Research Paper
Optimal grid resolution for the detection lead time of cyclogenesis in the North Indian ocean

https://doi.org/10.1016/j.jastp.2020.105289Get rights and content

Highlights

  • Early detection of pre-cyclonic vortices in the atmospheric column using OW technique.

  • Sensitivity experiments with five severe cyclones during pre- and post-monsoon seasons.

  • Characteristic differences in eddy lead detection time for pre- and post-monsoon cyclones.

  • Sensitivity of WRF model grid resolution to lead detection time of pre-cyclogenesis.

Abstract

Early detection of disturbances in the atmospheric column, its spatio-temporal behavior and evolution prior to tropical cyclogenesis over warm ocean surface is a challenging problem. High-resolution atmospheric models help to better understand the atmospheric dynamical conditions and instability mechanism triggered by intense air-sea interaction leading to the formation of tropical cyclogenesis. Present study performed a series of model-based sensitivity experiments to evaluate tropical cyclogenesis lead detection time for five cyclone cases with varying grid resolutions ranging between 27 km and 12 km in the north Indian Ocean. High-resolution Weather Research and Forecast (WRF) model examined the evolution of pre-cyclonic eddy vortices in the vertical atmospheric column associated with Aila, Thane, Mora, Ockhi, and the recent Vayu cyclones. Detection technique was based on Okubo-Weiss (OW) parameter that is very sensitive to varying grid resolutions. Study examined the sensitivity of pre-cyclogenesis lead detection time to WRF model grid resolutions. Significant findings from the study indicate a good skill in genesis prediction with approximately four days (~90 h) lead time in all cases irrespective of pre- and post-monsoon seasons and the domain size. First part of this study reveals characteristic difference in the eddy detection between pre- and post-monsoon cyclones and its lead detection time. The lead detection time using a grid resolution of 27 km resulted in a lag-time of nearly one day (~24 h) for pre-monsoon cyclogenesis as compared to the post-monsoon cases. Interestingly, the pre-cyclonic eddies are detected at higher atmospheric levels (450–650 hPa) for the pre-monsoon unlike post-monsoon developing cases in the lower tropospheric levels (800–950 hPa). Secondly, a 15 km grid resolution with OW technique would suffice early tropical cyclogenesis detection found optimum in context to lead detection time (~15 h) that better resolves the structure of pre-cyclonic eddies useful for advanced prediction.

Introduction

Tropical Cyclogenesis (TCG) characteristics, its evolution and early detection are challenging problems in tropical meteorology investigated during the past five decades or more. Though several studies were focused to understand TCG characteristics for the global ocean basins the skill level of prediction still remains questionable. The inherent physical processes that are quite complex involves both kinematics and thermodynamics interacting at different spatio-temporal scales through the boundary layer in the air-sea interface having a mutual feedback in the ocean-atmosphere system. The overall process is controlled by the atmospheric dynamical conditions at different vertical levels as well as the water mass characteristics extending below mixed layer depths and that play an important role in the overall flux exchange thereby modifying the characteristics of the atmospheric strata conducive for development of pre-cyclonic cyclogenesis. Precisely there is a lack of clarity and proper knowledge that primes the initiation and sustainment of mesoscale vortex circulation in TCG (Farfan and Zehnder, 1997). There has been consistent efforts to evaluate and better understand the background environmental physical processes and favorable conditions that enhances the atmospheric humidity levels in the mid-atmospheric levels that initiates the development of pre-cyclonic vortex in the mid- and lower atmospheric column (Yoo et al., 2015).

The initiation stage of cyclogenesis involves the development of atmospheric Mid-Level Vortices (MLVs) having horizontal scales ~ 100 km and rotating convective towers having horizontal scales ~ 10 km. The MLVs can persist in the atmosphere for duration of several days whereas the convective towers persist for few hours. Also it is extensively documented in literature that the maxima of mesoscale vorticity field are the key element for TCG detection scheme (Sinclair, 1997; Walsh, 1997). Popular theories on TCG and studies conducted for the global ocean basins postulates the fact that pre-cyclonic eddies initially develop in the upper atmosphere (Emanuel, 1986; Rotunno and Emanuel (1987); Simpson et al. (1997); Ritchie and Holland (1997); Bister and Emanuel (1997); Karyampudi and Pierce (2002); Montgomery et al. (2006); Nolan (2007); Dunkerton et al. (2009)). Further, there is a descending motion of eddies cluster directed towards the lower atmospheric boundary layer and finally its formation over the warm ocean surface (Dunkerton et al., 2009; Bhaskaran et al., 2018). Theories on cyclogenesis also proposed that the subsequent development of smaller vortex inside a tropical disturbance with time rapidly transformed into a tropical cyclone (Nicholls and Montgomery, 2013). Numerous cyclone vortex detection schemes have been established that used the intensity and strength of vorticity at a lower tropospheric level as one threshold for the cyclogenesis development such as the studies by Montgomery and Enagonio (1998); Vitart et al. (1999); Davis and Bosart (2001); Rogers and Fritsch (2001); Hendricks et al. (2004); Hodges et al. (2012); Tory et al. (2013a, 2013b, 2013c); Strachan et al. (2013) and Zarzycki et al. (2017). Transient nature of the atmospheric flow conditions and their intrinsic time dependence makes the detection characterization a complex problem. Also, in the detection schemes the occurrence of some disturbances that attains the minimum threshold spanning for short time duration can also lead to reporting of false alarms. Therefore, it is very essential to maintain a continuous monitoring of disturbances for a minimum time span in order to filter out the non-cyclonic eddies. In the detection schemes, it is evident from literature that many successful attempts were reported using the key independent parameter termed as the Okubo-Weiss (OW) (Zong and Wu, 2015; Tory et al., 2013a, 2013b, 2013c). This technique can be used to unveil the synoptic growth of atmospheric eddies and thereby provide the scope to identify cyclonic circulation prevalent at the pre-mature stages (Tory et al., 2013a, 2013b, 2013c; Rutherford et al., 2018).

There are also studies attempted by several researchers on the detection and observed variances in developing and non-developing cases (Simpson et al., 1968; McBride and Zehr, 1981; Berry and Hewson, 2007; Kerns and Zipser, 2009; Ross et al., 2009). More recently, Tory et al. (2013a, 2013b, 2013c) indicated that at centre the existence of a strain free rotation enveloped with either a rotational or deformation OW transition is an indicator for developing cases. Rutherford et al. (2018) established a threshold and structural criteria for cyclogenesis vorticity maxima in Atlantic Ocean using the OW detection tool. Their findings indicate a minimum required threshold value of 30 radians per 72 h for a tropical depression to form at an atmospheric level of 700 hPa. In addition, it was also reported that there is no restriction governing the alignment of vertical shear to be vertical in nature.

Tropical disturbances with varied intensity forms over the BoB and Arabian Sea (AS), and they are more frequent over the northern parts of BoB and eastern part of AS. The frequency is higher in BoB basin as compared to the AS. The environmental conditions conducive for tropical cyclogenesis formation can vary geographically and also seasonal dependence. In a broader perspective warm ocean temperature and weak vertical wind shear are the two necessary but not sufficient conditions for tropical cyclogenesis formation. There is a tendency for the wind shear to be weak during early summer. It increases as the tropical oceans warm strengthening the winter Hadley cell. As the cyclogenesis and its detection is partly influenced by both atmospheric and ocean parameters, it is imperative to understand the role and influence of mesoscale eddies and its variability during cyclogenesis during pre- and post-monsoon episodes. The pre-monsoon period covers the months of March-April-May (MAM) and the post-monsoon covers October-November-December (OND). Researchers have attempted to incorporate the seasonal dependency in North Indian Ocean (NIO) basin especially in the AS and BoB to determine the bimodal and annual cyclogenesis distributions and associated environmental factors (Camargo et al., 2007; Kikuchi and Wang, 2010; Evan and Camargo, 2011). Later Yanase et al. (2012) attempted to study the seasonal and intra-seasonal modulations highlighting on the environmental factors liable for cyclogenesis formation in the BoB basin. Their analysis have proven that the Genesis potential Index (GPI) and four factors (vertical shear, lower-tropospheric absolute vorticity, relative humidity over mid-tropospheric layers and potential intensity) for cyclogenesis are influenced by the combination of both seasonal and intra-seasonal characteristics.

The Weather Research and Forecasting (WRF) is a widely acclaimed mesoscale numerical weather prediction model designed for both atmospheric research and operational forecasting applications. The model can be used for wide range of meteorological applications with scales ranging from tens of meters to thousands of kilometers. The WRF model is being widely used by the scientific community for real-time weather forecast over the globe designed to serve both research and operational needs. The modeling system in WRF comprises of two dynamical solvers, referred to as the ARW (Advanced Research WRF) core and the NMM (non-hydrostatic Mesoscale Model) core. In context to the present study, there are reports that used high-resolution cloud resolving simulations using WRF to investigate the TC evolution stages from pre-existing warm core weak cyclonic vortex (Nolan, 2007; Yoo et al., 2015) in a realistic manner. Intensity of TCs and its sensitivity to horizontal grid resolution were investigated by several researchers over the global ocean basins. For practical needs, it is very essential to select the appropriate model domain size as well as grid resolutions as they play a major role in determining the quality of the model results (Sahoo et al., 2018). In a recent study related to cyclone tracks and intensity, Singh et al. (2019) reported on pronounced variations in intensity and cyclone tracks for changes made in the initial conditions. Also they mentioned on the importance of appropriate domain size selection and nesting options that has a direct bearing on effective representation of atmospheric eddies.

The inferences and limitations reported in the aforementioned studies has motivated the present study to critically examine and evaluate the role and influence of grid resolution and detection lead time for severe pre- and post-monsoon cyclones that developed over the AS and BoB basins. The study therefore focused on understanding the seasonal dependency of pre- and post-monsoon cyclones utilizing OW detection scheme as well as the sensitivity of WRF model grid resolution in the lead time of detection. The current study is structured into six sections. The first section covers aspect on the general introduction, problem formulation and objectives of the study. Methodology and model description along with the details of boundary condition and domain specifications are covered in Section 2. Details of the cyclone case studies are discussed in Section 3. The results and discussion of the current study are described in Section 4, and finally the conclusions listed in Section 5.

Section snippets

Model setup and methodology

The WRF model used in the present study is a mesoscale numerical model which is capable of simulating intense tropical cyclones (TCs) with realistic structures (Gentry, 2007). The atmospheric domain is simulated using the ideal gas law along with the conservation of thermal energy, horizontal momentum, and conservation of mass (more details are provided in Table 1). In context to maximum intensity and circulation features, the selection of optimum grid resolution plays an important role in WRF

Case studies for severe cyclones in the NIO basin

About 80 tropical cyclones form over the global ocean basins annually, and 5–6% of this total number form over the NIO basin (Niyas et al., 2009). The East coast of India that borders the Bay of Bengal is considered as the most vulnerable and susceptible regions in the world in context to risk associated with tropical cyclones and extreme wind-waves. Frequency of cyclones is higher in the Bay of Bengal as compared to the Arabian Sea. In a changing climate scenario and during recent years, the

Results and discussions

The induced or dynamic tertiary circulation can result in four types of circulation that includes eddies, turbulence, large-scale vertical waves, and Foehn winds. Eddies generated in the atmospheric column can be studied through observational measurements and numerical modeling. There are regions in the atmospheric column possessing enhanced eddy activity and the weather systems tend to evolve through baroclinic instability and subsequent dissipation. This study mainly emphasize on the

Summary and conclusions

Conducive synoptic scale environmental conditions favor the progression and evolution of pre-cyclonic vortices at different atmospheric levels. Thereafter with time the vortices subsides down leading to tropical cyclogenesis formation over the warm ocean surface detected by satellites. The present study is an effort to understand the spatio-temporal evolution of pre-cyclonic vortices in the atmospheric column that leads to tropical cyclogenesis. Several numerical experiments were performed to

Acknowledgement

The authors sincerely thank MHRD, Government of India for the financial support. This study was conducted for the DST Centre of Excellence (CoE) in Climate Change studies established at IIT Kharagpur as a part of the ongoing project ‘Wind-Waves and Extreme Water Level Climate Projections for East Coast of India’. The authors also acknowledge the working group on WRF model for carrying out this study.

References (52)

  • A. Okubo

    Horizontal dispersion of floatable particles in the vicinity of velocity singularities such as convergences

  • K.S. Singh et al.

    Impact of PBL and convection parameterization schemes for prediction of severe land-falling Bay of Bengal cyclones using WRF-ARW model

    J. Atmos. Sol. Terr. Phys.

    (2017)
  • G.J. Berry et al.

    African easterly waves during 2004 – analysis using objective techniques

    Mon. Weather Rev.

    (2007)
  • P.K. Bhaskaran et al.

    Tropical cyclogenesis for the north Indian Ocean region (book chapter), title: climate change and coastal and marine ecosystems in India

    Ministry of Environment, Forest & Climate Change, UNFCCC CoP-24, Katowice, Poland December

    (2018)
  • M. Bister et al.

    The genesis of Hurricane Guillermo: TEXMEX analysis and a modeling study

    Mon. Weather Rev.

    (1997)
  • S.J. Camargo et al.

    Use of a genesis potential index to diagnose ENSO effects on tropical cyclone genesis

    J. Clim.

    (2007)
  • J. Dudhia

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

    J. Atmos. Sci.

    (1989)
  • T.J. Dunkerton et al.

    Tropical cyclogenesis in tropical wave critical layer: easterly waves

    Atmos. Chem. Phys.

    (2009)
  • K.A. Emanuel

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

    J. Atmos. Sci.

    (1986)
  • A.T. Evan et al.

    A climatology of Arabian Sea cyclonic storms

    J. Clim.

    (2011)
  • L.M. Farfan et al.

    Orographic influence in the synoptic-scale circulations associated with the genesis of Hurricane Guillermo

    Mon. Weather Rev.

    (1997)
  • A.O. Fierro et al.

    The impact of horizontal grid spacing on the microphysical and kinematic structures of strong tropical cyclones simulated with the WRF-ARW model

    Mon. Weather Rev.

    (2009)
  • M.S. Gentry

    Sensitivity of WRF Simulations of Hurricane Ivan to Horizontal Resolution

    (2007)
  • M.S. Gentry et al.

    The sensitivity of WRF simulations of Hurricane Ivan to choice of cumulus parameterization

  • E.A. Hendricks et al.

    The role of vertical hot towers in the formation of tropical cyclone Diana (1984)

    J. Atmos. Sci.

    (2004)
  • K. Hodges et al.

    A comparison of recent reanalysis datasets using objective feature tracking: storm tracks and tropical easterly waves

    Mon. Weather Rev.

    (2012)
  • S.Y. Hong et al.

    A new vertical diffusion package with an explicit treatment of entrainment processes

    Mon. Weather Rev.

    (2006)
  • V.M. Karyampudi et al.

    Synoptic-scale influence of the Saharan air layer on tropical cyclogenesis over the eastern Atlantic

    Mon. Weather Rev.

    (2002)
  • B. Kerns et al.

    Four years of tropical ERA-40 vorticity maxima tracks. Part II: differences between developing and non-developing disturbances

    Mon. Weather Rev.

    (2009)
  • K. Kikuchi et al.

    Formation of tropical cyclones in the northern Indian Ocean associated with two types of tropical intra-seasonal oscillation modes

    J. Meteor. Soc. Japan

    (2010)
  • X. Li et al.

    Sensitivity of numerical simulations of the early rapid intensification of Hurricane Emily to cumulus parameterization schemes in different model horizontal resolutions

    Journal of the Meteorological Society of Japan. Ser. II

    (2009)
  • Y.L. Lin et al.

    Bulk parameterization of the snow field in a cloud model

    J. Clim. Appl. Meteorol.

    (1983)
  • J.L. McBride et al.

    Observational analysis of tropical cyclone formation. Part II: comparison of nondeveloping versus developing systems

    J. Atmos. Sci.

    (1981)
  • E.J. Mlawer et al.

    Radiative transfer for inhomogeneous atmosphere: RRTM, a validated correlated-k model for the longwave

    J. Geophys. Res. Atmos.

    (1997)
  • M.T. Montgomery et al.

    Tropical cyclogenesis via convectively forced vortex Rossby waves in a three-dimensional quasi-geostrophic model

    J. Atmos. Sci.

    (1998)
  • M.T. Montgomery et al.

    A vertical hot tower route to tropical cyclogenesis

    J. Atmos. Sci.

    (2006)
  • Cited by (0)

    View full text