atmosphere

Article Observing and Modeling the Response of Placentia Bay to an

Guangjun Xu 1,2,3,4,5,6 , Guoqi Han 3,*, Changming Dong 2,4,*, Jingsong Yang 5 and Brad DeYoung 6 1 Ocean Remote Sensing and Information Technology Laboratory, Guangdong Ocean University, Zhanjiang 524088, China; [email protected] 2 Southern Laboratory of Ocean Science and Engineering (Guangdong Zhuhai), Zhuhai 519000, China 3 Northwest Atlantic Fisheries Centre, Fisheries and Oceans Canada, St. John’s, NL A1C 5X1, Canada 4 Oceanic Modeling and Observation Laboratory, Nanjing University of Information Science and Technology, Nanjing 210044, China 5 State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, Ministry of Natural Resources, Hangzhou 310012, China; [email protected] 6 Department of Physics and Physical Oceanography, Memorial University of , St. John’s, NL A1B 3X7, Canada; [email protected] * Correspondence: [email protected] (G.H.); [email protected] (C.D.); Tel.: +1-709-772-4326 (G.H.); +86-25-5869-5733 (C.D.)  Received: 5 October 2019; Accepted: 11 November 2019; Published: 19 November 2019 

Abstract: An extratropical cyclone reported to have the largest wind speed in Newfoundland in more than a decade landed on the island of Newfoundland on 11 March 2017. The oceanic responses in Placentia Bay on the southeast coast of Newfoundland to the winter storm were examined using observed data and the Finite-Volume Community Ocean Model (FVCOM). The peak non-tidal water level increase, i.e., storm surge, reached 0.85 m at St. Lawrence and 0.77 m at Argentia on Placentia Bay. slightly decreased after the storm passage according to buoy and satellite measurements. Root mean square differences (RMSD) of the magnitude of storm surge between model results and observations are 0.15 m. The model sea surface temperature showed a small decrease, consistent with observations, with RMSDs from 0.19 to 0.64 ◦C at buoy stations. The simulated surface current changes agree with buoy observations, with model-observation velocity difference ratios (VDR) of 0.75–0.88. It was found that, at Argentia (St. Lawrence), the peak storm surge in Placentia Bay was dominantly (moderately) associated with the inverse barometric effect, and the subsequent negative surge was mainly due to the wind effect at both stations. The sea surface cooling was associated with oceanic heat loss. In the momentum balance, the Coriolis, pressure gradient, and advection terms were all important during the storm, while the first two terms were predominant before and after the storm.

Keywords: storm surge; FVCOM; extratropical cyclone

1. Introduction Tropical cyclones (such as hurricanes) and extratropical cyclones can generate strong winds and high coastal storm surges. The combination of storm surge and wind waves can produce significant inundation and cause severe damage in coastal zones [1,2]. Placentia Bay is located on the southeast coast of Newfoundland in the northwestern Atlantic Ocean, bordered by the Avalon Peninsula to the east and the Burin Peninsula to the west (see Figure1). It supports essential fisheries such as Atlantic cod [ 3] and emerging aquaculture. It is also an important marine transport route, especially for crude oil tankers. Therefore, it is important to

Atmosphere 2019, 10, 724; doi:10.3390/atmos10110724 www.mdpi.com/journal/atmosphere Atmosphere 2019, 10, 724 2 of 20 improve the understanding of oceanic responses to passing tropical and extratropical cyclones in Placentia Bay, for emergency response, fisheries management, aquaculture regulation and operation, Atmosphereand environment 2019, 10, x protection.FOR PEER REVIEW 2 of 20

Figure 1. (a) Map showing the study area. The black line is the track of the extratropical cyclone during Figure 1. (a) Map showing the study area. The black line is the track of the extratropical cyclone during 11~12 March 2017 and the background color shows the bathymetry of the region. The red box shows the March 11~12, 2017 and the background color shows the bathymetry of the region. The red box shows model domain. (b) Map of the model domain. Red squares are tide-gauge stations (SL, St. Lawrence; the model domain. (b) Map of the model domain. Red squares are tide-gauge stations (SL, St. AG, Argentia) and the blue triangles indicate locations of observation buoys (HP: Head of Placentia Lawrence; AG, Argentia) and the blue triangles indicate locations of observation buoys (HP: Head of Bay; RI: Red Island Shoal; MP: Mouth of Placenta Bay; NB: Nickerson Banks). Placentia Bay; RI: Red Island Shoal; MP: Mouth of Placenta Bay; NB: Nickerson Banks). Occasionally, Atlantic tropical storms pass over Placentia Bay in summer and fall. Using observational Occasionally, Atlantic tropical storms pass over Placentia Bay in summer and fall. Using data and modeling results, Han et al. [4] and Ma et al. [5] studied storm surges and other oceanic responses observational data and modeling results, Han et al. [4] and Ma et al. [5] studied storm surges and (temperature, currents) to on 21 September 2010 off Newfoundland. Han et al. [6] used other oceanic responses (temperature, currents) to Hurricane Igor on September 21, 2010 off satellite thermal imagery and ocean color data to examine the impacts of Hurricane Igor on the sea surface Newfoundland. Han et al. [6] used satellite thermal imagery and ocean color data to examine the temperature (SST) and phytoplankton over the Grand Banks of Newfoundland. Ma et al. [7] investigated impacts of Hurricane Igor on the sea surface temperature (SST) and phytoplankton over the Grand the oceanic responses of Placentia Bay to Hurricanes Igor on 21 September 2010 and Leslie on 11 September Banks of Newfoundland. Ma et al. [7] investigated the oceanic responses of Placentia Bay to 2012 using a coastal ocean model and in situ observations. In winter, the Newfoundland area is frequently Hurricanes Igor on September 21, 2010 and Leslie on September 11, 2012 using a coastal ocean model battered by extratropical cyclones. Extratropical cyclones are a dominant source of mid-latitude and in situ observations. weather, and their various formation and evolution mechanisms have been studied using a wide range In winter, the Newfoundland area is frequently battered by extratropical cyclones. Extratropical of observational, theoretical, and modeling approaches [8]. Extratropical cyclones are also the principal cyclones are a dominant source of mid-latitude weather, and their various formation and evolution cause of storm surge events [9]. Currently, many studies are focused on increasing the prediction mechanisms have been studied using a wide range of observational, theoretical, and modeling precision of extratropical cyclone tracks and assessing the links between the extratropical cyclones and approaches [8]. Extratropical cyclones are also the principal cause of storm surge events [9]. extreme events [10–13]. Using coupled air-sea model solutions, Nelson and He [14] found that air-sea Currently, many studies are focused on increasing the prediction precision of extratropical cyclone interactions during winter extratropical cyclone outbreaks enhanced ocean heat loss over the Gulf tracks and assessing the links between the extratropical cyclones and extreme events [10–13]. Using Stream and supported rapid extratropical cyclone intensification. Nevertheless, studies on oceanic coupled air-sea model solutions, Nelson and He [14] found that air-sea interactions during winter responses of Newfoundland waters to extratropical cyclones are limited. extratropical cyclone outbreaks enhanced ocean heat loss over the Gulf Stream and supported rapid According to online news [15], an exatratropical cyclone passed over Newfoundland on 11 March extratropical cyclone intensification. Nevertheless,1 studies on oceanic responses of Newfoundland 2017 (Figure1), with peak wind speeds up to 44 m s − at St. John’s International Airport. Meteorologists waters to extratropical cyclones are limited. described this storm as the strongest in more than a decade because the peak wind speed broke the According to online news [15], an exatratropical cyclone passed over Newfoundland on March record in some regions of Newfoundland. Three-quarters of Newfoundland was affected by the 11, 2017 (Figure 1), with peak wind speeds up to 44 m s−1 at St. John’s International Airport. extratropical cyclone, resulting in widespread power outage. Meteorologists described this storm as the strongest in more than a decade because the peak wind To understand the oceanic responses (such as storm surge, ocean temperature evolution, speed broke the record in some regions of Newfoundland. Three-quarters of Newfoundland was and current change) in Placentia Bay to the strong extratropical cyclone, we analyze in situ observations affected by the extratropical cyclone, resulting in widespread power outage. and remote sensing data, and apply a coastal ocean circulation model in Placentia Bay and adjacent To understand the oceanic responses (such as storm surge, ocean temperature evolution, and shelf waters. Furthermore, we aim to discuss the differences between the oceanic responses of Placentia current change) in Placentia Bay to the strong extratropical cyclone, we analyze in situ observations Bay to winter and summer storms, in reference to Ma et al. [7]. and remote sensing data, and apply a coastal ocean circulation model in Placentia Bay and adjacent shelf waters. Furthermore, we aim to discuss the differences between the oceanic responses of Placentia Bay to winter and summer storms, in reference to Ma et al. [7]. This manuscript is arranged as follows: In Section 2, we describe the data used and the model setup. Section 3 presents the oceanic responses to the extratropical cyclone using observations and model results. Section 4 discusses the storm surge mechanisms and the momentum balance during the storm. Finally, Section 5 gives the summary and conclusion.

Atmosphere 2019, 10, 724 3 of 20

This manuscript is arranged as follows: In Section2, we describe the data used and the model setup. Section3 presents the oceanic responses to the extratropical cyclone using observations and model results. Section4 discusses the storm surge mechanisms and the momentum balance during the storm. Finally, Section5 gives the summary and conclusion.

2. Data and Model Description

2.1. Data

2.1.1. Remote Sensing Data 20-Hz Jason-3 along-track altimetry data are produced by CNES (National Center for Space Studies, France) in the framework of the PEACHI project (Prototype for Expertise on AltiKa for Coastal, Hydrology and Ice [16]) and downloaded from the French Archiving, Validation, and Interpretation of Satellite Oceanographic data (AVISO, www.aviso.altimetry.fr). Each track repeats exactly every 9.9156 days. The 20-Hz data have an along-track spatial resolution of about 300 m. An ascending track (Pass 039) of Jason-3 passed over the domain during the extratropical cyclone event. We make corrections for Jason-3 altimetric range data in the Ku band, including ionospheric delay and sea state bias provided by CNES, radiometer wet tropospheric delay provided by the Jet Propulsion Laboratory (JPL), and dry tropospheric delay provided by the European Center for Medium-range Weather Forecasting (ECMWF). We also correct the altimetry data for geocentric ocean tide using Goddard Ocean Tide 4.8 (GOT4.8 [17]), solid earth tide [18], and pole tide [19]. Finally, we calculate the temporal sea surface height anomalies relative to the sea level averaged from 18 February 2016 to 18 July 2017. Multiscale Ultrahigh Resolution (MUR) Level 4 SST data [20,21] with spatial resolution of 0.01 0.01 are obtained from the Group for High Resolution Sea Surface Temperature (GHRSST, ◦ × ◦ https://www.ghrsst.org/). The data are produced at the JPL Physical Oceanography Data Distributed Active Archive Center (PO.DAAC), based on in situ data, the Advanced Very High Resolution Radiometer (AVHRR) data, and the Advanced Microwave Scanning Radiometer-EOS (AMSR-E) data.

2.1.2. In Situ Data Tide-gauge data are obtained from the Canadian Hydrographic Service (http://tides.gc.ca/). The data used in this study are hourly time series of water level at the tide-gauge stations St. Lawrence (SL, 46.92◦ N, 55.39◦ W) and Argentia (AG, 47.30◦ N, 53.98◦ W) (red squares in Figure1). Buoy data, including SST, currents at 0.5 m below sea surface, and 10 m wind speed, are downloaded from Environment Canada and SmartAtlantic (http://www.smartatlantic.ca/PlacentiaBay/). The data used are hourly observations at Head of Placentia Bay (HP, 47.76◦ N, 54.07◦ W), Red Island Shoal (RI, 47.32◦ N, 54.12◦ W), Mouth of Placentia Bay (MP, 46.98◦ N, 54.70◦ W), and Nickerson Bank (NB, 46.44◦ N, 53.39◦ W) (blue triangles in Figure1). Meteorological data are obtained from the Canadian Weather, Climate and Hazard Service (https: //www.canada.ca/en/services/environment/weather.html). Hourly wind speed and wind direction at SL were included in this study.

2.1.3. Numerical Model Data This study makes use of a high-resolution global reanalysis dataset, which include hourly 10 m wind, heat flux, mean sea level pressure and sea surface height from National Centers for Environmental Prediction (NCEP) Climate Forecast System Version2 (CFSv2, https://rda.ucar.edu/,[22]). The horizontal resolution for wind, heat flux, and mean sea level pressure is 0.2◦, while that for sea surface height is 0.5◦. Hourly temperature and salinity data are obtained from the HYbrid Coordinate Ocean Model (HYCOM [23]). Computations of HYCOM are carried out on a Mercator grid between 78◦ S and 76◦ N Atmosphere 2019, 10, 724 4 of 20

with horizontal resolution of 1/12◦ (about 7 km in our study area) and 32 vertical layers ranging from 0 to 5500 m. Atmosphere 2019, 10, x FOR PEER REVIEW 4 of 20 2.2. Model 2.2. Model 2.2.1. FVCOM (3.2) Ocean Circulation Model 2.2.1. FVCOM (3.2) Ocean Circulation Model With the combined advantages of horizontal grid flexibility and computational efficiency, the FVCOMWith the modelcombined [24] advantages is suitable forof horizontal simulations grid in areaflexibility with complexand computational coast lines. efficiency, FVCOM the has FVCOMbeen widely model used [24] in is simulating suitable for geophysical simulations dynamic in areaprocesses with complex in coastal coast waters lines. FVCOM [5,7,25,26 has]. For been the widelypresent used application, in simulating we run the geophysical model with dynamic the three-dimensional processes in coastal setup and waters in the [5,7,25,26]. prognostic For mode. the presentA barometric application, pressure we term run isthe added model in thewith momentum the three-dimensional equations to accountsetup and for in the the eff pectrognostic of atmospheric mode. Apressure barometric change pressure on sea level.term is The added k-ε turbulence in the momentum model from equations the General to account Ocean Turbulence for the effect Model of atmospheric(GOTM, http: pressure//www.gotm.net change/ ,[ on27 ]) sea is usedlevel. to The calculate k-ε turbulence the vertical model mixing. from the General Ocean Turbulence Model (GOTM, http://www.gotm.net/, [27]) is used to calculate the vertical mixing. 2.2.2. Model Domain and Grid 2.2.2. Model Domain and Grid The model domain includes Placentia Bay and adjacent shelf waters (see Figure1b). The geometry of theThe domain model includes domain several includes banks Placentia and channels. Bay and adjacent shelf waters (see Figure 1b). The geometryThe triangularof the domain grid includes (Figure2 several) is designed banks usingand channels. the Surface Water Model System (SMS, www. aquaveo.comThe triangular/software grid/sms-learning (Figure 2)). is The designed grid resolution using isthe lowest Surface along Water theopen Model boundary System over (SMS, the www.aquaveo.com/software/smscontinental shelf (3–5 km) and highest-learning). along The the grid coastlines resolution in the is inner lowest bay along (50m). the The open domain boundary grid overconsists the continental of 31 unequally shelf spaced (3–5 km) sigma and levelshighest in along the vertical, the coastlines as well in as the 25,414 inner nodes bay and(50 m). 47,629 The elementsdomain gridin the consists horizontal. of 31 Theunequally model spaced bathymetry sigma is levels derived in the from vertical, General as Bathymetricwell as 25,414 Chart nodes of theand Oceans47,629 elements(GEBCO, inhttp: the// horizontal.www.gebco.net The /model), which bathymetry is in a one is arc-minute derived from grid andGeneral updated Bathymetric in 2008. Chart To minimize of the Oceanspressure (GEBCO, gradient http://www.gebco.net/), errors [28], the bathymetry which is smoothedis in a one using arc-minute the same grid method and updated as in Ma in et2008. al. [29To]. minimizeThe integration pressure time gradient step is errors set as [28], 0.6 s the for bathymetry the external is mode smoothed and 6 using s for thethe internalsame method mode. as in Ma et al. [29]. The integration time step is set as 0.6 s for the external mode and 6 s for the internal mode.

FigureFigure 2. 2. A Amodel model mesh mesh with with 25,414 25,414 nodes nodes and and47,629 47,629 elements elements for Placentia for Placentia Bay and Bay adjacent and adjacent shelf watersshelf waters..

2.2.3. Model Forcing, Open Boundary Conditions and Initial Conditions The model is forced by spatially and temporally varying winds, air pressure and heat fluxes at the sea surface. Tidal and non-tidal sea levels, temperature and salinity are specified at the open boundary. Hourly, 0.2° wind speeds at 10 m above sea surface and barometric pressure field at the sea surface obtained from CFSv2 are interpolated onto the elements and nodes, respectively. Eight dominant tidal constituents (semi-diurnal: M2, S2, N2 and K2; and diurnal: K1, O1, P1 and Q1) are obtained from OTIS (OSU Tidal Inversion Software [30]) and specified along the open boundaries.

Atmosphere 2019, 10, 724 5 of 20

2.2.3. Model Forcing, Open Boundary Conditions and Initial Conditions The model is forced by spatially and temporally varying winds, air pressure and heat fluxes at the sea surface. Tidal and non-tidal sea levels, temperature and salinity are specified at the open boundary. Hourly, 0.2◦ wind speeds at 10 m above sea surface and barometric pressure field at the sea surface obtained from CFSv2 are interpolated onto the elements and nodes, respectively. Eight dominant tidal constituents (semi-diurnal: M2, S2, N2 and K2; and diurnal: K1, O1, P1 and Q1) are obtained from OTIS (OSU Tidal Inversion Software [30]) and specified along the open boundaries. Non-tidal sea level elevations along the open boundaries are extracted from CFSv2. A difference between non-tidal sea level elevations from CFSv2 and tide-gauge data at stations AG and SL is noted while processing the data. The difference of 0.58 cm averaged for the two stations and over the model period is added to every open boundary node. River runoff is not considered since it is very low in winter. Initial sea level and velocity of the model are set to zero. The temperature and salinity conditions are initialized from the HYCOM monthly mean output of February 2017. The model is run from 14 February to 31 March in 2017. The model reaches an approximate dynamic equilibrium after running for 15 days. Hourly results from March 1 to March 31 are analyzed to examine the influence of the extratropical cyclone on sea level, temperature, salinity, and currents in Placentia Bay.

2.2.4. Model Validation Metrics The model results are compared with observational data and evaluated qualitatively and quantitatively by examining their correlation coefficient (cor) and the root mean square differences (RMSD) [7,31]: q 1 X 2 RMSD = (Am Ao) , (1) N − where Am and Ao are the model output and observational data, respectively. In addition, the velocity difference ratio (VDR) and the speed difference ratio (SDR) are included to validate the currents. VDR is defined as the ratio of the sum of the squared magnitudes of the vector velocity differences to the sum of the squared magnitudes of the observed velocities, that is: P V V 2 VDR = m o , (2) | P − 2 | Vo | | where Vm and Vo are the model velocities and the observational velocities, respectively. VDR = 0 means exact agreement. VDR < 1 means some agreement; the closer to 0 the better the agreement. VDR 1 means little or no agreement. Similarly, SDR is defined as the ratio of the sum of squared ≥ speed differences to the sum of the squared magnitudes of the observations, that is:

P 2 ( Vm Vo ) SDR = | | − | | . (3) P 2 Vo | | VDR is usually larger than SDR, since the former considers not only speed but also direction.

3. Results

3.1. Extratropical Storm Detection Extratropical storms are identified from CFSv2, using an automated storm detection and tracking methodology [32,33]. Storms are detected according to the minima of air pressure at sea level. We consider a storm for further study based on the following criteria: (i) there is a closed low pressure contour with central air pressure less than 1005 hPa, which is generated in the extratropical region from 25◦ N to 55◦ N; (ii) the low-pressure system lasts for at least 24 h; and (iii) the storm track is from southwest to northeast. Atmosphere 2019, 10, 724 6 of 20

As shown in Figure3, a low-pressure system formed over the northwest Atlantic on 10 March 2017, moving northeastward and gradually intensifying. In the morning of 11 March 2017, the extratropical cyclone arrived on the southwestern Newfoundland Shelf and headed north toward Placentia Bay. It made landfall at about 13:30 (UTC, similarly hereinafter) and crossed over Newfoundland in the afternoon. By the evening of 11 March, the storm center was located on the northeastern Newfoundland AtmosphereShelf. It stayed2019, 10, inx FOR the PEER area REVIEW for about two days, gradually weakening and eventually dissipating.6 of 20

Figure 3. Wind fields (vectors) and sea level pressure (shadows) from CFSv2. Figure 3. Wind fields (vectors) and sea level pressure (shadows) from CFSv2. 3.2. Wind 3.2. Wind The predominant wind over the study domain during the winter is eastward [34]. When the extratropicalThe predominant cyclone approachedwind over the Newfoundland, study domain theduring dominant the winter wind is waseastward southwestward [34]. When overthe extratropicalPlacentia Bay cyclone (Figure 4 approached). As the storm Newfoundland, center moved the over dominant Placentia wind Bay, the was observed southwestward wind shifted over Placentianorthwestward. Bay (Figure The wind4). As direction the storm finally center turned moved northeastward over Placentia after Bay, the the storm observed center wind left shifted the bay. 1 northwestward.The maximum hourly The wind wind direction speed was finally recorded turned at northeastward 26.6 m s− over after Placentia the storm Bay during center theleft periodthe bay. of The10–16 maximum March 2017. hourly Figure wind5 showsspeed was the comparisonrecorded at between26.6 m s−1 CFSv2 over Placentia and observed Bay during 10-m windsthe period at four of Marchlocations 10– in16, the 2017. model Figure domain. 5 shows the comparison between CFSv2 and observed 10-m winds at four locations in the model domain.

Atmosphere 2019, 10, x FOR PEER REVIEW 7 of 20

Atmosphere 2019, 10, x 724 FOR PEER REVIEW 77 of of 20 20

Figure 4. Hourly wind vectors (m s−11 ) from buoys at HP (a), RI (b), and MP (c) from March 10 to 13, Figure 4. Hourly wind vectors (m s− ) from buoys at HP (a), RI (b), and MP (c) from 10 to 13 March 2017. 2017. Figure 4. Hourly wind vectors (m s−1) from buoys at HP (a), RI (b), and MP (c) from March 10 to 13, 2017.

FigureFigure 5. 5. ComparisonComparison between between buoy buoy (red) (red) and and CFSv2 CFSv2 (blue) (blue) zonal zonal winds winds ( (leftleft panels panels)) and and meridional meridional 1 windswinds ( (rightright panels panels)) (m (m s s−1−) )at at HP, HP, RI, RI, MP MP,, and and NB. NB. The The horizontal horizontal black black dashed dashed lines lines denote U = 00 m m Figure1 5. Comparison1 between buoy (red) and CFSv2 (blue) zonal winds (left panels) and meridional ss−1− andand V V = =0 0m m s−1 s.− The. The vertical vertical black black dashed dashed lines lines denote denote the the time time of oflandfall. landfall. winds (right panels) (m s−1) at HP, RI, MP, and NB. The horizontal black dashed lines denote U = 0 m s−1 and V = 0 m s−1. The vertical black dashed lines denote the time of landfall.

Atmosphere 2019, 10, 724 8 of 20 Atmosphere 2019, 10, x FOR PEER REVIEW 8 of 20

3.3.3.3. Storm Storm Surge Surge WhenWhen the the stor stormm reaches reaches the the bay bay in in the the morning morning of of March 11 March 11, 2017, 2017, the the water water level level increases. increases. PeakPeak surges of 0.85 mm andand 0.770.77 m m are are reached reached at at SL SL (Figure (Figure6a) 6 anda) and AG AG (Figure (Figure6b), 6 respectively,b), respectively, a few a fewhours hours after after the stormthe storm makes makes landfall. landfall. Afterwards Afterwards a sharp a sharp decrease decrease in sea in sea level level is seen is seen at SL at andSL and AG. AG.Negative Negative surges surges with with magnitudes magnitudes similar similar to the to preceding the preceding positive positive surges surges occur occur in the in evening. the evening. There Thereis indication is indication of high-frequency of high-frequency oscillations, oscillations, which which may be may associated be associated with seiches with seiches [7]. [7].

Figure 6. Time series of hourly non-tidal sea level anomalies (m) from tide-gauge observations (red) Figureand model 6. Time results series (blue) of hourly at AG non (a) and-tidal SL sea (b). level The anomalies horizontal (m) black from dashed tide- linesgauge denote observations non-tide (red) SLA and= 0 m.model The results vertical (blue) black at dashed AG (a) lines and denoteSL (b). The the timehorizontal of landfall. black dashed lines denote non-tide SLA = 0 m. The vertical black dashed lines denote the time of landfall. We compare the non-tidal sea level anomalies (defined as non-tidal sea levels minus their mean overWe the compare period from the 1non to 31-tida Marchl sea 2017)level anomalies from the model (defined solutions as non with-tidal the sea tide-gauge levels minus observations their mean at overAG andthe SL.period Tide-gauge from March data 1 are to first31, 2017) detided from by the using model the solutions T-Tide toolbox with the [35]. tide Eight-gauge tidal observations constituents at(M2, AG S2, and N2, SL. K2, Tide K1,-gauge O1, P1, data and are Q1) first are removed detided by from using the model the T-Tide results toolbox by using [35]. the Ei harmonicght tidal constituentsanalysis. The (M2, temporal S2, N2, evolutions K2, K1, O1, of hourly P1, and non-tidal Q1) are sea removed level anomalies from the frommodel 10 results to 16 March by using 2017 the are harmonicshown in Figureanalysis.6. CorrelationThe temporal coe evolutionsfficients of of the hourly model non solutions-tidal sea of non-tidal level anomalies sea level from anomalies March and 10 toobservations 16, 2017 areat shown AG and in Figure SL tide-gauge 6. Correlation stations coef areficients 0.74 and of the 0.81, model respectively solutions (Figure of non6-).tidal The sea peak level of anomaliessea level anomalies and observations at AG is at observed AG and atSL 15:00 tide- ongauge 11 March, stations while are 0.74 the and model 0.81, peak respectively occurs at (Figure 14:00 (a 6phase). The discrepancypeak of sea oflevel one anomalies hour), with at aAG magnitude is observed about at 15:00 0.1 m on lower March than 11, the while peak the of model 0.77 m peak from occuobservations.rs at 14:00 At (a SL,phase water discrepancy levels reach of one 0.85 hour), m at 14:00 with on a magnitude 11 March, but about the 0.1 peak m oflower about than 0.47 the m peak from ofthe 0.77 model m from solutions observations. occurs four At hours SL, water earlier. levels After reach peaking, 0.85 seam at level 14:00 starts on March to decrease 11, but rapidly the peak at both of aboutstations. 0.47 The m averagedfrom the model-observationmodel solutions occurs RMSD four for hours water levelearlier. is 0.15After m peaking, from 10 Marchsea level to 16starts March, to decreasesubstantially rapidly smaller at both than stations. the peak The surge averaged values. model-observation RMSD for water level is 0.15 m from TheMarch Jason-3 10 to March altimeter 16, passessubstantially over Placentia smaller than Bay the (Figure peak7 a)surge during values. the extratropical cyclone. We calculateThe Jason along-track-3 altimeter seapasses surface over heightPlacenti anomaliesa Bay (Figure based 7a) on during the altimetry the extratropical product cyclone. (Figure 7Web). calculateThe storm along surge-track is about sea surface 0.5 m near height RI (Figureanomalies7b) atbased 15:07 on on the 11 altimetry March, consistent product (Figure with the 7b). model The stormresult surge but much is about lower 0.5 than m near the tide-gaugeRI (Figure 7 valueb) at 15:07 of 0.75 on m March at nearby 11, consistent AG. The model with the underestimation model result butis mainly much duelower to than the one-hour the tide-gauge phase value error asof indicated0.75 m at innearby the comparison AG. The model with theunderestimation tide-gauge data is mainly(Figure 6duea). Whileto the theone- causeshour phase for the error phase as errorindicated at this in magnitudethe comparison is complicated with the tide and-gauge di fficult data to (Figureidentify, 6 likelya). While major the factors causes are for wind the phase forcing error and non-tidalat this magnitude sea levels is specified complicated at the and open difficult boundary. to identify,The Jason-3 likely underestimation major factors are may wind be forcing due to and errors non in-tidal altimetric sea levels measurements specified at the and open environmental boundary. The Jason-3 underestimation may be due to errors in altimetric measurements and environmental and geophysical corrections [36]. The sea surface height anomalies decrease offshore (Figure 7b). The

Atmosphere 2019, 10, 724 9 of 20

andAtmosphere geophysical 2019, 10, correctionsx FOR PEER REVIEW [36]. The sea surface height anomalies decrease offshore (Figure79 b).of 20 The RMSD of sea surface height anomalies between model and low-pass filtered altimeter observations isRMSD 0.12 m. of sea surface height anomalies between model and low-pass filtered altimeter observations is 0.12 m.

FigureFigure 7. 7.(a ()a Ground) Ground track track of Passof Pass 039 039 from from Jason-3 Jason satellite-3 satellite (blue (blue line). line). The black The black line denotes line denotes the track the oftrack the extratropical of the extratropical cyclone. cyclone. The red The squares red squares represent represent the tide-gauge the tide stations-gauge atstations SL and at AG. SL (andb) Jason-3 AG. (b) along-trackJason-3 along sea-track level sea anomalies level anomalies with model withresults model alongresults Pass along 039 Pass on 039 11 Marchon March 2017. 11, The2017. blue The line blue denotesline denotes along-track along- seatrack level sea anomalies,level anomalies, the yellow the yellow line denotes line denotes low-pass low filtered-pass filtered along-track along sea-track level sea anomalies,level anomalies, the red the line red denotes line denotes FVCOM FVCOM model resultsmodel results and the and black the triangle black triangle denotes denotes from tide-gauge from tide- observationsgauge observations at AG. at AG. 3.4. Temperature 3.4. Temperature Comparisons of time series of SST based on buoy measurements and model solutions at HP, RI, Comparisons of time series of SST based on buoy measurements and model solutions at HP, RI, MP and NB are shown in Figure8. SST from buoy observations and model results both decrease slightly, MP and NB are shown in Figure 8. SST from buoy observations and model results both decrease with correlation coefficients from 0.67 to 0.81 at the four buoys. The averaged SST at the four buoy slightly, with correlation coefficients from 0.67 to 0.81 at the four buoys. The averaged SST at the four stations from model results drop by 0.45 C from 17:00 on 10 March to 11:00 on 13 March, while the buoy stations from model results drop by◦ 0.45 °C from 17:00 on March 10 to 11:00 on March 13, while averaged SST from buoy observations decrease by 0.33 C during the same period. The RMSDs the averaged SST from buoy observations decrease by 0.33◦ °C during the same period. The RMSDs are mostly less than 0.5 C at the buoy stations. Therefore, there is an overall good agreement are mostly less than 0.5 °C◦ at the buoy stations. Therefore, there is an overall good agreement between between the observations and the model. Previous studies off the Canadian Atlantic coast reported the observations and the model. Previous studies off the Canadian Atlantic coast reported model- model-observation SST differences of 1 C during Hurricane Juan [37] and 1 to 1.5 C during Hurricanes observation SST differences of 1 °C during◦ Hurricane Juan [37] and 1 to 1.5 °C during◦ Hurricanes Igor Igor and Leslie [7]. and Leslie [7].

Atmosphere 2019, 10, 724 10 of 20 Atmosphere 2019, 10, x FOR PEER REVIEW 10 of 20

Figure 8. Comparison between buoys (red) and modeled (blue) sea surface SST at HP (a), RI (b), MP (c), Figure 8. Comparison between buoys (red) and modeled (blue) sea surface SST at HP (a), RI (b), MP and NB (d). The horizontal black dashed lines denote SST = 0 ◦C. The vertical black dashed lines (denotec), and theNB time(d). The of landfall. horizontal black dashed lines denote SST = 0 °C. The vertical black dashed lines denote the time of landfall. Next, we examine satellite observations. A slight drop of SST appears in satellite data in Placentia Bay andNext, surrounding we examine waters satellite after observations. the passage of A the slight extratropical drop of cyclone SST appears (Figure in9 a,b). satellite The biggestdata in Placentia Bay and surrounding waters after the passage of the extratropical cyclone (Figure 9a,b). The drop in the spatially averaged SST for Placentia Bay is 0.89 ◦C from 10 March to 14 March (Figure9f). biggestThe discrepancies drop in the in spatially the SST changesaveraged between SST for satellite Placentia and Bay buoy is 0.89 data °C may from be ascribedMarch 10 to to the March different 14 (Figuremeasurement 9f). The depths discrepancies used: the in seathe SST surface changes for the between former satellite and 0.5 and m below buoy data it for may the latter.be ascribed The sea to thesurface different cooling measurement from satellite depths observations used: the (Figure sea surface9a,b) isfor more the former prominent and in0.5 the m innerbelow and it for middle the latter. bay Thethan sea in thesurface outer cooling bay, since from the satellite outer bayobservations can be strongly (Figure influenced 9a,b) is more by relativelyprominent warm in the shelf inner water and middlethrough bay advection than in and the lateralouter bay, mixing. since The the distribution outer bay can of SST be strongly from the influenced FVCOM solutions by relatively (Figure warm9c,d) shelfpresents water a similarthrough pattern advection overall. and lateral Nevertheless, mixing. thereThe distribution are some notable of SST spatialfrom the di FVCOMfferences solu betweentions (Figureobserved 9c and,d) model presents SST a changes, similar patternfor which overall. we are Nevertheless,unable to pinpoint there exact are causes. some notable The maximum spatial differences between observed and model SST changes, for which we are unable to pinpoint exact SST drop from 10 March to 13 March is 0.54 ◦C (Figure9f). causes. The maximum SST drop from March 10 to March 13 is 0.54 °C (Figure 9f). The SST averaged for the three buoy stations drops by 0.42 ◦C from 10 March to 13 March, which represents the maximal SST change during the storm (Figure9f). Figure9e shows the time series of daily air temperatures observed from buoys at RI and NB and from weather station at SL. The air temperature is about 0 ◦C before the storm, and drops about 7 ◦C during the storm (from 9 March to 12 March).

AtmosphereAtmosphere2019 2019, 10, 10, 724, x FOR PEER REVIEW 1111 of of 20 20

Figure 9. (a,b) Five-day averaged sea surface temperatures observed from satellites before (06–10 Figure 9. (a,b) Five-day averaged sea surface temperatures observed from satellites before (March 06– March) and after (11–16 March) the passage of the extratropical cyclone. The red squares denote the 10) and after (March 11–16) the passage of the extratropical cyclone. The red squares denote the region region in which the spatially averaged sea surface temperatures are calculated. (c,d) same as (a,b) in which the spatially averaged sea surface temperatures are calculated. (c,d) same as (a,b) except for except for the sea surface temperatures from FVCOM solutions. (e) Time series of daily averaged air the sea surface temperatures from FVCOM solutions. (e) Time series of daily averaged air temperature from buoys at RI and NB and from the weather station at AG. (f) Time series of daily temperature from buoys at RI and NB and from the weather station at AG. (f) Time series of daily spatially averaged sea surface temperature from satellite observations (red), buoy measurements (blue) spatially averaged sea surface temperature from satellite observations (red), buoy measurements and FVCOM solutions (green) with standard deviations shown as the shadows. (blue) and FVCOM solutions (green) with standard deviations shown as the shadows. 3.5. Sea Surface Current The SST averaged for the three buoy stations drops by 0.42 °C from March 10 to March 13, which Modeled surface currents are compared with measurements taken at buoys HP, RI, and MP represents the maximal SST change during the storm (Figure 9f). Figure 9e shows the time series of (Figure 10). The buoy measurements show that the storm-induced surface current changes are mainly daily air temperatures observed from buoys at RI and NB and from weather station at SL. The air associated with the zonal component, which is consistent with the dominant wind changes in the zonal temperature is about 0 °C before the storm, and drops about 7 °C during the storm (from March 9 to direction (Figure5). As the storm approaching, surface currents at HP (Figure 10a,b) change gradually March 12). from eastward to westward and then drastically from westward back to eastward soon after landfall. The3.5. surface Sea Surface currents Current at RI are relatively weak even during the extratropical cyclone (Figure 10c,d) due to its location. At MP, the surface currents turn westward before the extratropical cyclone moves over Modeled surface currents are compared with measurements taken at buoys HP, RI, and MP (Figure 10). The buoy measurements show that the storm-induced surface current changes are mainly

Atmosphere 2019, 10, x FOR PEER REVIEW 12 of 20 associated with the zonal component, which is consistent with the dominant wind changes in the zonal direction (Figure 5). As the storm approaching, surface currents at HP (Figure 10a,b) change Atmosphere 2019, 10, 724 12 of 20 gradually from eastward to westward and then drastically from westward back to eastward soon after landfall. The surface currents at RI are relatively weak even during the extratropical cyclone (Figurethe region 10c (Figure,d) due 10to e,f).its location. After the At storm MP, the center surface passes, currents the surface turn westward currents atbefore MP become the extratropical eastward cycloneunder the moves wind over forcing. the region (Figure 10e,f). After the storm center passes, the surface currents at MP become eastward under the wind forcing.

Figure 1 10.0. TimeTime series of buoy-measuredbuoy-measured seasea surfacesurface currentscurrents (red) (red) compared compared to to model model results results (blue) (blue) at 1 1 atHP HP (a ,(ba,), b RI), (RIc,d ()c, and d) and MP MP (e,f ).(e, The f). The horizontal horizontal black black dashed dashed lines lines denote denote U = 0U m = s0− mand s−1 and V = V0 m= 0 s −m . sThe−1. The vertical vertical black black dashed dashed lines lines denote denote the the time time of landfall. of landfall.

Statistics calculated calculated between observed observed and and model model currents currents show show that that the the correlation correlation coefficients coefficients are greater greater than than 0.5 0.5 for for the zonalthe zonal component, component, which which undergoes undergoes much larger much changes larger than changes the meridional than the meridionalcomponent component during the storm.during The the SDRstorm. values The ofSDR the values currents of duringthe cur therents storm during event the (from storm 00:00 event on (from11 March 00:00 to on 12:00 March on 1211 March)to 12:00 are on smallMarch varying 12) are fromsmall0.23 varying to 0.34 from at the0.23 three to 0.34 buoys, at the indicating three buoys, the indicatingmagnitudes the of magnitudes the model results of the agreemodel with results the observations.agree with the On observations. the other hand, On the the other VDR valueshand, the are VDRlarger, values varying are from larger, 0.75 var toying 0.88, from indicating 0.75 to fair 0.88, agreement indicating between fair agreement model solutions between and model observed solutions data. and observedThere is nodata. indication of near-inertial oscillation in neither observed nor modeled currents, since the oceanThere had is no little indication stratification of near in-inertial the vertical. oscillation Previous in neither studies observed [7] showed nor strong modeled inertial currents, oscillation since theafter ocean the passagehad little of stratification hurricanes atin MPthe vertical. in summer Previous when s thetudies ocean [7] wasshowed well strong stratified inertial in the oscillation vertical. afterVertical the stratificationpassage of hurricanes favors the at generation MP in summer of the near-inertialwhen the ocean oscillation was well during stratified a storm in the [7]. vertical. Vertical stratification favors the generation of the near-inertial oscillation during a storm [7]. 4. Discussion

4.4.1. Discussi Storm Surgeon Mechanisms

4.1. StormThere Surge are severalMechanisms processes generating storm surge, including winds, , tide-surge interaction [38]. Ma et al. [7] investigated the storm surges in Placentia Bay during There are several processes generating storm surge, including winds, atmospheric pressure, Hurricanes Igor on 21 September 2010 and Leslie on 11 September 2012. They found the peak storm tide-surge interaction [38]. Ma et al. [7] investigated the storm surges in Placentia Bay during surge is significantly influenced by local wind-driven and inverse barometric effects during Leslie at Hurricanes Igor on September 21, 2010 and Leslie on September 11, 2012. They found the peak storm Argentia and St. Lawrence, but predominately due to remote forcing during Igor, depending on the surge is significantly influenced by local wind-driven and inverse barometric effects during Leslie at storm track location. Argentia and St. Lawrence, but predominately due to remote forcing during Igor, depending on the To identify the dominant processes for the storm surge during this winter storm, we carried out a storm track location. series of model experiments with three different forcing configurations, namely, atmospheric forcing To identify the dominant processes for the storm surge during this winter storm, we carried out only, tide forcing only, and atmospheric and tide forcing combined. The corresponding non-tidal sea a series of model experiments with three different forcing configurations, namely, atmospheric levels are denoted as ηNT, ηTO, and ηAT, respectively. Then the tide-surge interaction ηTSI for the storm

Atmosphere 2019, 10, 724 13 of 20 Atmosphere 2019, 10, x FOR PEER REVIEW 13 of 20 forcing only, tide forcing only, and atmospheric and tide forcing combined. The corresponding non- follows ηTSI = ηAT ηTO ηNT. The inverse barometric response is defined as the sea surface height tidal sea levels are denoted− − as 휂푁푇, 휂푇푂, and 휂퐴푇, respectively. Then the tide-surge interaction ηTSI variation caused by an atmospheric pressure fluctuation patm: for the storm follows 휂푇푆퐼 = 휂퐴푇 − 휂푇푂 − 휂푁푇. The inverse barometric response is defined as the sea surface height variation caused by an atmospheric pressurepatm fluctuation 푝 : η = ,푎푡푚 (4) − ρ푝0푎푡푚g 휂 = − , (4) 휌0푔 where η is the sea surface height variation, g is the gravitational acceleration, and ρ0 is the reference where 휂 is the sea surface height variation, 푔 is the gravitational acceleration, and 휌0 is the water density. The wind effect can then be determined by subtracting η from ηNT. 휂 휂 referenceThe barometricwater density. pressure The wind declines effect by 54can hPa then from be 00:00determined to 15:00 by on subtracting 11 March, when from the extratropical푁푇. cycloneThe movesbarometric toward pressure Placentia declines Bay. by The54 hPa lowest from barometric00:00 to 15:00 pressure on March is 11, 954 when hPa atthe AG, extratropical inducing cyclonea sea surface moves height toward increase Placentia of Bay. 0.58 The m that lowest accounts barometric for 75% pressure of the is observed 954 hPa at storm AG, inducing surge (0.77 a sea m) surface(Figure height11a). Atincrease SL, the of sea 0.58 surface m that height accou increasents for 75% caused of the by observed the inverse storm barometric surge (0.77 eff ectm) reached(Figure 11a).0.46 m,At aboutSL, the 56 sea percent surface of height the observed increase stormcaused surge by the (Figure inverse 11 barometricb). Therefore, effect the reached peak storm 0.46 m, surge about is 56mainly percent associated of the withobserved the inverse storm barometric surge (Figure response 11b). atTherefore, AG, and the to a peak lesser storm degree surge at SL. is mainly associated with the inverse barometric response at AG, and to a lesser degree at SL.

FigureFigure 11. TimeTime series series of of the the non non-tidal-tidal sea sea level level anomalies anomalies (m) (m) from from observations (red) (red) and and model resultsresults (blue), (blue), the the inverse inverse barometric barometric effect effect (purple), (purple), the the tide tide-surge-surge interaction interaction (green) (green),, and and the the wind wind effecteffect (black) at AG ( a) and SL ( b))..

ItIt is is well well known known that that Ekman Ekman transport transport induced induced by by winds winds plays plays an an important important role role in in gen generatingerating coastalcoastal storm storm surge. surge. In In this this extratropical extratropical cyclone cyclone event, event, the the wind wind effect effect is is much much weaker weaker than than the the inverseinverse barometric barometric e effectffect for for the the peak peak surge surge at AG at AG(Figure (Figure 11a), 11 duea)， todue quite to low quite atmospheric low atmospheric pressure pressureand relatively and relatively weak winds weak before winds and before during and landfall. during Stronglandfall. eastward Strong windeastward with wind speed with more speed than 1 more15 m sthan− (Figure 15 m s −112 (Figurea) lasts 12 fora) several lasts for days several after days landfall. after Thelandfall. eastward The eastward wind induces wind Ekman induces transport Ekman transport which pushes water out of the bay, causing the water level in the bay to drop by about 1.3

Atmosphere 2019, 10, x FOR PEER REVIEW 14 of 20

m from 14:00 to 21:00 on March 11. Therefore, the large negative surge is mainly due to the wind effect. Atmosphere Mecking2019, 10, 724 et al. [39] used a barotropic shallow-water model to simulate coastal sea level change14 of 20 around the southeastern Newfoundland in response to Tropical Storm Helene (2000) passing over the Grand Banks. It was found that a combination of wind stress and atmospheric pressure forcing whichcaused pushes a meteorologicalwater out of the tsunami. bay, causing The dynamic the water response level in to the the bay wind to drop forcing by about and atmospheric 1.3 m from 14:00 to 21:00pressure on 11 forcing March. resulted Therefore, in sea the level large increa negativese and decrease surge is mainlyat Argentia, due respectively. to the wind Note effect. that the dynamic pressure response here is different from the inverse barometric effect.

−11 FigureFigure 12. Hourly12. Hourly wind wind vectors vectors (m (m s −s )) fromfrom CFSv2 CFSv2 (blue) (blue) and and weather weather station station observations observations (red) (red)at at AG (AGa) and (a) and SL (SLb) ( fromb) from 10 March to 14 March10 to 14, 2017. 2017.

MeckingAt SL, et the al. observed [39] used wind a barotropic speed and shallow-water directions are quite model different to simulate from those coastal of seaCFSv2 level from change around12:00 the to southeastern 13:00 on March Newfoundland 11 (Figure 12b). in The response northwestward to Tropical winds Storm would Helene induce (2000) northeastward passing over the Ekman transport, causing the water level to increase. However, the southeastward winds from CFSv2 Grand Banks. It was found that a combination of wind stress and atmospheric pressure forcing caused would induce offshore Ekman transport, causing the water level to decrease in the model results. As a meteorologicala result, the model tsunami. would The underestimate dynamic response the contribution to the wind of forcingwinds to and the atmospheric peak storm surge pressure at SL. forcing resultedFollow ining sea Ma level et increaseal. [7], we and reconstruct decrease wind at Argentia, fields by blending respectively. windNote observations that the at dynamic the weather pressure responsestations here and is buoys different with from CFSv2 the and inverse use the barometric reconstructed effect. wind fields to force the model. As shown inAt Figure SL, the 13 observed, the model wind results speed forced and by directions the reconstructed are quite winds different are similar from thoseto those of forced CFSv2 by from the 12:00 to 13:00CFSv2 on winds, 11 March though (Figure the reconstructed 12b). The northwestward winds agree well with winds the would observed induce winds northeastward in both magnitude Ekman transport,and direction. causing Therefore, the water the level underestimation to increase. However, of the peak the storm southeastward surge may be winds related from to something CFSv2 would induceelse. off Whileshore we Ekman are not transport, able to pinpoint causing exact the waterproblems, level the to quality decrease of the in theopen model boundary results. condition As a result, the modelmay be would one of them, underestimate since SL is theclose contribution to the western of open winds boundary. to the peak storm surge at SL. Following Ma et al. [7], we reconstruct wind fields by blending wind observations at the weather stations and buoys with CFSv2 and use the reconstructed wind fields to force the model. As shown in Figure 13, the model results forced by the reconstructed winds are similar to those forced by the CFSv2 winds, though the reconstructed winds agree well with the observed winds in both magnitude and direction. Therefore, the underestimation of the peak storm surge may be related to something else. While we are not able to pinpoint exact problems, the quality of the open boundary condition may be one of them, since SL is close to the western open boundary. Atmosphere 2019, 10, 724 15 of 20 Atmosphere 2019, 10, x FOR PEER REVIEW 15 of 20

1 Figure 13 13.. ((aa)) Hourly Hourly wind wind vectors vectors (m s −1−) )from from weather weather station station observations observations (red) (red) and and reconstruction reconstruction (blue) at SL from March 10 to 16 10 March to 16, 2017.2017. ((b)) TimeTime seriesseries ofof thethe non-tidalnon-tidal seasea levellevel anomalies (m) from observations (red) (red) and and model model results results forced forced by the by reconstructed the reconstructed wind fields wind (blue) fields at (blue)SL. The at horizontal SL. The horizontalblack dashed black lines dashed denote lines non-tide denote SLAnon-tide= 0 SLA m. The= 0 m. vertical The vertical black dashedblack dashed lines denotelines denote the time the timeof landfall. of landfall. 4.2. Momentum Balance 4.2. Momentum Balance The momentum balance is analyzed at MP. The momentum equations and the continuity equation The momentum balance is analyzed at MP. The momentum equations and the continuity used in FVCOM [40] are: equation used in FVCOM [40] are: 휕푢 1 휕푝 휕푢 휕푢 휕푢 휕 휕푢 = 푓푣⏟ − − (푢! + 푣 + 푤! ) + (!퐾푚 ) +! ∂u 휕푡⏟ 1 ∂p ∂u휌⏟0 휕푥 ∂u ∂⏟u 휕푥 ∂ 휕푦 ∂ u 휕푧 ∂ 휕푧⏟∂u 휕푧∂ ∂u = + + + + + 푎푐푐푒푙푒푟푎푡𝑖표푛f v 퐶표푟𝑖표푙𝑖푠 u v w Km 푣푒푟푡𝑖푐푎푙 푑𝑖푓푓푢푠𝑖표푛 , (5) ∂t − ρ0 ∂x 푝푟푒− 푠푠푢푟푒∂x 푔푟푎푑𝑖푒푛푡∂y ∂z 푎푑푣푒푐푡𝑖표푛∂z ∂z ∂x A ∂x ∂y A∂y |{z} |{z} 휕 휕푢 휕 휕푢 (5) Coriolis |{z} | (풜{z ) + }(풜| ),{z } | {z } acceleration pressure gradient ⏟휕푥 advection 휕푥 휕푦 vertical 휕푦 di f f usion horizontal di f f usion ℎ표푟𝑖푧표푛푡푎푙 푑𝑖푓푓푢푠𝑖표푛 ! ! ! ! ∂v 1 ∂p ∂ v ∂v ∂ v ∂ ∂ v ∂ ∂ v ∂ ∂ v = 휕푣f u u 1 +휕푝v + w 휕푣+ 휕푣Km 휕푣+ 휕 + 휕푣 , (6) ∂t |{z}− − = ρ−⏟0푓푢∂y − − ∂x ∂y −∂(z푢 +∂푣z +∂z푤 ) ∂+x A∂x(퐾푚 ∂y)A∂+y |{z} 휕푡⏟ 휌⏟0 휕푦 ⏟ 휕푥 휕푦 휕푧 휕푧⏟ 휕푧 퐶표푟𝑖표푙𝑖푠|{z} | {z } | {z } | {z } 푎푐푐푒푙푒푟푎푡𝑖표푛Coriolis 푎푑푣푒푐푡𝑖표푛 푣푒푟푡𝑖푐푎푙 푑𝑖푓푓푢푠𝑖표푛 acceleration pressure gradient 푝푟푒푠푠푢푟푒 푔푟푎푑𝑖푒푛푡advection vertical di f f usion horizontal di f f usion 휕 휕푣 휕 휕푣 (6) (풜 ) + (풜 ), ⏟휕푥∂u ∂휕푥 v ∂ 휕푦w 휕푦 + + = 0, (7) ∂xℎ표푟𝑖푧표푛푡푎푙∂y 푑𝑖푓푓푢푠𝑖표푛∂z where x, y, and z are the zonal, meridional,휕푢 and휕푣 vertical휕푤 coordinates, respectively; u, v and w are + + = 0, (7) the x, y, and z velocity components, respectively;휕푥 휕푦 f is휕푧 the Coriolis parameter; g is the gravitational whereacceleration; x, y, andρ0 isz theare referencethe zonal, water meridional, density; andp is vertical the pressure; coordinates,Km is the respectively; vertical eddy u, viscosity;v and w andare theis x the, y, horizontaland z velocity eddy components, viscosity. respectively; f is the Coriolis parameter; g is the gravitational A acceleration;Based on ρ0 the is the model reference results, water four density; dominant p termsis the arepressure; considered: Km is Coriolis,the vertical pressure eddy viscosity; gradient, andadvection, 풜 is the and horizontal horizontal eddy diff usionviscosity. terms. The other terms are smaller by a few orders of magnitude. FigureBased 14 shows on the the model anomalies results, of thefour main dominant terms interms the x -are and considered:y-directions Coriolis, at 5 m below pressure the surfacegradient, at advectionMP. The anomalies, and horizontal are relative diffusion to the terms. means The over other the terms period are fromsmaller 8 March by a few to 14orders March. of magnitude. Before the Figureextratropical 14 shows cyclone, the anomalies the variations of the of main the Coriolisterms in termthe x and- and pressure y-directions gradient at 5 termm below were the large surface and atthus MP. played The anomalies dominant are roles relative in the to momentum the means balance over the in period both directions, from March while 8 to those March of 14. the Before advection the extratropical cyclone, the variations of the Coriolis term and pressure gradient term were large and thus played dominant roles in the momentum balance in both directions, while those of the advection

Atmosphere 2019, 10, 724 16 of 20 Atmosphere 2019, 10, x FOR PEER REVIEW 16 of 20 termterm and and horizontal horizontal diffusion diffusion term term were were relatively relatively small small.. As As the the storm storm was was approaching, approaching, magnitudes magnitudes of the pressure gradient term and the Coriolis term changed from less than 1 10 5 m s 2 to more of the pressure gradient term and the Coriolis term changed from less than 1 × 10×−5 m− s−2 to more− than than 2 10 5 m s 2. The variability of the advection term increased by about 1 10 5 m s 2 as well. 2 × 10−5× m s−2−. The −variability of the advection term increased by about 1 × 10−5 m ×s−2 as− well. −After the storm,After thethestorm, Coriolis the term Coriolis and the term pressure and the gradient pressure term gradient became term dominant becamedominant again. again.

Figure 14. Anomalies of Coriolis term (blue), advection term (red), pressure gradient term (green) and Figurehorizontal 14. Anomalies diffusion term of Coriolis (purple) term in the(blue), x-direction advection (a )term and (red), the y-direction pressure gradient (b). term (green) and horizontal diffusion term (purple) in the x-direction (a) and the y-direction (b). The root mean square (RMS) values of the Coriolis and pressure gradient terms are about four timesThe of root other mean terms square in the (RMS)x and valuesy directions of the Coriolis during and the pressure pre- and gradient post-storm terms periods are about (Table four1), timessuggesting of other that terms the momentum in the x and balance y directions is predominated during the by the pre Coriolis- and post and-storm pressure periods gradient (Table terms. 1), suggestingDuring the that storm the eventmomentum (from 12:00balanc one is 10 predominated March to 12:00 by on the 12 Coriolis March), and the pressure RMS values gradient of all terms. terms Duringincreased. the storm The Coriolis event (from and pressure 12:00 on gradientMarch 10 terms to 12:00 were on aboutMarch one 12), to the two RMS times values greater of all than terms the increased.other terms. The Coriolis and pressure gradient terms were about one to two times greater than the other terms. Table 1. Root mean square values of the terms in the momentum balance at MP. (units: 10−6 m s−2)

5 m

During Storm 1 Before and After Storm 2 x 8.74 5.64 Coriolis y 14.85 7.50 x 8.17 6.84 pressure gradient y 12.21 8.31 x 6.42 2.08 advection y 6.13 2.06

Atmosphere 2019, 10, 724 17 of 20

6 2 Table 1. Root mean square values of the terms in the momentum balance at MP. (units: 10− m s− ).

5 m c During Storm 1 Before and After Storm 2 x 8.74 5.64 Coriolis y 14.85 7.50 x 8.17 6.84 pressure gradient y 12.21 8.31 Atmosphere 2019, 10, x FOR PEER REVIEW 17 of 20 x 6.42 2.08 advection x y4.42 6.131.98 2.06 horizontal diffusion x 4.42 1.98 horizontal diffusion y 2.52 0.85 y 2.52 0.85 1 During the storm refers to the period from 12:00 on March 10 to 11:00 on March 12; 2 before and after 1 2 stormDuring refers the to storm the refersperiod to thefrom period 00:00 from on 12:00March on 1009 Marchto 11:00 to11:00 on March on 12 March; 10 and beforethe period and after from storm 12:00 refers on to the period from 00:00 on 09 March to 11:00 on 10 March and the period from 12:00 on 12 March to 00:00 on 14 March. March 12 to 00:00 on March 14. 4.3. Differences in Responses to Winter and Summer Storms 4.3. Differences in Responses to Winter and Summer Storms Using satellite observations, Han et al. [6] found that Hurricane Igor in September 2010 made Using satellite observations, Han et al. [6] found that Hurricane Igor in September 2010 made SST decrease by about 6 ◦C for the area shallower than 200 m southeast of Newfoundland. The SST SST decrease by about 6 °C for the area shallower than 200 m southeast of Newfoundland. The SST cooling in Placentia Bay during Hurricane Igor was about 3 ◦C and that during Hurricane Leslie was cooling in Placentia Bay during Hurricane Igor was about 3 °C and that during Hurricane Leslie was about 2 ◦C[7]. During these tropical storms, colder subsurface water was brought upward to the near about 2 °C [7]. During these tropical storms, colder subsurface water was brought upward to the near surface layer through vertical mixing, leading to sea surface cooling. surface layer through vertical mixing, leading to sea surface cooling. In contrast, the present study reveals a different mechanism of sea surface cooling in response to a In contrast, the present study reveals a different mechanism of sea surface cooling in response winter storm. Figure 15 shows the temporal evolution of temperature at MP during 9–13 March 2017. to a winter storm. Figure 15 shows the temporal evolution of temperature at MP during March 9–13, The temperature is nearly uniform in the vertical before the storm. The storm brings much colder 2017. The temperature is nearly uniform in the vertical before the storm. The storm brings much air (Figure9e) and enhances oceanic heat loss at sea surface. As a result, the ocean temperature in colder air (Figure 9e) and enhances oceanic heat loss at sea surface. As a result, the ocean temperature the upper 100 m decreases by about 0.3 ◦C. Note that the vertical distribution of modelled oceanic in the upper 100 m decreases by about 0.3 °C . Note that the vertical distribution of modelled oceanic temperature needs to be validated against observations which however are unavailable at the present. temperature needs to be validated against observations which however are unavailable at the Using a regional coupled air–sea model, Nelson and He [14] also found that a winter extratropical present. Using a regional coupled air–sea model, Nelson and He [14] also found that a winter cyclone enhanced ocean heat loss over the Gulf Stream. extratropical cyclone enhanced ocean heat loss over the Gulf Stream.

Figure 15. Temporal evolution of temperature during the winter storm. The black dashed line denotes Figure 15. Temporal evolution of temperature during the winter storm. The black dashed line denotes the time when the extratropical cyclone made landfall. The contour interval is 0.05 C. the time when the extratropical cyclone made landfall. The contour interval is 0.05 °C◦ . In terms of ocean currents, Ma et al. [7] showed strong inertial oscillation after the passage of hurricanesIn terms at of MP ocean in summer currents, when Ma theet al. ocean [7] showed was well strong stratified inertial in the oscillation vertical. af Verticalter the stratificationpassage of hurricanes at MP in summer when the ocean was well stratified in the vertical. Vertical stratification favors the generation of the near-inertial oscillation during a storm [7]. In contrast, there is no indication of near-inertial oscillation in neither observed nor modeled currents in response to the winter storm (Figure 10), since the ocean had little vertical stratification.

5. Summary and Conclusion Using satellite measurements, in situ data and model results, we investigate the response of Placentia Bay to the extratropical cyclone of March 11, 2017, which is reported to have the largest wind speed in Newfoundland in more than a decade. The peak wind speed observed by the buoys in Placentia Bay reaches 26.6 m s−1 during the extratropical cyclone event. The storm surge induced by the extratropical cyclone is detected by tide-gauge stations and the Jason−3 altimeter. The

Atmosphere 2019, 10, 724 18 of 20 favors the generation of the near-inertial oscillation during a storm [7]. In contrast, there is no indication of near-inertial oscillation in neither observed nor modeled currents in response to the winter storm (Figure 10), since the ocean had little vertical stratification.

5. Summary and Conclusions Using satellite measurements, in situ data and model results, we investigate the response of Placentia Bay to the extratropical cyclone of 11 March 2017, which is reported to have the largest wind speed in Newfoundland in more than a decade. The peak wind speed observed by the buoys in 1 Placentia Bay reaches 26.6 m s− during the extratropical cyclone event. The storm surge induced by the extratropical cyclone is detected by tide-gauge stations and the Jason-3 altimeter. The maximum storm surges at St. Lawrence (SL) and Argentia (AG) were 0.85 m and 0.77 m, respectively, during the storm. Both buoy and satellite observations show a slight SST drop, 0.42 ◦C from the former and 0.65 ◦C from the latter. The Finite-Volume Community Ocean Model (FVCOM) is employed to examine the oceanic response to the storm in Placentia Bay. The magnitude of storm surge from the model generally agrees with observations, with RMSDs of about 0.15 m. However, the model is noted to underestimate the peak surges at SL. The model captures the slight SST drop during the storm. The root-mean-square differences (RMSDs) of SST between the model and buoy observations are less than 0.7 ◦C. The model surface current changes agree fairly with observations, with SDRs of 0.23–0.34 and VDRs of 0.75–0.88 ◦C. The model storm surge results indicate that the inverse barometric response plays an important role in generating the peak surge at AG and SL due to quite low air pressure and relatively weak winds before and during landfall. The subsequent negative surge is mainly due to the wind effect associated with relatively strong eastward winds at both stations. The model ocean temperature change reveals that the sea surface cooling is associated with the oceanic heat loss during the storm that brings much colder air. There is no indication of near-inertial oscillation in response to the winter storm, due to little oceanic stratification in the vertical. The variations of major terms in the governing equation at the 5 m depth at MP indicate that the storm perturbs the pre-storm momentum balance which is dominated by the Coriolis term and the pressure gradient term. As a result, the advection term and horizontal diffusion term become important as well during the storm. The Coriolis term and the pressure gradient term become dominant again after the storm passes. In future work, we plan to improve the open boundary conditions by using a regional ocean model output, which may help improve the storm surge simulation. Including precipitation and evaporation in the atmospheric forcing can be another aspect for enhancement. Furthermore, the FVCOM model can be coupled with an atmospheric model and a surface wave model to better simulate atmosphere-ocean-wave interactions during winter storms. In addition, more observations, such as vertical profiles data, should be collected and applied to validate the model results.

Author Contributions: Conceptualization, G.X. and G.H.; methodology: G.X.; validation: G.X. and G.H.; formal analysis: G.X.; writing: G.X., G.H., and C.D.; supervision: G.H., C.D., J.Y., and B.D. Funding: This research was funded by National Key Research and Development Program of China (2017YFA0604100, 2016YFA0601803), the National Natural Science Foundation of China (41476022, 41490643), the Startup Foundation for Introducing Talent of Nanjing University of Information Science and Technology (2014r072), the Program for Innovation Research and Entrepreneurship team in Jiangsu Province (2191061503801), the National Programme on Global Change and Air-Sea Interaction (GASI-03-IPOVAI-05), the National Science Foundation of China (OCE 06-23011), the foundation of China Ocean Mineral Resources R and D Association (DY135-E2-2-02, DY135-E2-3-01), and Jiangsu Province, PR China for providing financial support through the Programme for Postgraduate Cultivation and Innovation (KYLX16_0925). This work is also supported by China Scholarship Council and by the Surface Water and Ocean Topography—Canada (SWOT), and the Canadian Space Agency. Acknowledgments: The authors thank NASA/JPL (https://podaac.jpl.nasa.gov/dataset/), AVISO (https://www. ghrsst.org/), Canadian Hydrographic Service (http://www.charts.gc.ca/), SmartAtlantic (http://www.smartatlantic. ca/PlacentiaBay/), AZMP (https://www.canada.ca/en/services/environment/weather.html), and the National Ocean Atmosphere 2019, 10, 724 19 of 20

Partnership Program (https://rda.ucar.edu/) for providing the data used in this study. GX thanks Zhimin Ma and Yingda Xie for helping set up the FVCOM model and Nancy Chen for help processing and analyze satellite altimetry data. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Bernier, N.B.; Thompson, K.R. Predicting the frequency of storm surges and extreme sea levels in the Northwest Atlantic. J. Geophys. Res. Ocean. 2006, 111, C10009. [CrossRef] 2. Chen, C.; Beardsley, R.C.; Luettich, A.L., Jr.; Westerink, J.J.; Wang, H.; Perrie, W.; Xu, Q.; Donahue, A.S.; Qi, J.; Lin, H.; et al. Extratropical storm inundation tested: Inter-model comparisons in Scituate, Massachusetts. J. Geophys. Res. Ocean. 2013, 118, 5054–5073. [CrossRef] 3. Bradbury, I.R.; Snelgrove, P.V.; Fraser, S. Transport and development of eggs and larvae of Atlantic cod, Gadus morhua, in relation to spawning time and location in coastal Newfoundland. Can. J. Fish. Aquat. Sci. 2000, 57, 1761–1772. [CrossRef] 4. Han, G.; Ma, Z.; Chen, D.; DeYoung, B.; Chen, N. Observing storm surges from space: Hurricane Igor off Newfoundland. Sci. Rep. 2012, 2, 1010. [CrossRef][PubMed] 5. Ma, Z.; Han, G.; DeYoung, B. Oceanic responses to Hurricane Igor over the Grand Banks: A modeling study. J. Geophys. Res. Ocean. 2015, 120, 1276–1295. [CrossRef] 6. Han, G.; Ma, Z.; Chen, N. Hurricane Igor impacts on the stratification and phytoplankton bloom over the Grand Banks. J. Mar. Syst. 2012, 100, 19–25. [CrossRef] 7. Ma, Z.; Han, G.; DeYoung, B. Modelling the response of Placentia Bay to hurricanes Igor and Leslie. Ocean Model. 2017, 112, 112–124. [CrossRef] 8. Catto, J.L. Extratropical cyclone classification and its use in climate studies. Rev. Geophys. 2016, 54, 486–520. [CrossRef] 9. Orton, P.M.; Hall, T.M.; Talke, S.A.; Blumberg, A.F.; Georgas, N.; Vinogradov, S. A validated tropical-extratropical flood hazard assessment for New York Harbor. J. Geophys. Res. Ocean. 2016, 121, 8904–8929. [CrossRef] 10. Hirsch, M.E.; DeGaetano, A.T.; Colucci, S.J. An East Coast Winter Storm Climatology. J. Clim. 2001, 14, 882–899. [CrossRef] 11. Colle, B.A.; Booth, J.F.; Chang, E.K.M. A review of historical and future changes of extratropical cyclones and associated impacts along the US East Coast. Curr. Clim. Chang. Rep. 2015, 1, 125–143. [CrossRef] 12. DeGaetano, A.T. Predictability of seasonal east coast winter storm surge impacts with application to New York’s Long Island. Meteorol. Appl. 2008, 15, 231–242. [CrossRef] 13. Booth, J.F.; Rieder, H.E.; Lee, D.E.; Kushnir, Y. The paths of extratropical cyclones associated with wintertime high-wind events in the northeastern United States. J. Appl. Meteorol. Clim. 2015, 54, 1871–1885. [CrossRef] 14. Nelson, J.; He, R. Effect of the Gulf Stream on winter extratropical cyclone outbreaks. Atmos. Sci. Lett. 2012, 13, 311–316. [CrossRef] 15. Hurricane-force Winds Wreak Havoc in Newfoundland. Available online: https://www.theglobeandmail. com/news/national/brier-play-stopped-as-extreme-winds-batter-newfoundland/article34275289/ (accessed on 17 April 2017). 16. Valladeau, G.; Thibaut, P.; Picard, B.; Poisson, J.C.; Tran, N.; Picot, N.; Guillot, A. Using SARAL/AltiKa to improve Ka-band altimeter measurements for coastal zones, hydrology and ice: The PEACHI prototype. Mar. Geod. 2015, 38 (Suppl. 1), 124–142. [CrossRef] 17. Ray, R.D. Precise comparisons of bottom-pressure and altimetric ocean tides. J. Geophys. Res. Ocean. 2013, 118, 4570–4584. [CrossRef] 18. Cartwright, D.E.; Edden, A.C. Corrected tables of tidal harmonics. Geophys. J. R. Astron. Soc. 1973, 33, 253–264. [CrossRef] 19. Wahr, J.M. Deformation induced by polar motion. J. Geophys. Res. Solid Earth 1985, 90, 9363–9368. [CrossRef] 20. Martin, M.; Dash, P.; Ignatov, A.; Banzon, V.; Beggs, H.; Brasnett, B.; Cayula, J.F.; Cummings, J.; Donlon, C.; Gentemann, C.; et al. Group for High Resolution Sea Surface Temperature (GHRSST) analysis fields inter-comparisons. Part 1: A GHRSST multi-product ensemble (GMPE). Deep Sea Res. Part II 2012, 77–80, 21–30. [CrossRef] Atmosphere 2019, 10, 724 20 of 20

21. Dash, P.; Ignatov, A.; Martin, M.; Donlon, C.; Brasnett, B.; Reynolds, R.W.; Banzon, V.; Beggs, H.; Cayula, J.F.; Chao, Y.; et al. Group for High Resolution Sea Surface Temperature (GHRSST) analysis fields inter-comparisons—Part 2: Near real time web-based level 4 SST Quality Monitor (L4-SQUAM). Deep Sea Res. Part II 2012, 77–80, 31–43. [CrossRef] 22. Saha, S.; Moorthi, S.; Wu, X.; Wang, J.; Nadiga, S.; Tripp, P.; Behringer, D.; Hou, Y.T. The NCEP climate forecast system version 2. J. Clim. 2014, 27, 2185–2208. [CrossRef] 23. Bleck, R.; Halliwell, G.R., Jr.; Wallcraft, A.J.; Carroll, S.; Kelly, K.; Rushing, K. Hybrid Coordinate Ocean Model (HYCOM) User’s Manual: Details of the Numerical Code. Available online: http://hycom.rsmas.miami.edu (accessed on 12 May 2017). 24. Chen, C.; Liu, H.; Beardsley, R.C. An unstructured grid, finite-volume, three-dimensional, primitive equations ocean model: Application to coastal ocean and estuaries. J. Atmos. Ocean. Technol. 2003, 20, 159–186. [CrossRef] 25. Rego, J.L.; Li, C. Storm surge propagation in Galveston Bay during Hurricane Ike. J. Mar. Syst. 2010, 82, 265–279. [CrossRef] 26. Han, G.; Ma, Z.; DeYoung, B.; Foreman, M.; Chen, N. Simulation of three-dimensional circulation and hydrography over the Grand Banks of Newfoundland. Ocean Model. 2011, 40, 199–210. [CrossRef] 27. Burchard, H.; Bolding, K.; Villarreal, M.R. GOTM, a General Ocean Turbulence Model: Theory, Implementation and Test Cases; Technical Reports; Space Applications Institute: Gujarat, India, 1999. 28. Mellor, G.L.; Ezer, T.; Oey, L.Y. The pressure gradient conundrum of sigma coordinates ocean models. J. Atmos. Ocean. Technol. 1994, 11, 1126–1134. [CrossRef] 29. Ma, Z.; Han, G.; DeYoung, B. Modelling temperature, currents and stratification in Placentia Bay. Atmos. Ocean 2012, 50, 244–260. [CrossRef] 30. Egbert, G.D.; Erofeeva, S.Y. Efficient inverse modeling of barotropic ocean tides. J. Atmos. Ocean. Technol. 2002, 19, 183–204. [CrossRef] 31. Dong, C.; McWilliams, J.C.; Hall, A.; Hughes, M. Numerical simulation of a synoptic event in the Southern California Bight. J. Geophys. Res. Ocean. 2011, 116, C05018. [CrossRef] 32. Jiang, J.; Perrie, W. The impacts of climate change on autumn north Atlantic midlatitude cyclones. J. Clim. 2007, 20, 1174–1187. [CrossRef] 33. Yao, Y.; Perrie, W.; Zhang, W.; Jiang, J. Characteristics of atmosphere-ocean interactions along North Atlantic extratropical storms tracks. J. Geophys. Res. Atmos. 2008, 113, D14124. [CrossRef] 34. Han, G.; Lu, Z.; Wang, Z.; Helbig, J.; Chen, N.; DeYoung, B. Seasonal variability of the Labrador Current and shelf circulation off Newfoundland. J. Geophys. Res. Ocean. 2008, 113, C10013. [CrossRef] 35. Pawlowicz, R.; Beardsley, B.; Lentz, S. Classical tidal harmonic analysis including error estimates in MATLAB using T_TIDE. Comput. Geosci. 2002, 28, 929–937. [CrossRef] 36. Han, G.; Ma, Z.; Chen, N.; Chen, N.; Yang, J.; Chen, D. storm surges off observed by Jason-1 and Jason-2 satellite altimeters. Remote Sens. Environ. 2017, 198, 244–253. [CrossRef] 37. Sheng, J.; Zhai, X.; Greatbatch, R.J. Numerical study of the storm-induced circulation on the Scotian Shelf during Hurricane Juan using a nested-grid ocean model. Prog. Oceanogr. 2006, 70, 233–254. [CrossRef] 38. Bunya, S.; Dietrich, J.C.; Westerink, J.J.; Ebersole, B.B.; Smith, J.M.; Atkinson, J.H.; Jensen, R.; Resio, D.T.; Luettich, R.A.; Dawson, C.; et al. A high-resolution coupled riverine flow, tide, wind, wind wave, and storm surge model for Southern Louisiana and Mississippi. Part I: Model Development and Validation. Mon. Weather Rev. 2010, 138, 345–377. [CrossRef] 39. Mecking, J.V.; Fogarty, C.T.; Greatbatch, R.J.; Sheng, J.; Mercer, D. Using atmospheric model output to simulate the meteorological tsunami response to Tropical Storm Helene (2000). J. Geophys. Res. Ocean. 2009, 114, C10005. [CrossRef] 40. Chen, C.; Huang, H.; Beardsley, R.C.; Liu, H.; Xu, Q.; Cowles, G. A finite volume numerical approach for coastal ocean circulation studies: Comparisons with finite difference models. J. Geophys. Res. Ocean. 2007, 112, C03018. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).