1. Introduction
 Tropical Cyclones are very intense   characterized by a low-pressure center, a closed low-level atmospheric circulation, strong winds, and a spiral arrangement of thunderstorms that produce heavy rain.important at the coast as they can lead to high waves, storm surges, inland inundation, and erosion in populated areas resulting in severe economical and societal damage. Prediction of these events is a challenging task, around the world various models were used to predict these events. All these models were very sensitive to forcing fields like wind speed, track and mean sea level pressure anomalies. A wide variety of tropical cyclone surface wind fields have been used to drive storm surge prediction models, ranging from parametric wind models (Holland 1980, 2010), to steady-state dynamic PBL models, to interactive objective kinematic analysis (Cox et al., 1995), and to sophisticated non-hydrostatic Numerical Weather Prediction (NWP) models. Though parametric wind models have distinct operational advantages that maximize the number of hours of forecast utility, they cannot reproduce a fine scale features such as spiral rain bands. Difficulties in storm modeling  be related to the smoothness of model simulated/predicted wind fields because of the limitations of spatial resolution not capturing secondary depression, fronts, and the sting jet phenomenon (Browning 2004). Improved atmospheric model resolution has been found to improve the representation of storms, producing sharper Mesoscale features as well as better representing convective events (Mass et al., 2002). In India, several earlier works on storm surge modeling based on above methods concluded that resolving the atmospheric forcing of the cyclone is very important along with detailed coastal geometry for surge development (\cite{Sig_enza_Ortiz_2017}, \cite{1977}). Hence, the motivation of this work is to examine the application of numerical weather prediction model (WRF-ARW) to compute the realistic maximum wind speed, lower sea level pressure and reduced track error values by improving the track simulation at high domain resolution towards application to storm surge simulations. This study mainly focused to explore the impact of atmospheric model resolution and initial conditions in simulating the cyclone track, maximum winds and intensity. In these regions several studies have been carried-out  for prediction of track and intensity of tropical cyclones using numerical weather prediction (NWP) models (Pattanayak and Mohanty 2008). But those studies are confined to coarse resolution grids and are not aimed towards applications in storm surge prediction. Bhaskaran et al., (2013) have used WRF simulated winds and pressure fields as inputs to ocean model to study the water level rise at the coast based on ensemble tracks. Their study was confined to coarse domain resolution at 9 km.
Thus, one of the primary focuses of the present work was to reconstruct the wind fields and their forcing under different horizontal resolutions using WRF model for predicting the cyclone. The results will be implied in future studies towards track prediction and intensity under various initial conditions and to be used in the storm surge studies.
1.1. Description of Hudhud cyclone
The Hudhud cyclone was the very severe cyclonic storm occurred over the Bay of Bengal during 07-14th October 2014. It developed over low-pressure area near Tenasserim coast and adjoining North Andaman Sea in the morning of 6th October 2014. It emerged into Southeast Bay of Bengal and continued to move west-northwestwards intensifying into a severe cyclonic storm in the morning of 09th October and into a very severe cyclonic storm in the afternoon of 10th October. It crossed north Andhra Pradesh coast over Visakhapatnam (VSP) between 1200 and 1300 hrs IST (0630 – 0730 UTC) of 12th October. IMD reported the maximum sustained wind speed of 180 kmph at the time of landfall. The lowest estimated central pressure was 950 hPa from 0000 UTC of 12th to 0600 UTC of 12th October. The maximum estimated wind speed was 185 kmph. The system moved in a west northwestward direction over the sea area and then northwestward at the time of landfall. The lowest observed pressure of 950.3 hPa was reported by Visakhapatnam station at 0700 UTC (at the time of landfall) of 12th October and the maximum sustained wind speed of 130 kmph at 0500 UTC. The damage was quite severe and the VSCS, Hudhud mainly affected North Andhra Pradesh and adjoining south Odisha. In Andhra Pradesh, 46 people lost their lives, and 43 were injured. About 41,269 houses were damaged. Thousands of animals including poultry/duck perished. 2.3 lakh Hectares of land was submerged (Medha and Sunitha, 2015). The IMD post cyclonic survey (IMD report 2014) reported maximum of storm surge of 1.4 meters above the astronomical tide at Visakhapatnam at the landfall.