Elsevier

Atmospheric Research

Volume 262, November 2021, 105779
Atmospheric Research

Impact of surface roughness parameterizations on tropical cyclone simulations over the Bay of Bengal using WRF-OML model

https://doi.org/10.1016/j.atmosres.2021.105779Get rights and content

Highlights

  • Dynamic variations in surface drag improves on the prediction of pre-monsoon cyclones.

  • Post-monsoon storms are highly sensitive to the variations in moist enthalpy coefficients.

  • Storm induced ocean response is sensitive to changes in surface drag and enthalpy coefficients.

  • Garratt based enthalpy coefficients better capture the variations in moisture budget parameters.

Abstract

The impact of different surface roughness schemes in air-sea coupling on simulating tropical cyclones (TCs) over the Bay of Bengal is analyzed using Weather Research and Forecasting-Ocean Mixed layer (WRF-OML) modeling system. The sensitivity of three surface roughness schemes is tested by conducting three experiments for seven TCs namely Phailin (2013), Lehar (2013), Hudhud (2014), Vardah (2016), Gaja (2018), Fani (2019) and Amphan (2020). The first experiment (Opt0) is configured with surface drag and moist enthalpy from Garratt formulations, the second (Opt1) uses surface drag from Donelan and constant moist enthalpy formulations and the third experiment (Opt2) employs the modified moist enthalpy from Garratt formulations along with Donelan drag. Results of predicted track, intensities, precipitation and structure of the cyclones are highly sensitive to the surface exchange coefficients formulated through the roughness parameterization. The Opt2 followed by Opt1 experiments captured the deepening and mature phases of the storm close to the observed estimates for both pre- and post-monsoon TCs, by improving the ocean-atmosphere feedback of surface energy fluxes. Comparison of simulated moisture convergence, transport and precipitation with observations for all cyclones suggest that the Opt2 provides an improved representation of surface enthalpy fluxes playing a crucial role in the moisture transport and cloud microphysical processes. The improved simulation of TCs with Opt1 and Opt2 over Opt0 is attributed to the realistic simulation of mixed layer deepening and sea surface temperature cooling effects in the model. An interesting result found from the WRF-OML simulations with three roughness schemes is the maximum moisture transport is mainly concentrated on the right sectors while the moisture convergence and precipitation are distributed on the left sectors of the cyclones.

Introduction

Tropical Cyclones (TCs) are intense cyclonic vortices that develop over tropical oceans under favorable ocean-atmospheric conditions. TCs are associated with strong surface winds, heavy precipitation, occasionally induce storm surges and create huge havoc all along the coast during their landfall. The rising sea surface temperatures and enhanced ocean heat potentials under global warming scenario indirectly promote the tropical cyclone genesis and intensification potential (Sun et al., 2017) and it is considered as one of the great concerns for the populated coastal regions. The coastal Indian areas with a high density of population are highly vulnerable to cyclone disasters due to the TCs that form over the North Indian Ocean (NIO) in pre-monsoon (April and May) and post-monsoon (October, November and December) periods. The yearly frequency of TCs formed over the Arabian Sea (AS) and Bay of Bengal (BoB) is in the ratio of 1:4 (Dube et al., 1997). Accurate prediction of TCs with reference to their movement and intensity is highly crucial in disaster management. In particular, precise forecasting of TCs over NIO is a challenging task for numerical forecasters due to the shorter lifespan of TCs compared to the other basins.

The favorable conditions for the development of TCs include increased sea surface temperature (SST; ≥ 26.5 °C), relatively week vertical shear of horizontal winds, low-level cyclonic vorticity, minimum Coriolis force, incipient low pressure, conditional instability of the atmosphere and high mid-tropospheric relative humidity (Gray, 1975; Palmen, 1948; Riehl, 1948). The incipient vortex intensifies into TC over warmer ocean regions and decays when it passes over colder ocean regions and after crossing the coast (Gray, 1968). The intensification of TC is determined by the transport of air-sea fluxes, underlying mesoscale processes and inner core dynamics, moist convection and cloud-scale physical processes (Emanuel, 1986; Weatherford and Gray, 1988; Xu and Wang, 2010; Ma et al., 2015). The movement of the TCs mainly depends upon the large-scale steering currents (easterlies in tropics and westerlies in mid-latitudes), beta effect and their interaction with storm-scale motions (Chan and Gray, 1982; Chan, 2005). The existing numerical models can able to capture the large-scale steering flow precisely, but still, there is a challenge to resolve the inner core dynamics of TCs and the mesoscale processes associated with lower boundary conditions (Zhang et al., 2011). In the last two decades, the skill of numerical models for prediction of cyclone's track has been greatly improved but the intensity prediction still remains a challenging problem (Rappaport et al., 2009). Thus, advanced high-resolution ocean-atmospheric models with data assimilation, air-sea flux parameterization, and improved physics are being adopted for achieving better accuracy in TC forecasting in all the tropical ocean basins including the NIO (Gopalakrishnan et al., 2011; Srinivas et al., 2010; Srinivas et al., 2016; Raju et al., 2011; Chandrasekar and Balaji, 2012, Chandrasekar and Balaji, 2016; Tallapragada et al., 2013a, Tallapragada et al., 2013b; Yesubabu et al., 2014, Yesubabu et al., 2020; Subramani et al., 2014; Greeshma et al., 2015, Greeshma et al., 2019; Rao et al., 2018; Bonaldo et al., 2018; Ramakrishna et al., 2019; Baisya et al., 2020).

The mechanism of the TC intensification from the air-sea interaction perspective is well explained by the wind-induced surface heat exchange (WISHE) theory (Emanuel, 1995). As per the WISHE paradigm, the intensification of the TC vortex occurs through the supply of energy from the ocean surface to the atmosphere in the form of heat and moisture fluxes which are modulated by the winds. The surface fluxes critically depend on the surface winds and upper ocean heat content (Vissa et al., 2012). The upper-ocean acts as a bridge for TC intensification through atmosphere-wave coupling and sometimes also acts as a barrier due to negative feedback of cyclone which induces upper ocean cooling by enhanced vertical mixing during the passage of the storm (Balaguru et al., 2012). Since the turbulent heat and moisture fluxes highly influence the intensification of the TCs over the ocean (Shen and Ginis, 2003; Wang et al., 2001; Green and Zhang, 2013), the accurate parameterization of fluxes in numerical models is crucial for the better prediction of these storms.

The surface exchange coefficients of moist enthalpy (Ck) and drag/ momentum (CD) are the two most important parameters that affect the air-sea fluxes in the simulation of TCs (Ooyama, 1969; Emanuel, 1995). Several studies explained the role of the surface roughness schemes in the energy transfer between ocean and atmosphere through thermodynamic coupling and its influence on the simulation of TCs using coupled models (Bender and Ginis, 2000; Lin et al., 2005; Olabarrieta et al., 2012; Chen et al., 2013; Samala et al., 2013; Srinivas et al., 2016; Kwon and Kim, 2017; Singh and Tyagi, 2019; Ricchi et al., 2019; Greeshma et al., 2019). In particular, the surface drag can modulate the momentum and moist enthalpy which influence the intensification of primary and secondary circulations through the boundary layer spin-up mechanism (Shapiro and Willoughby, 1982; Bui et al., 2009). Moreover, CD is a key parameter that controls the relationship between maximum wind and minimum pressure near the surface (Bao et al., 2012; Smith et al., 2013). Previous studies reported the limitations of existing formulations for bulk exchange coefficients over the sea surface especially for the very strong winds during cyclones (Bender and Ginis, 2000; Lin et al., 2005; Davis et al., 2008; Chen et al., 2013; Srinivas et al., 2016; Greeshma et al., 2019). The upper-ocean surface continuously generates waves which alter the roughness and static stability over the sea surface. Many researchers therefore, analyzed the impact of exchange coefficients and surface drag on simulations of maximum sustained winds of TCs (Bao et al., 2002, Bao et al., 2012; Davis et al., 2008; Nolan et al., 2009). Therefore detailed analyses of the surface wind stress or drag parameterization is essential for the tropical cyclone simulation (Hsu et al., 2017).

The surface waves enhance the surface drag by modulating the roughness of the ocean surface, primarily for the mean winds above 5 ms−1 (Smith, 1988; Fairall et al., 2003). In the early years, there were no direct flux calculations beyond the maximum steady winds of 22 ms−1 in the open sea, for which data were extrapolated from low wind speeds. Previous studies have computed the CD and surface roughness length for TC winds using GPS dropsonde wind speed profiles (Powell et al., 2003), stepped frequency microwave radiometer fixed on the aircraft (Bell et al., 2012) and buoys' anemometer (Potter et al., 2015). According to Powell et al. (2003), CD remains constant with wind speed above 33 ms−1. Donelan et al. (2004) also demonstrated similar results by tank experiment for wind speeds more than 30 ms−1. Apart from CD, the higher ocean heat content associated with the presence of deeper ocean mixed layer depth (MLD) can favor the intensification of the TCs by enhancing moist enthalpy fluxes during the passage of the TCs (Shay and Uhlhorn, 2008). At the air-sea interface moisture transport in the area of high ocean heat content enhances moist enthalpy fluxes to the cyclones and helps to its rapid intensification phase of the TC (Jaimes et al., 2015). The changes in moist enthalpy exchange coefficients therefore, play an essential role on the intensification of TC (Ooyama, 1969; Braun and Tao, 2000; Bryan, 2012; Green and Zhang, 2014).

In this study, the sensitivity of TC track, intensity and structure predictions to the surface roughness parameterization is tested for seven TCs using the atmosphere-ocean model (WRF-OML). We considered seven intense cyclones Phailin (2013), Lehar (2013), Hudhud (2014), Vardah (2016), Gaja (2018), Fani (2019) and Amphan (2020) that formed over the BoB in NIO during 2013–2020. In the OML model, CD mainly controls the mixing processes at the ocean surface (Pollard et al., 1973). The intensification of simulated tropical storms in numerical models depends on the moisture and energy fluxes across the air-sea interface which is influenced by the moist enthalpy coefficient (Ck). A change of moist enthalpy coefficient (Ck) leads to a change in buoyancy which in turn affects the wind stress (Green and Zhang, 2014). The main purpose of the present study is to understand the role of ocean surface bulk exchange coefficients of drag (CD) and moist enthalpy (Ck) on the intensification of the TCs. Though the ocean physical processes are highly relevant for the TC evolution, there are very few studies attempted over BoB on analyzing the performance of surface roughness schemes using ocean-atmospheric models (Singh and Tyagi, 2019; Greeshma et al., 2019). This study employs the ocean mixed layer (OML) model which is a simplified 1-D ocean model to compute the physical process of wind-driven ocean mixing and mixed layer deepening. During the passage of TC, the vertical mixing and resultant sea surface cooling are the major ocean surface feedbacks supplied to the atmospheric model. Many studies highlighted that the storm-induced SST feedback can be grossly represented using computationally simple 1-D mixed layer models (Yablonsky and Ginis, 2009; Ginis et al., 2010; Wang and Duan, 2012). The model estimates the effect of entrainment of cold subsurface waters to the surface proportional to supplied surface wind stress. Numerous studies have suggested that OML can be run with limited computational resources for studying the ocean response on TCs and it adequately reproduces the cyclone-induced SST cooling (Yablonsky and Ginis, 2009; Ginis et al., 2010; Greeshma et al., 2015; Yesubabu et al., 2020; Li et al., 2020). In particular, although OML is a 1-D model, it is capable of estimating the storm-induced SST feedback at the same time step to the atmospheric model thereby realistically representing the ocean-atmosphere interaction (Pollard et al., 1973; Wang and Duan, 2012). Also, few studies (Wang and Duan, 2012; Yesubabu et al., 2020) highlighted that the OML has tendency to underestimate the TC induced SST cold wake, depending on the characteristics of TC, in particular the size and translation movement.

The paper is structured as follows. The cyclonic storms considered for this study are described in Section 2; the details of the model initialization, sensitivity experiments and observations used for the study are discussed in Section 3. Section 4 discusses the results of the simulations for the seven TCs. The summary and conclusions of the analysis are provided in Section 5.

Section snippets

Description of cyclonic storms

The details of the seven TCs formed over the BoB considered in this study are provided in Table 1. Among the seven TCs, five storms (Phailin, Lehar, Hudhud,Vardah and Gaja) are post-monsoon cyclonic storms and the remaining two (Fani and Amphan) are pre-monsoon storms. The detailed history of track and intensity of these TCs are available from annual reports of IMD (http://www.rsmcnewdelhi.imd.gov.in).

The extremely severe cyclonic storm (ESCS) Phailin initially formed as a low pressure over the

Model domain and physics configuration

In this study, we have used the Advanced Research Version (V4.2.1) of Weather Research and Forecast model (WRF-ARW, Skamarock and Klemp, 2008) along with its ocean mixed layer model (OML, Pollard et al., 1973). A series of sensitivity experiments with the surface roughness parameterizations are conducted with WRF-OML to assess the impact of moist enthalpy and drag coefficients between the ocean and atmosphere on the simulated storm characteristics (Table 2). The model is configured with two

Results and discussions

The results of simulations are presented from the model inner domain (3 km) along with the corresponding observations. Below we present the results of simulations for track and intensity using different roughness schemes for all TCs. This is followed by diagnostic analysis for parameters such as MLD, SST, surface energy fluxes, precipitation, surface winds, RMW, radius-height sections of tangential winds and temperature deviation to infer the differences by different schemes and to assess their

Summary and conclusions

In this study, we explored the sensitivity of the three surface roughness schemes for seven BoB tropical cyclones (Phailin, Lehar, Hudhud, Vardah, Gaja, Fani, and Amphan) representing different intensities and seasons. The WRF-OML simulations of the seven TCs revealed large sensitivity of the structure and intensity predictions to a large extent and the track predictions to a moderate extent to the roughness parameterization. Results indicate that variation in the surface wind stress (CD) and

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors thank the India Meteorological Department for providing the access to observational datasets like best track data of tropical cyclone and cyclone reports. The authors acknowledge NCAR/USA for the public assess of the WRF-ARW model and NCEP for the access of the Global Forecast System model analysis used in the numerical simulations. Special Sensor Microwave Imager remote sensing data and Tropical Rainfall Measuring Mission data are obtained from NASA.The MLD data collected from the

Author statement

Nanaji Rao Nellipudi: Data curation, Formal analysis, Software, Investigation, Writing - Original draft.

V. Yesubabu: Conceptualization, Methodology, Writing- Reviewing and Editing, Supervision.

C.V. Srinivas: Methodology, Writing- Reviewing and Editing, Supervision.

Naresh Krishna Vissa: Writing- Reviewing and Editing.

Sabique Langodan: Writing- Reviewing and Editing.

References (105)

  • D. Bonaldo et al.

    Wind storminess in the Adriatic Sea in a climate change scenario

    ActaAdriatica

    (2018)
  • S.A. Braun et al.

    Sensitivity of high-resolution simulations of Hurricane Bob (1991) to planetary boundary layer parameterizations

    Mon. Wea. Rev.

    (2000)
  • W. Brutsaert

    A theory for local evaporation (or heat transfer) from rough and smooth surfaces at ground level

    Water Resourc. Res.

    (1975)
  • G.H. Bryan

    Effects of surface exchange coefficients and turbulence length scales on the intensity and structure of numerically simulated hurricanes

    Mon. Wea. Rev.

    (2012)
  • H.H. Bui et al.

    Balanced and unbalanced aspects of tropical cyclone intensification

    Q. J. R. Meteorol. Soc.

    (2009)
  • J.C.L. Chan

    The physics of tropical cyclone motion

    Annu. Rev. Fluid Mech.

    (2005)
  • J.C.L. Chan et al.

    Tropical cyclone movement and surrounding flow relation- ships

    Mon. Wea. Rev.

    (1982)
  • R. Chandrasekar et al.

    Sensitivity of tropical cyclone Jal simulations to physics parameterizations

    J. Ear. Syst. Sci.

    (2012)
  • R. Chandrasekar et al.

    Impact of physics parameterization and 3DVAR data assimilation on prediction of tropical cyclones in the Bay of Bengal region

    Nat. Haz.

    (2016)
  • H. Charnock

    Wind stress on a water surface

    Q. J. R. Meteorol. Soc.

    (1955)
  • F. Chen et al.

    Coupling an advanced land–surface hydrology model with the Penn State–NCAR MM5 modeling system. Part I: model implementation and sensitivity

    Mon. Wea. Rev.

    (2001)
  • S.S. Chen et al.

    Directional wind– wave coupling in fully coupled atmosphere–wave–ocean models: results from CBLAST-hurricane

    J. Atmos. Sci.

    (2013)
  • J.A. Cummings et al.

    Ocean data impacts in Global HYCOM

    J. Atmos. Ocean. Technol.

    (2014)
  • R.A. Dare et al.

    Sea surface temperature response to tropical cyclones

    Mon. Wea. Rev.

    (2011)
  • H.P. Dasari et al.

    On the movement of tropical cyclone LEHAR.Ear

    Syst. Env.

    (2017)
  • C.A. Davis et al.

    Prediction of landfalling hurricanes with the advanced hurricane WRF model

    Mon. Wea. Rev.

    (2008)
  • M.A. Donelan et al.

    On the limiting aerodynamic roughness of the ocean in very strong winds

    Geo. Res. Let.

    (2004)
  • S.K. Dube et al.

    Storm surge in the Bay of Bengal and Arabian Sea: the problem and its Prediction

    Mausam

    (1997)
  • J. Dudhia

    Numerical study of convection observed during winter monsoon experiment using a mesoscale two-dimensional model

    J. Atmos. Sci.

    (1989)
  • K.A. Emanuel

    An air–sea interaction theory for tropical cyclones. Part I: steady-state maintenance

    J. Atmos. Sci.

    (1986)
  • K.A. Emanuel

    Sensitivity of tropical cyclones to surface exchange coefficients and a revised steady-state model incorporating eye dynamics

    J. Atmos. Sci.

    (1995)
  • C.W. Fairall et al.

    Bulk parameterization of air–sea fluxes: Updates and verification for the COARE algorithm

    J. Clim.

    (2003)
  • J.R. Garratt

    Review of drag coefficients over oceans and continents

    Mon. Wea. Rev.

    (1997)
  • W. Gemmill et al.

    Daily Real Time Global Sea Surface Temperature High Resolution Analysis

    (2007)
  • I. Ginis et al.

    Numerical modeling of air–sea interaction in tropical cyclones

  • S. Glenn et al.

    Stratified coastal ocean interactions with tropical cyclones

    Nat. Commun.

    (2010)
  • S. Gopalakrishnan et al.

    Hurricane Weather and Research and Forecasting (HWRF) Model: Scientific Documentation

    (2011)
  • W.M. Gray

    Global view of the origin of tropical disturbances and storms

    Mon. Wea. Rev.

    (1968)
  • W.M. Gray

    Tropical Cyclone Genesis

    (1975)
  • B.W. Green et al.

    Impacts of air–sea flux parameterizations on the intensity and structure of tropical cyclones

    Mon. Wea. Rev.

    (2013)
  • B.W. Green et al.

    Sensitivity of tropical cyclone simulations to parametric uncertainties in air–sea fluxes and implications for parameter estimation

    Mon. Weather Rev.

    (2014)
  • M. Greeshma et al.

    Sensitivity of tropical cyclone predictions in the coupled atmosphere–ocean model WRF-3DPWP to surface roughness schemes

    Met. App.

    (2019)
  • M.M. Greeshma et al.

    Real-time numerical simulation of tropical cyclone Nilam with WRF: experiments with different initial conditions, 3D-Var and ocean mixed layer model

    Nat. Haz.

    (2015)
  • H. Hersbach

    The ERA5 global reanalysis

    Q. J. R. Met. Soc.

    (2020)
  • J. Hsu et al.

    Estimates of surface wind stress and drag coefficients in typhoon megi

    J. Phys. Oceanogr.

    (2017)
  • G.J. Huffman et al.

    The TRMM Multisatellite Precipitation Analysis (TMPA): quasi-global, multiyear, combined-sensor precipitation estimates at fine scales

    J. Hydro. Meteorol.

    (2007)
  • M.J. Iacono et al.

    Radiative forcing by long–lived greenhouse gases: calculations with the AER radiative transfer models

    J. Geo. Res.

    (2008)
  • IMD 2020a. Report on Cyclonic disturbances over North Indian Ocean during 2019. RSMCTropical Cyclones Report No....
  • IMD

    Report on Super Cyclonic storm Amphan. 16–21 May 2020: Summary

    (2020)
  • B. Jaimes et al.

    Enthalpy and momentum fluxes during hurricane earl relative to underlying ocean features

    Mon. Wea. Rev.

    (2015)
  • Cited by (11)

    View all citing articles on Scopus
    View full text