanka. Terrain height contours are drawn every 250 m. Vectors and shading 710

denote the 10 m wind vectors and hourly accumulated rainfall (mm) at 2200 UTC 15 May 2016, 711

respectively. 712

713

714

715

716

717

718

719

720

721

722

723

724

725

726

727

728

729

730

731

732

733

734

735

736

737

738

739

740

741

742

743

744

745

30

746

747

748

749

750

751

752

753

754

755

756

757

758

759

760

761

762

763

764

765

766

767

768

769

770

Fig. 1: (a) Accumulated rainfall distribution of the event from 15–16 May 2016, observed by 771

Department of Meteorology weather stations. (Elevation is shaded with 250m contours, showing 772

areas above 250m of the central mountains) (b) Distribution map of the number of people affected 773

by the floods and landslides due to the extreme event from 14–20 May 2016 (Source: DMC 774

(2016)). (c) Map of the study domain with topography. Shaded point circle refers to the location of 775

Colombo weather station (6.90oN; 79.87oE) in the western coast of Sri Lanka. A-B transect used in 776

vertical cross sections. 777

Figures

(a) (b)

(c)

31

Fig. 2: The Sea level pressure (hPa) (in blue colored contours in 2 hPa interval), Surface winds (10m) 778

(m s-1) (in vector arrows colored based on magnitude) and 500 hPa level geopotential height (in 779

green colored contours in 10m interval in the range of 5700 – 6000m) from 6-hourly National 780

Centers for Environmental Prediction Climate Forecast System version 2 (NCEP-CFSv2) data 781

valid at (a) 0600 UTC 14 May (b) 0600 UTC 15 May and (c) 0600 UTC 16 May 2016. Meteosat-7 782

VISSR (visible and infrared spin-scan radiometer) Indian Ocean data coverage thermal infrared 783

cloud imageries (Dundee 2016) 2(d, e, f) valid at corresponding time as in 2(a, b, c). 784

(a) (d)

(b) (e)

(c) (f)

32

785

786

787

788

789

790

791

792

793

794

795

796

797

798

799

800

801

802

803

804

805

806

807

808

809

810

811

812

813

814

Fig. 3: Accumulation daily rainfall for 15 and 16 May 2016 (a-b) observed by the Department of 815

Meteorology, and corresponding accumulated rainfall from the model simulation (c-d). Pointed 816

locations correspond to the stations with maximum daily rainfall amount. Elevation shows above 817

250m of mean sea level only. 818

819

820

821

822

(a) (b)

(c) (d)

33

823

824

Fig. 4: Time series of observed (red line), and model simulated (blue line) rainfall at Colombo weather 825

station (6.90oN; 79.87oE) from 0000 LST 15 May to 0000 LST 17 May 2016. 826

827

828

829

Fig. 5: Scatter plot of hourly rainfall derived from the model results verses the observed hourly rainfall 830

for Colombo weather station location from 0000 LST 15 May to 0000 LST 17 May 2016. 831

832

833

834

34

835

Fig. 6: Model simulated water vapor fluxes at 925hPa and 10m wind vectors at (a) 1200 UTC 14 May, 836

(b) 0600 UTC, (c) 1200 UTC, (d) 1500 UTC 15 May and (e) 0000 UTC, (f) 0600 UTC 16 May 2016. 837

Elevation shows above 250m of mean sea level only (shaded/line terrain contour interval 250m). 838

839

840

(a) (d)

(b) (e)

(c) (f)

35

841

Fig. 7: Model simulated Low level (925 hPa) wind convergence (negative), divergence (positive) and 842

zonal and meridional divergent wind vectors at (a) 1200 UTC 14 May, (b) 0600 UTC, (c) 1200 843

UTC, (d) 1500 UTC 15 May and (e) 0000 UTC, (f) 0600 UTC 16 May 2016. Elevation shows above 844

250m of mean sea level only (terrain contour (line) interval 250m). 845

(a) (d)

(b) (e)

(c) (f)

36

846

847

848

849

850

851

852

853

854

855

856

857

858

859

860

861

862

863

864

865

866

867

868

869

870

871

872

873

874

875

876

877

Fig. 8: Model simulated convective available potential energy (CAPE) (colored) (units: J Kg-1) and 878

equivalent potential temperature (contours at an interval of 4 K) at lowest model level (~160m 879

above surface) valid at (a) 1200 UTC 14 May, (b) 1800 UTC 14 May, and (c) 0000 UTC 15 May 880

2016. Blank areas denote areas with CAPE below 500 J Kg-1. 881

882

883

(a)

(b)

(c)

37

884

885

886

887

888

889

890

891

892

893

894

895

896

897

898

899

900

901

902

903

904

905

906

907

908

909

910

911

912

913

914

915

916

917

Fig. 9: Model simulated equivalent potential temperature (in line contours at an interval of 3 K), 918

vertical velocity (shaded contours) in vertical cross section of the A-B transect (given in Fig. 1c) 919

valid at (a) 0600 UTC, (b) 1200 UTC, (c) 1500 UTC and (d) 2100 UTC 15 May 2016. Simulated 920

Radar reflectivity and wind vectors (e-h) in corresponding times respectively as in Fig.9 (a, b, c, d). 921

(a) (e)

(c) (g)

(d) (h)

(f) (b)

38

922

923

924

925

926

927

928

929

930

931

932

933

934

Fig. 10: The 48h (15-16 May, 2016) accumulated rainfall difference between the sensitivity cases and 935

control run. (a) Topo-Z (No Topography), (b) Topo-Q (topographic height 25% of the control run), 936

(c) Topo-H (topographic height 50% of the control run). Elevation shows above 250m of mean sea 937

level only (terrain contour interval in 100m) for the respective cases. 938

939

940

941

942

943

Fig. 11: Percentage change (%) of 48h (15-16 May, 2016) accumulated total rainfall (average / 944

maximum) for cases Topo-Z (No Topography), Topo-Q (topographic height 25% of the control run) 945

and Topo-H (topographic height 50% of the control run) cases with respect to the control 946

simulation for (a) Colombo municipal region (rectangle A in Fig. 10) and (b) western slope of the 947

central mountains (rectangle B in Fig. 10) 948

949

(a)

(b)

(a)

(b) (c)

39

950

951

952

953

954

955

956

957

958

959

960

961

962

963

964

965

966

967

968

969

970

971

972

973

974

975

976

977

Fig. 12: Schematic presentation of the possible mechanism for the observed extreme rainfall in the 978

western coast of Sri Lanka, showing interactions of westerlies and LPS, and effects of central 979

mountains of Sri Lanka. Terrain height contours are drawn every 250 m. Vectors and shading 980

denote the 10 m wind vectors and hourly accumulated rainfall (mm) at 2200 UTC 15 May 2016, 981

respectively. 982

983

984

985

986

987

A B

Convergence Zone

Westerly Winds Central Mountains

Low Pressure System

40

988

989

990

991

List of Tables 992

993

Table 1: Overview of the configurations of physics and dynamics used in the WRF model 994

995

Table 2: R-squared value, Standard error of the regression, Mean Absolute Error and Mean 996

Fractional Bias for the hourly rainfall of Colombo weather station from 0000 LST 15 May - 997

0000 LST 17 May, 2016. 998

999

1000

1001

1002

1003

1004

1005

1006

1007

1008

1009

1010

1011

1012

1013

1014

41

1015

1016

1017

Table 1: Overview of the configurations of physics and dynamics used in the WRF

model

Model features Configurations

Dynamics Non-hydrostatic

Model domain 1 (3°18' N - 14°12' N, 74°14' E - 88°9' E)

Center of the domain 8.8 ⁰N, 81.2 ⁰E

Horizontal grid 500 X 400 (zonal and meridional respectively)

Horizontal grid distance 3 km

Map projection Mercator

Horizontal grid system Arakawa C-grid

Vertical coordinates / levels Terrain-following hydrostatic pressure vertical

coordinate with 45 vertical levels

Model top 50hPa

Four-dimensional data

assimilation (FDDA) Yes (Gird nudging)

Feedback No

Microphysics Lin et al. scheme (Lin et al., 1983)

Longwave Radiation scheme Rapid Radiative Transfer Model (RRTM) scheme

(Mlawer et al., 1997)

Shortwave Radiation scheme Dudhia scheme (Dudhia 1989)

Surface layer

parameterization

Monin–Obukhov surface layer scheme (Beljaars

1995)

Planetary boundary layer

(PBL) Yonsei University (YSU) scheme (Hong et al., 2006)

Cumulus scheme (CPS) Kain-Fritsch (new Eta) scheme (Kain 2004)

Land surface processes Unified Noah land surface model (Tewari et al.,

2004)

1018

1019

1020

1021

42

1022

1023

1024

1025

1026

Table 2: R-squared value, Standard error of the regression, Mean Absolute

Error and Mean Fractional Bias for the hourly rainfall of Colombo weather

station from 0000 LST 15 May - 0000 LST 17 May, 2016.

Index Abbreviation Value

Coefficient of determination (R-squared) R2 0.7

Standard error of the regression (mm) S 5.6

Mean Absolute Error (mm) MAE 1.7

Mean Fractional Bias (%) MFB 0.4

1027