TYPE Original Research

PUBLISHED 01 May 2024

DOI 10.3389/fmars.2024.1332883

Tropical cyclone wave data

OPEN ACCESS assimilation impact on air-
EDITED BY
Sudheer Joseph,
Indian National Centre for Ocean Information
ocean-wave coupled Hurricane
Services, India

REVIEWED BY
Harvey (2017) forecast
Charles Reid Nichols,
Marine Information Resources Corporation, Sue Chen 1*, James A. Cummings 2, Jayaram Veeramony 3
United States
Vijay Tallapragada, and Justin S. Tsu 1
NCEP Environmental Modeling Center (EMC),
United States
1
Naval Research Laboratory, Marine Meteorology Division, Monterey, CA, United States, 2 Science
Applications International Corporation, Monterey, CA, United States, 3 Naval Research Laboratory,
*CORRESPONDENCE Ocean Sciences Division, Stennis Space Center, MS, United States
Sue Chen
sue.chen@nrlmry.navy.mil

RECEIVED 03 November 2023

The impact of surface wave assimilation on hurricane track and intensity
ACCEPTED 02 April 2024
PUBLISHED 01 May 2024 forecasts has been investigated using a fully coupled air-ocean-wave tropical
cyclone data assimilation and forecast modeling system. A new 3DVAR wave
CITATION
Chen S, Cummings JA, Veeramony J and assimilation method in the Navy Coupled Ocean Data Assimilation system
Tsu JS (2024) Tropical cyclone wave data (NCODA) maps the 1D wave energy spectra from buoys to 2D directional wave
assimilation impact on air-ocean-wave
coupled Hurricane Harvey (2017) forecast. energy spectra using the maximum likelihood method (MLM) and corrects the
Front. Mar. Sci. 11:1332883. wave model forecast component directional wave energy spectra. The Coupled
doi: 10.3389/fmars.2024.1332883
Ocean/Atmosphere Mesoscale Prediction System for Tropical Cyclone
COPYRIGHT Prediction (COAMPS-TC) is used to conduct three Hurricane Harvey (2017) air-
© 2024 Chen, Cummings, Veeramony and Tsu.
This is an open-access article distributed under ocean-wave coupled data assimilation and forecasting experiments with and
the terms of the Creative Commons Attribution without the wave data assimilation. Hurricane Harvey traversed through the
License (CC BY). The use, distribution or
Western Gulf of Mexico from 24 August to 1 September, 2017 and made landfall
reproduction in other forums is permitted,
provided the original author(s) and the in the Texas and Louisiana coast. Validation of track, maximum wind speed,
copyright owner(s) are credited and that the significant wave height, and mean absolute wave periods show wave assimilation
original publication in this journal is cited, in
accordance with accepted academic
of the 1D wave energy spectra from 13 National Data Buoy Center (NDBC) buoys
practice. No use, distribution or reproduction reduced the forecast errors of these parameters compared to experiments
is permitted which does not comply with without the wave assimilation. In spite of this positive outcome, the wave
these terms.
assimilation is unable to reduce Harvey’s 0-120 h forecast mean wave
direction errors and correlation compared to the NDBC buoy time series

KEYWORDS

tropical cyclone, hurricane, wave data assimilation, coupled model prediction,

air-sea interaction

1 Introduction
Tropical Cyclones (TC) are one of the most severe natural disasters that cause
devastating damage on societies worldwide from the combination of winds, tornados,
flood, and storm surge in both coastal and inland regions. The advancements of TC
dynamic models in recent decade have made steady inroads to improve deterministic and

Frontiers in Marine Science 01 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

probability forecasts of TC track, intensity, winds, and waves the surface wind and pressure field will generally result in higher
(Heming et al., 2019; Cangialosi et al., 2020). In particular, skill in the wave forecasts. However, direct improvement of wave
increased capabilities to observe the ocean wave energy spectra forecasts using data assimilation is also possible, but the sparsity of
routinely by operational buoy observing system or by targeted real-time wave data is a core issue for wave assimilation systems,
sampling of TC provide an opportunity to investigate the wave especially under tropical cyclone (TC) conditions. Buoy networks
spectra data assimilation impact on TC prediction. Inclusion of provide highly accurate estimates of wave spectrum statistics, but
wave observations to improve operational forecasting of wave these data are sparse due to the cost of deployment and
conditions has high impact in the coastal zones where a large maintenance of the buoys. In the U.S., the National Buoy Data
population lives. Center (NDBC) deploys fixed buoys near the coast that collect a
The fundamental concept underpinning spectral wave variety of observations useful in the monitoring of Hurricane wave
modeling is the energy balance equation. This states that the conditions. These measurements include accelerometers or
evolution of the wave spectrum is the sum of three source terms inclinometers on board the buoys that measure the heave,
describing the input of energy from the wind, the nonlinear transfer acceleration, and the vertical displacement of the buoy hull
of energy within the wave variance spectrum, and the dissipation of (NDBC, 2009; Hall et al., 2022), as well as winds and ocean
energy from wave breaking or shallow water processes (Komen temperatures. The in situ wave observations complement the
et al., 1994). Wave models compute an explicit representation of all satellite remote sensing of surface waves from altimeters, which
three source terms and the evolution of the wave spectrum, without has been available since the 1980s and assimilated by many
a priori assumptions about the spectral shape. Even though various operational centers worldwide. The altimeter wave estimates are
data assimilation techniques have been used to improve model valuable but they carry considerable uncertainty and the spatial and
predictions (see Bannister, 2017 for a review) in an operational temporal sampling characteristics of the altimeters limit their
environment, these techniques have been applied largely to the effectiveness (Wittmann and Cummings, 2004; Tran et al., 2010;
atmospheric and ocean circulation models. Wave model runs with Fan et al., 2012; Seemanth et al., 2021).
assimilated data have been shown to significantly reduce errors in More fundamentally, wave forecasting is a forcing problem
modeled wave parameters. Lionello et al. (1992) used an optimal while numerical weather prediction is an initial value problem. A
interpolation technique (OI) to assimilate significant wave heights considerable part of the ocean wave field is formed directly by
derived from satellite altimeter data to correct the total energy in the surface winds and near-surface pressure fluctuations in the
wave spectrum. A similar technique was used by Voorrips et al. atmosphere (i.e., wind sea components). As a result, errors in the
(1997) to assimilate significant wave height data from wave buoys. forecasting of wind-sea wave conditions are reduced when the skill
Smit et al. (2021) took the OI technique further by assimilating bulk of the weather models is improved with data assimilation. The
wave parameters including significant wave height and wave period assimilation of wave observations correct wave model errors, but if
from a large distributed buoy network, 129 free-drifting Sofar the wave field is corrected without also correcting the atmospheric
Spotters buoys, to analyze the potential impact on the total forcing field to match, the wave field will rapidly return to the state
energy. They reported the forecast improvement from wave dictated by the forcing. In this regard, a coupled ocean, atmosphere,
assimilation extends to 4 days. In all these cases, the aim was to wave modeling system that includes data assimilation in all of the
improve the wave energy in the modeled domain. Veeramony et al. coupled system components would be expected to maintain forecast
(2010) and Orzech et al. (2014) showed that for smaller domains skill for much longer than an uncoupled system.
focused on the nearshore region, a 4D variational assimilation Building upon the previous aforementioned studies, this study
(4DVar) system assimilating the wave spectrum to correct the investigates a wave model assimilation methodology that assimilate
wave boundary conditions provided significant improvements in both routinely available observations of wave energy spectra from
reproducing the directional wave spectra in the domain. Compared 13 NDBC buoys (Table 1; Kuik et al., 1988; Riley et al., 2019) and
to the 3DVAR data assimilation system, the added computation satellite altimeter derived altimeter significant wave height
cost of using 4DVAR may not be feasible for operational measurements in a fully coupled air-ocean-wave tropical cyclone
implementation because of wall time constrain. Conversely, data assimilation and forecast modeling system. Because there are
recent studies using a 3DVAR significant wave height many NDBC buoys in the Gulf of Mexico (GOM) region, the
assimilation show improvement of wave forecast up to 24 h Hurricane Harvey (2017) case is ideally used to study the impact of
(Saulter et al., 2020; Byrne et al., 2023). Using the most the directional wave spectra assimilation.
computational expansive ensemble-based assimilation of Hurricane Harvey entered the Gulf of Mexico on 23 August,
significant wave height method can further extend the wave rapidly intensified and made landfall near Port Aransas, Texas
forecast improvement out to 2.5 days (Houghton et al., 2023). around 10 p.m. on 25 August (03 UTC 26 August) as a devastating
The success of data assimilation in atmospheric and Saffir-Simpson category (CAT) 4 TC (Saffir 1973; Simpson, 1974).
oceanographic forecast systems is made possible by a considerable Hurricane Harvey’s rapid intensification near landfall was due to a
amount of routine observational data from a variety of sources. pre-existing very warm (~ 31°C), and well-mixed ocean
Data assimilation in atmospheric forecasts indirectly improve ocean temperature along the Texas Bight (Potter et al., 2019). They
and wave models through improved estimates of surface wind attributed the absence of cold ocean water in the shallow Texas
stresses, surface pressure, and air-sea fluxes (Sanabia et al., 2013; Bight shelf enabled Hurricane Harvey to draw from this warm
Chen et al., 2017). For wave models, specifically, improvements in ocean energy source and rapidly intensified from CAT3 to CAT4

Frontiers in Marine Science 02 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

TABLE 1 A list of 13 NDBC SCOOP payload buoy attribute used in the

COAMPS WW3 Hurricane Harvey wave assimilation experiment.

NDBC Size/type Location Buoy

Buoy ID Water depth
42001 3-meter foam buoy 25.926 N 3200 m
89.662 W

42002 3-meter foam buoy 26.055 N 3088 m

93.646 W

42020 3-meter foam buoy 26.955 N 84 m

96.687 W

42019 2.1-meter ionomer 27.910 N 83.5 m

foam buoy 95.345 W

42035 3-meter foam buoy 29.237 N 15 m

94.404 W

42036 3-meter discus buoy 28.501 N 50.9 m FIGURE 1

84.508 W Hurricane Harvey’s best track from 00 UTC 24 August to 00 UTC 2

September 2017. The colors are Harvey’s intensities shown on the
42039 3-meter foam buoy 28.787 N 281 m legend. Diamonds and 5-digit numbers are the NDBC buoy stations.
86.007 W The best track is from IBTrACS.

42040 3-meter foam buoy 29.207 N 192 m

88.237 W
(NCOM, Martin et al., 2013), and the WAVEWATCH III (WW3)
42055 3-meter foam buoy 22.140 N 3508 m
wave model (WW3DG, T.W.I.D.G, 2019). The coupling between
SCOOP payload 94.112 W
the atmospheric model, NCOM, and WW3 is through the Earth
42056 3-meter foam buoy 19.820 N 4554 m
System Modeling Framework NUOPC (Theurich et al., 2016) with a
84.945 W
12-minute coupling frequency.
42057 3-meter discus buoy 16.973 N 412 m COAMPS domains are configured focusing on the GOM region
81.575 W
and the exchange of forcing and feedback are three-way, i.e.,
42058 3-meter discus buoy 14.844 N 4131 m between air-ocean, air-wave, and ocean-wave, using exchange
75.061 W
grids between each pair of component models (Campbell et al.,
42059 3-meter foam buoy 15.300 N 4761 m 2010). The atmospheric model uses the COAMPS-TC physics suite
67.483 W but with three fixed domains with 36, 12, and 4 km horizontal grid
spacing. The model forecast Harvey track and maximum wind
before landfall. The Harvey track moved back offshore on 29 speed (intensity) every 6 hours is determined by running a NOAA
August, and subsequently made a final landfall west of Cameron, Geophysical Fluid Dynamics Laboratory (GFDL) tracker (Marchok,
Louisiana on 30 August while maintaining tropical storm strength 2021). As for the ocean and wave forecast component models,
(Figure 1). Harvey’s heavy precipitation caused wide spread severe NCOM is configured to use one domain with 6 km horizontal grid
flooding, especially in the Huston, Texas area. The extreme flooding spacing and WW3 is also one domain with 12 km horizontal
was compounded by heavy precipitation, river overflow, and spacing. The WW3 model is setup to run with 34 frequencies
onshore ocean water surge (Valle-Levinson et al., 2020). In from 0.0418 Hz with interval of 1.1 Hz and 48 directions.
combination with Hurricane Harvey’s high winds and tornadoes Wave forcing to the atmospheric component model is by a sea-
(Wurman and Kosiba, 2018; Nowotarski et al., 2021), Harvey state dependent Charnock parameter (Charnock, 1955) which is
resulted in ~125 billion U.S. dollars of damage along the Texas computed in WW3 based on the Janssen (1991) wind-wave
and Louisiana coastal zone (National Hurricane Center, 2017). generation. Wave forcing to NCOM ocean model turbulent kinetic
energy (TKE) includes, (a) in the coastal and offshore ocean region:
Stokes drift currents from waves and the wave orbital motion near the
2 Materials and methods ocean bottom (Allard et al., 2012; Smith et al., 2013), and (b) in the
surf zone: the additional depth-dependent wave radiation stress to
2.1 Coupled model configuration transfer the wave momentum from breaking waves to the ocean
circulation (Martin et al., 2020). The atmospheric forcing in WW3
The coupled air-ocean-wave model used in this study is the uses the 10 m wind speed, and NCOM forcing in WW3 are the sea
Coupled Ocean/Atmosphere Mesoscale Prediction System surface height and surface currents (Allard et al., 2012; Chen et al.,
(COAMPS, Chen et al., 2003; Chen et al., 2015; Allard et al., 2015). Atmospheric forcing to the ocean includes sea level pressure,
2010). The coupled model configuration used in this study 10 m wind stress, surface heat and moisture fluxes, and shortwave
consists of the COAMPS atmospheric model for tropical cyclone radiation. NCOM forcing to the atmosphere is by the sea surface
(COAMPS-TC, Doyle et al., 2012), the Navy Coastal Ocean Model temperature (Campbell et al., 2010).

Frontiers in Marine Science 03 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

Global model forecasts from the NOAA Global Forecasting the sea surface and infers SWH from the shape and position of the
System (GFS), NRL Global Ocean Forecasting System (GOFS 3.1; leading edge of the waveform data. The altimeter SWH retrieval
Metzger et al., 2010), and global WW3 provide the initial and lateral algorithms have difficulty deriving SWH in very low wave height
boundary conditions for COAMPS simulations. Additional to the environments and also near shore where radar reflections from land
GFS initial condition, a TC bogus is used to replace the GFS contaminate the signal. SWH is an integral wave parameter and can
circulation in TC areas based on the JTWC warning message be derived from 2D directional wave spectra obtained
(Komaromi et al., 2021). Procedures to insert a TC bogus vortex from observations.
into COAMPS-TC consist of several steps. Step 1 removes the GFS
TC vortex winds. This is accomplished by removing the vorticity 2.2.2 Spectral wave data
and divergence component of the GFS winds. Step 2 performs direct Spectral wave data are obtained from fixed buoys. For fixed
solver iteration to balance the pressure, non-divergent winds, and buoys the wave measurements are inferred from the accelerometers
thermodynamics variables using a non-linear balancing equation or inclinometers on board the buoys that measure the heave
and variational adjustment method (Barker, 1980). Step 3 computes acceleration or the vertical displacement of the buoy hull during
a Rankin vortex (Holland et al., 2010) that uses the radius of the wave acquisition time. A Fast Fourier Transform (FFT) is
maximum wind radius to build a sloping eyewall wind structure at applied to the data by the processor on board the buoy to
the TC location from the JTWC warning message. The bogus vortex transform the data from the temporal domain into the frequency
depth and structure of the secondary circulation are set empirically. domain. From this transformation wave energies with their
Step 4 blends large-scale winds with the Rankin vortex. The final associated frequencies are derived. A limitation of the fixed buoy
step 5 inserts the TC bogus vortex in nest 3 using a Barnes analysis network is that the buoys are primarily positioned in shelf and
scheme (Barnes, 1964). coastal areas and do not provide open-ocean directional wave
Eight 5-day coupled COAMPS simulations for Hurricane observations. However, they provide valuable in situ observations
Harvey are carried out from 12 UTC 24 to 00 UTC 30 August, to monitor and improve model initial ocean and wave conditions
2017 cycling every 12-h using the Navy Coupled Ocean Data ahead and underneath the hurricanes. For the study, 13 NDBC
Assimilation (NCODA) for the ocean (Cummings and Smedstad, buoys that observed sea surface temperature and wave spectrum are
2013) and wave component. NCODA assimilates ocean and wave assimilated into NCODA (Figure 1).
observations from both satellite and in situ data sources.
Previous studies have shown assimilation of ocean observation 2.2.3 Ocean and wave model data
improved air-ocean coupled COAMPS-TC intensity forecast assimilation method
(Sanabia et al., 2013; Chen et al., 2017). The current study uses a NCODA uses a 3DVAR scheme to initialize the ocean model
new wave assimilation method in NCODA. Three fully coupled air- and a 2DVAR scheme to initialize the wave model. The
ocean-wave COAMPS experiments are used to quantify wave implementation is based on a “weakly” coupled data assimilation
assimilation impact on Harvey’s forecast. They include approach whereby the coupled model is used to provide
experiments without the ocean and wave assimilation (NDA), background information for separate ocean, atmosphere, and
with the ocean assimilation only (ODA), and with the wave wave model analyses. The analysis increments generated from
assimilation only (WDA). Experiment ODA assimilates surface these separate analyses are then added back into the next cycle of
and subsurface ocean temperature and salinity from buoys, floats, coupled model background fields. The mismatch in time scales
drifters, and satellite. The ODA also assimilates the Altimeter sea among the atmosphere, ocean, and wave models is a major difficulty
surface height observations. Experiment WDA only assimilates the as each of the models must use the same time window in the weakly
2D directional wave spectrum derived from buoys and altimeter coupled assimilation. We have taken the approach of using an
significant wave height. appropriately short time window of 12 hours for all of the
model components.
NCODA is an oceanographic implementation of the Navy
2.2 Assimilation method and wave data Variational Atmospheric Data Assimilation System (NAVDAS), a
3DVAR technique developed for Navy numerical weather
Two sources of wave field observations are assimilated in the prediction systems (Daley and Barker, 2001). The 3DVAR
COAMPS WW3 model and are described here: analysis variables are temperature, salinity, geopotential (dynamic
height), and u, v vector velocity components. All ocean variables are
2.2.1 Significant wave height analyzed simultaneously in three dimensions. The horizontal
Significant wave height (SWH) is defined as the average wave correlations are multivariate in geopotential and velocity, thereby
height (from trough to crest) of the highest third of the waves in a permitting adjustments to the mass fields to be correlated with
given sampling period. Satellite altimeters are the primary means of adjustments to the flow fields. The velocity adjustments (or
providing near-global measurements of SWH. The measurements increments) are in geostrophic balance with the geopotential
are made at nadir along the satellite tracks with resolutions of 5 to 7 increments, which, in turn, are in hydrostatic agreement with the
km. The standard algorithm for retrieving SWH from satellite temperature and salinity increments. The NCODA 3DVAR
altimeters fits a modeled shape to the radar echo waveforms from problem is formulated in observation space as:

Frontiers in Marine Science 04 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

xa = xb + Pb H T (HPb H T + R)−1 ½y − H(xb ) Fourier coefficients (r1, r2), the mean wave direction (alpha1) and
the principal wave direction (alpha2) based on the Fourier series
where xa is the analysis vector, xb is the background vector, Pb is expansion originally developed by Longuet-Higgins et al. (1963).
the background error covariance matrix, H is the forward operator, This required a modification to the H forward operator in the
R is the observation error covariance matrix, and y is the 3DVAR. For other data types, the forward operator is spatial
observation vector. HP b HT is approximated directly by the interpolation performed in two or three dimensions by fitting a
background error covariance between observation locations, and surface to a 4x4 or a 4x4x4 grid point target and evaluating the
PbHT directly by the error covariance between observation and grid surface at the observation location.
locations. The quantity [y-H(xb)] is referred to as the innovation
Benoit (1992), (1994) describes the methods commonly used to
vector, [y-H(xa)] is the residual vector, and xa-xb is the increment reconstitute the directional wave spectra from the reported values.
(or correction) vector.
Of these, there are two main methods that are widely used: The
The oceanographic error covariances in the 3DVAR are maximum likelyhood method (MLM), which was developed and
separated into a background error variance and a correlation. The
presented in detail by Oltman-Shay and Guza (1984), and the
correlation is further separated into a horizontal and a vertical maximum entropy method (MEM) developed and presented in
component. Horizontal correlation length scales vary with location
detail by Lygre and Krogstad (1986). Earle et al. (1999) show that
and vertical correlation length scales vary with location and depth. while MEM provides better resolution, it produces artificially
Flow-dependence is introduced by modifying the horizontal
narrow directional spreading. For the purposes of this study, it
correlations with a tensor obtained from a model forecast was determined that the accuracy in directional spreading provided
background field that is used as a proxy for the circulation field.
by MLM is more important than the resolution since the model
Example flow fields include sea surface height, sea surface resolution used is 7.5°, which is reasonable for this application, but
temperature, and potential vorticity. The flow-dependent tensor
not high enough to necessitate using MEL. Once the 2D spectral
spreads innovations along rather than across the flow field density is obtained from the values reported by the buoys, the
gradients, which is a desirable outcome in the analysis. The
observed spectra are compared to the model spectra using nearest
discontinuous and non-homogeneous influence of coastlines in neighbor. The differences in wave energy as a function of frequency
the analysis uses a second tensor based on distance to land. Near
and direction are computed which form the (y – H(xb)) innovations
the coast the tensor rotates and stretches the horizontal correlations in the 3DVAR.
along the coast while minimizing or removing correlations into the
land. Error correlations are expected to be anisotropic near
coastlines due to horizontal advection from coastal currents that
flow parallel to the coast.
3 Results
Assimilation of wave observations into the wave model
component of the coupled system required several modifications
3.1 Buoy observations
to the 3DVAR. First, the wave model prognostic variable is wave
Buoys 42002, 42019, 42020, and 42035 have a minimum
energy as a function of location, direction, and frequency. As such,
distance to Harvey’s track in GOM about 74, 57, 37, and 37 nm
all analysis updates or corrections to the wave model must be done
respectively during Harvey’s traverse toward Texas coast. Buoy
to the model forecast directional wave spectra at each model grid
42002 and 42019 are located to the right of Harvey’s track and buoy
point. Second, the error covariances were modified to include not
42020 is located to the left of the Harvey’s track (Figure 1).
only correlations between observations and grid points but also
Interestingly because Harvey’s track moved back out to sea after
correlations within wave spectra in terms of frequency and
the first landfall on 26 August, this movement put buoy 42035 to the
direction. Assimilation of integral measures of wave energy, such
right of Harvey’s track at this time before the first landfall, then
as altimeter SWH, is achieved in two steps. First, an analyzed SWH
switched to the left of the track after Harvey’s final landfall on 30
increment field is added to the WW3 SWH forecast (Hf) to produce
August. Buoys 42002, 42019, and 42020 all have high values of
a corrected SWH analysis field (Ha). The analyzed wave model
spectral wave density > 50 m2 Hz-1 around 25-26 August when
spectrum (Fa) as a function of frequency (f) and direction (Q) is
these buoys were closest to the Harvey track. During this period, all
then obtained from the ratio of analyzed and forecast SWH fields to
three buoys exhibit a dominant wind-sea wave with range of wind
produce an updated forecast spectrum (Ff) using a simple scaling
speed (ms-1) of 10-27 ms-1 (1 ms-1 = 1.943844492 kt) mean wave
strategy,
period of 10-14s, and significant wave height of 3-7 m (top panel of
a = (H a =H f )2 Figures 2A–C). A research mooring south of Port Aransas at 20 m
water depth recorded a 10 m significant wave height on 26 August,
F a (f , Q) = aF f (f , Q)
2017 (Romero-Arteaga et al., 2022).
Assimilation of the NDBC buoy spectral wave data is more The significant wave height from buoy 42002 relaxes back down
straightforward in that the observations measure the directional to around 5 m after the passage of Harvey between 27-28 August
wave spectrum. However, the 2D wave data provided by the NDBC and has the signature of swell with mean wave period< 5 s. During
are in the form of the non-directional spectral wave density as a this time, buoys 42019 and 42020 mean wave period are ~ 6-8 s,
function of the frequency along with the normalized directional indicating a mixed wind and swell sea. All three buoys’ significant

Frontiers in Marine Science 05 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

wave height and mean wave period increase again after Harvey’s spectra density maxima on the two front quadrants of Harvey
track moved back out to the sea on 28 August. Compared to these before Harvey’s final landfall. The slow-down of Harvey’s track that
three buoys, buoy 42035 experiences a lower wind speed range of 8- turned back out to sea after the first landfall and subsequent
16 m s-1, mean wave period of 6-10 s, and 2-3 m significant wave landfalls reduced the wind intensity, hence the weaker 2D
height during 26-29 August (top panel of Figure 2D). The wind directional wave spectral density over two closest distance buoys
speed increases to 19 m s-1 but with receding significant wave height 42019 and 42035. The furthest buoy from Harvey’s track is 42002,
to 2 m and mean wave period ~ 7 s when Harvey’s track moved which has the largest 2D directional wave spectral density. For the
back out to sea. Buoy 42035 is located on the shallow Texas– earlier three time periods, waves from all four buoys show their
Louisiana continental shelf and because the wind direction (middle maximum 2D directional wave spectral density rotate from
panel of Figure 2D) during this time is northerly, prevents the high Harvey’s front left quadrant to rear-right quadrant. Additionally,
wave surge at this location. Buoy 42035 also has the lowest spectral buoy 42035 (Figure 4D) has the smallest 2D directional wave
wave density compared to the other three buoys (bottom panel spectral density compared to the other three buoys (Figures 4A–C).
of Figure 2D).
Figure 3 shows two 2D wave density spectra using MLM from
buoys 42002 (3a) and 42020 (3b) at 05 UTC (mid night Texaslocal 3.2 Wave data assimilation impact on
time) 25 August when Harvey reached a CAT 4 intensity. The COAMPS TC track and intensity forecast
maximum 2D wave spectra density are 0.36 and 0.68 m2 Hz-1,
respectively and located on the front-right quadrant (0° points to Three fully coupled air-ocean-wave COAMPS experiments are
north). The time series of mean 2D wave density spectra for all experiments without the ocean and wave assimilation (NDA), with
frequencies from these four buoys show four separate periods of 2D the ocean assimilation only (ODA), and with wave assimilation only
wave spectra density (m2 Hz-1 degree-1) on 25-26 August, 27-28 (WDA). Comparison of COAMPS forecast track from each
August, 28-29 August, and 30 August–1 September (Figure 4). All experiment shows all three COAMPS experiments have a smaller
four buoys show a fully developed sea from all directions after 30 track error over the ocean than after landfall (Figure 5). The best
August and maximum 2D directional wave spectra density track is from the International Best Track Archive for Climate
magnitudes at this time. They exhibit 1-3 2D directional wave Stewardship (IBTrACS) version 4 (Knapp et al., 2018) which shows

A C

B D

FIGURE 2
NDBC buoy observations from buoy (A) 42002, (B) 42019, (C) 42020, and (D) 42035. Top panel is the time series of buoy observed wind speed (m
s-1, blue line), mean wave period (s-1, orange line), and significant wave height (m, black line). Bottom panel is the time series of spectra wave density
(m2 Hz-1).

Frontiers in Marine Science 06 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

FIGURE 3
Buoy 42002 and 42020 2D directional wave spectral density (m2 Hz-1 degree-1) in polar coordinate on 05 UTC 25 August, 2017. At this time
Hurricane Harvey reached a CAT 4 intensity. The 2D directional wave spectral density is derived from the buoy 1D spectral wave density using the
MLM method.

Harvey degraded to tropical storm after the last landfall and moved ~ 3-4 nm compared to the no ocean and wave DA experiments
rapidly northeastward after 31 August. In contrast, the last 00 UTC (Table 2). The maximum reduction of absolute mean track error
28 August COAMPS runs for all three experiments have a from WDA experiment is about ~ 9 nm (1 nm = 1.852 km), i.e., 50
northward track and move much slower than the observation. nm compared to the NDA experiment’s 59 nm, at 66 h
The time series of mean absolute track error shows WDA (red forecast time.
line in Figure 6) is smaller for most of the forecast lead times. The 5- Figure 7A is a spaghetti plot of forecast maximum wind speed
day mean absolute track error for experiments NDA, ODA, and (kt) that show all COAMPS experiments have a small low intensity
WDA are 72.9, 73.2, and 69.1 nm, respectively. Assimilating wave bias compared to IBTrACS. The homogenous time series of mean
data clearly improves COAMPS Harvey track forecast by a mean of intensity bias of WDA experiment is better than the NDA and ODA

A C

B D

FIGURE 4
Time series of mean spectral wave density (m2 Hz-1 degree-1) from buoy (A) 42002, (B) 42019), (C) 42020, and (D) 42035.

Frontiers in Marine Science 07 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

TABLE 2 COAMPS 0-5 day forecast mean track and intensity errors for
each experiments.

Experiment 0-5-day mean 0-5-day mean

track error (nm) intensity error (kt)
No DA (NDA) 72.9 -5.5

Ocean 73.2 -8.6

DA (ODA)

Wave 69.1 -5.4

DA (WDA)

The forecast period is from 12UTC 24 August - 00 UTC 30 August, 2017. The sample size for
each experiment is 184, i.e. 8 forecast cycles times 23 of 6 hourly data.

Harvey track (Figure 8 green and brown color shadings). These

patterns are due to WDA having a slower translation speed than the
NDA experiment. The significant wave height forecast differences
FIGURE 5
A spegettii plot of COAMPS Harvey forecast tracks from the NDA show WDA with a persistently ~ 0.2 m higher wave height than
(blue lines), ODA (cyane lines), and OWDA (red lines) experiments. NDA along the Mississippi Alabama coastal region. For longer lead
Harvey’s best track is shown in black line. The squares are the
Harvey’s best track from IBTrACS maked at every 6 h interval.
time forecast cycles, WDA shows a larger/smaller wave height on
the right/left of forecast track than NDA (Figures 8A–C). These
along track forecast wave height difference patterns are due to
experiments except at the 20 h forecast lead time (Figure 7B). The slightly track differences and WDA possessing a larger right-
WDA experiment also has the smallest mean 5-day intensity error asymmetry wind than NDA. However, for shorter forecast lead
of -5.4 kt compared to NDA and ODA experiments’ -5.5 and -8.6 times (Figure 8D), the along track significant wave height
kt, respectively (Table 2). Notably, the ODA experiment has the is negligible.
largest mean -18 kt low intensity bias for forecast lead time > 60 h
after Harvey’s final landfall than the other two experiments.
Assimilating the wave observations clearly improves the intensity 3.3 Wave data assimilation impact on
forecast from assimilating ocean only observations for COAMPS COAMPS wave forecast
Harvey forecast.
Aside from the track position and intensity differences, the The validation of COAMPS WW3 forecasts uses NDBC buoy
WDA minus NDA 10 m wind speed differences valid at Harvey’s observation of wave parameters that are not assimilated by
first landfall on 03 UTC 26 August from four forecast cycles reveal NCODA. These are the significant wave height (m), mean wave
alternate positive and negative difference patterns along forecast direction (°) and absolute mean wave period (s). Note that NCODA
uses buoy’s 2D directional wave spectral density to derive
significant wave height for assimilation (see section 2.1).
COAMPS WW3 0–120-hour forecasts of these three variables
from each forecast cycle are first interpolated to the buoys 42002,
42019, 42020, and 42035 locations, then forecast bias error (forecast
minus observation) and root mean square error (RMSE) are
computed. COAMPS forecast time is valid at the observation
time. The correlation is evaluated using each pair of forecast and
observation time series at each buoy location. And lastly, The mean
of the last 5 forecast cycles is computed to account for two
DA cycling adjustment times for the NDA, ODA, and
WDA experiments.
For the wave parameters evaluated, all experiments have a
similar mean error trend and magnitude. The mean significant
wave height bias for forecast lead times under 60 h is about -0.5 m
and reduces to near 0 for longer lead forecast time greater than 60 h.
By the end of 120 h, the mean bias becomes positive at about 0.2 m.
FIGURE 6
The 0-120 h time series of COAMPS forecast mean absolute track The RMSE is less than 0.6 m for all forecast lead times (Figure 9).
error (nm) from the NDA (blue line), ODA (cyan line), and WDA (red The mean wave direction bias oscillates between -20° to 20° with
line) experiments. NDA is experiment without the ocean and wave
assimilation, ODA is with the ocean assimilation only, and WDA is
minimum and maximum RMSE of 10° and 90° (Figure 10). The
with the wave assimilation only. mean absolute wave period bias is less than 0.5 s with RMSE
between ~ 0.5-1.0 s (Figure 11). The mean absolute error difference

Frontiers in Marine Science 08 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

A B

FIGURE 7
(A) A speghettii plot of COAMPS forecast maximum wind speed (kt) and (B) COAMPS forecast maximum wind speed bias compared to IBTtraACS.

between WDA minus NDA and WDA minus ODA experiments wave height and mean wave period from all three experiments are
show the ODA experiment is as good as or better than the WDA larger than 0.5. But the mean wave direction correlation for all 3
assimilation. They both reduced the forecast error for all three wave experiments is lower than 0.5 with correlation of 0.34, 0.39, and 0.35
variables evaluated here compared to the NDA experiment. for the NDA, ODA, and WDA, respectively, A plausible cause of
However, the mean correlation of WDA is higher than ODA for lower wave direction correlation of WDA than ODA may be due to
significant wave height and absolute wave period but not for the MLM tendency to over broaden the 2D wave direction (Donelan
wave direction (Figure 12). The mean correlation of significant et al., 2015).

A B

C D

FIGURE 8
Maximum differences of COAMPS WDA minus NDA forecast validate at the Harvey’s first landfall time on 03 UTC 26 August, 2017. (A) 39 h forecast
from the 12 UTC 24 August cycle, (B) 27 h forecast from the 00 UTC 25 August cycle, (C) 15 h forecast from the 12 UTC 25 August cycle, and (D) 3
h forecast from the 00 UTC 26 August cycle. Color shadings are the wind speed (WSPD) differences. Blue contours represent significant wave height
(SIGH) differences with contour interval of 0.1 m. Solid blue lines are positive differences and dashed blue lines are negative differences. Black and
magenta arrowers depict the maximum differences of COAMPS WDA minus NDA forecast wind speed and direction, and current speed (CURR) and
direction, respectively. The maximum differences of wind & current vectors are shown on the title of each panel. Black-square line depicts Havey’s
best track from IBTrACS and gray- dot line is the WDA forecast track.

Frontiers in Marine Science 09 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

FIGURE 9
Time series of forecast mean (A) significant wave height (m) bias (dash lines) and RMSE (solid lines), and (B) mean absolute error differences between
the WDA minus NDA (red line) and WDA minus ODA (cyne line) experiments. The blue, cyane, and red lines in (A) are from the NDA, ODA, and WDA
experiments, respectively. The station and forecast cycles mean is computed at buoys 40002, 42019, 42020, and 42035 locations.

FIGURE 10
Time series of forecast mean (A) wave directional (°) bias (dash lines) and RMSE (solid lines), and (B) mean absolute error differences between the
WDA minus NDA (red line) and WDA minus ODA (cyne line) experiments. The blue, cyane, and red lines in (A) are from the NDA, ODA, and WDA
experiments, respectively. The station and forecast cycles mean is computed at buoys 40002, 42019, 42020, and 42035 locations.

Frontiers in Marine Science 10 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

FIGURE 11
Time series of forecast mean (A) absolute wave period (s) (dash lines) and RMSE (solid lines), and (B) mean absolute error differences between the
WDA minus NDA (red line) and WDA minus ODA (cyne line) experiments. The blue, cyane, and red lines in (A) are from the NDA, ODA, and WDA
experiments, respectively. The station and forecast cycles mean is computed at buoys 40002, 42019, 42020, and 42035 locations.

A B

FIGURE 12
The mean correlation of forecast and buoy observed time series for (A) significant wave height (m), (B) mean wave direction (°), and (C) mean
absolute wave period (s). The forecast cycle is from 12 UTC 25 August to 00 UTC 28 August, 2017.

Frontiers in Marine Science 11 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

4 Summary wave direction suggesting possible limitation of using the MLM

method to convert 1D wave spectral density observation from
The impact of 2D directional wave spectral density assimilation NDBC buoys to 2D directional wave spectral density. Donelan
for Hurricane Harvey forecasts is quantified using a fully coupled et al. (2015) shows MLM tends to broaden the wind direction
air-ocean-wave tropical cyclone model COAMPS. A 3DVAR wave compared to a wavelet method. To obtain a statistically significant
assimilation method is developed in COAMPS ocean data evaluation of wave buoy assimilation impact on Hurricane track
assimilation component NCODA that uses 1D wave spectral and intensity forecasts will require additional case studies to
density observations from NOAA NDBC wave buoys and satellite increase the sample size. However, it remains unclear how large a
altimeter significant wave height observation over GOM. The sample size is needed given fewer major hurricanes have made
conversion from 1D to 2D directional wave spectral density is landfall along the U.S. Gulf coastal zone.
achieved using the MLM method. The NDBC buoy wave Recent advancements in Hurricane airborne targeted 2D
assimilation impact on the COAMPS Hurricane Harvey intensity, directional wave spectral density observations such as from the
track, and wave forecasts sensitivity are quantified and validated Wide Swath Altimeter Radar (WSRA; Walsh et al., 2021),
using the IBTrACS and NDBC wave data for three air-ocean-wave Directional Wave Spectra Drifter (DWSD; Centurioni et al.,
coupled experiments with and without ocean or wave 2017), Surface Wave Instrument Float with Tracking (SWIFT;
data assimilation. Thomson et al., 2019), and Air-Launched Autonomous Micro
Results show that over the ocean both the ocean only (ODA) Observer (ALAMO; Sanabia and Jayne, 2020) float or satellite
and wave only (WDA) DA experiments improve the track forecast Synthetic Aperture Radar (SAR; Schuler et al., 2004) provide a
error for forecast lead times up to 4 days, compared to the no wave new opportunity to directly assimilate the 2D directional wave
or ocean DA (NDA) experiment,. The 0-4 day mean absolute track spectral density and use a shorter wave data assimilation window
error for NDA, ODA and WDA experiments are 56.1, 52.7, and for future hurricane wave research.
51.0 nm, respectively. The ocean DA or wave DA reduced the Improving the coastal landfall hurricane observation and
Harvey forecast absolute track bias by 3.4 and 5.1 nm, respectively. forecast continue to be the forefront of several U.S. community
All three experiments have a low mean 0-120 h forecast lead time efforts to strengthen the conduit between research and operation.
maximum wind speed errors ranging from ~ -2 to -18 m s-1. The These include the Coastal and Ocean Modeling Testbed program
ODA experiment has the highest mean maximum wind intensity from the Integrated Ocean Observing System (IOOS; Joyce et al.,
error over land. Compared to NDA and ODA experiments, WDA 2019), the Southeastern Universities Research Association Coastal
experiment’s wave data assimilation has the smallest mean 0-5 day Ocean Observing and Predicting program (SCOOP; Nichols and
intensity bias of -5.4 kts but has the highest mean intensity bias Wright, 2020), NOAA Forecast Improvement Program
while Harvey’s eye is over ocean. Between time periods of 12 UTC (Gopalakishnan et al., 2021), and in situ observation of the
24 August- 06 UTC 26 August and between 00 UTC 29 - 00UTC 30 tropical cyclone from the partnership of NOAA Advancing the
August, the mean intensity errors from NDA, ODA, and OWDA Prediction of Hurricanes Experiment (APHEX) program and Office
experiments are -9.0, -9.3, and -15.3 kts respectively. of the Naval Research Hurricane Rapid Intensification
As for the mean wave forecast error validations at four NDBC Departmental Research Initiative (Holbach et al., 2023). Further
buoy locations that are closest to the Harvey track, the mean research on wave assimilation methods of 2D directional spectral
significant wave height RMSE differences between ODA and wave density observations as well as forecast validation using a fully
WDA are less than 0.1 m. Compared to the 0-120 h COAMPS coupled air-ocean-wave tropical cyclone model such as COAMPS
forecast time series with buoys, the mean 5-day forecast cycle presented in this study are warranted to continue improving the
correlations between forecast and buoy observations show that coastal storm surge forecasts worldwide.
the ODA experiment has a slightly higher (0.01) correlation than
the WDA experiment. All three experiments have significant wave
height correlations greater than 0.5 with mean values of 0.62, 0.64, Data availability statement
and 0.63 for the NDA, ODA, and WDA experiments, respectively.
These COAMPS experiments also have mean correlations greater The raw data supporting the conclusions of this article will be
than 0.5 for the mean absolute wave period. The mean absolute made available by the authors, without undue reservation.
wave period correlations for the NDA, ODA, and WDA
experiments are 0.65, 0.66, and 0.67, respectively. However
COAMPS forecast correlations of mean wave direction are low (< Author contributions
0.5) for all three experiments. The mean wave direction correlations
for the NDA, ODA, and WDA experiments are 0.34, 0.39, and SC: Conceptualization, Data curation, Formal analysis, Funding
0.35, respectively. acquisition, Investigation, Methodology, Project administration,
Overall evaluation of wave DA impact on the fully air-ocean- Resources, Software, Supervision, Validation, Visualization,
wave coupled COAMPS Hurricane Harvey forecasts show that Writing – original draft, Writing – review & editing. JC: Data
wave assimilation improved Harvey’s track and wind intensity curation, Methodology, Software, Writing – review & editing. JV:
forecasts compared to the NDA and ODA experiments. However, Data curation, Methodology, Software, Writing – review & editing.
the wave DA experiment does not improve the correlation of mean JT: Formal analysis, Writing – review & editing.

Frontiers in Marine Science 12 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

Funding Conflict of interest

The author(s) declare financial support was received for the Author JC was employed by company Science Applications
research, authorship, and/or publication of this article. This International Corporation.
research is supported by the Office of Navy Research (ONR) The remaining authors declare that the research was conducted
grant N0001417WX01401. in the absence of any commercial or financial relationships that
could be construed as a potential conflict of interest.

Acknowledgments Publisher’s note

The computation resource is provided by the Department of All claims expressed in this article are solely those of the authors and
Defense (DoD) High Performance Computing Modernization do not necessarily represent those of their affiliated organizations, or
Program NRL-HPC 024 project. Dr. Tim Campbell is those of the publisher, the editors and the reviewers. Any product that
acknowledged for providing consultation on the selection of may be evaluated in this article, or claim that may be made by its
coupling extrapolation methods in COAMPS. manufacturer, is not guaranteed or endorsed by the publisher.

References
Allard, R. A., Campbell, T. J., Smith, T. S., Jensen, T. G., Chen, S., Cummings, J. A., vol. II. (Springer Berlin Heidelberg, Berlin, Heidelberg), 303–343. Available at: https://
et al. (2010). Validation test report for the coupled ocean atmospheric mesoscale link.springer.com/chapter/10.1007/978-3-642-35088-7_13.
prediction system version 5.0 (Washington DC, U.S.A.: Naval Research Laboratory), Daley, R., and Barker, E. (2001). NAVDAS: Formulation and diagnostics. Monthly
172. NRL/MR/7322-10-9283. Weather Rev. 129 (4), 869–883. doi: 10.1175/1520-0493(2001)129%3C0869:NFAD%
Allard, R. A., Smith, T., Jensen, T. G., Chu, P. Y., Rogers, E., Campbell, T., et al. (2012). 3E2.0.CO;2"https://doi.org/10.1175/1520-0493(2001)129<0869:NFAD>2.0.CO;2
Validation test report for the Coupled Ocean/Atmosphere Mesoscale Prediction System Donelan, M. A., Babanin, A., Sanina, E., and Chalikov, D. (2015). A comparison of
(COAMPS) version 5.0: ocean/wave component validation. (Washington DC, U.S.A.: methods for estimating directional spectra of surface waves. J. Geophys. Res. Oceans
Naval Research Laboratory), 91, NRL Memorandum Report NRL/MR/7320–12-9423. 120, 5040–5053. doi: 10.1002/2015JC010808
Bannister, R. N. (2017). A review of operational methods of variational and ensemble Doyle, J. D., Jin, Y., Hodur, R., Chen, S., Jin, H., Moskaitis, J., et al. (2012). “Real-time
variational data assimilation. Q. J. R. Meteorol. Soc. 143 (703), 607–633. doi: 10.1002/ tropical cyclone prediction using COAMPS-TC,” in Advances in geosciences, vol. 28 .
qj.2982. ISSN: 1477870X. Eds. C.-C. Wu and J. Gan (World Scientific Publishing Company, Singapore), 15–28.
Barker, E. H. (1980). Solving for temperature using unnaturally latticed hydrostatic doi: 10.1142/9789814405683_0002
equations. Mon. Wea. Rev. 108, 1260–1268. doi: 10.1175/1520-0493(1980)108,1260: Earle, M. D., Steele, K. E., and Wang, D. W. C. (1999). Use of advanced directional
SFTUUL.2.0.CO;2 wave spectra analysis methods. Ocean Eng. 26, 1421–1434. doi: 10.1016/S0029-8018
Barnes, S. L. (1964). A technique for maximizing details in numerical weather map (99)00010-4
analysis. J. Appl. Meteorology Climatology 3, pp.396–pp.409. doi: 10.1175/1520-0450 Fan, Y., Lin, S. J., Held, I. M., Yu, Z., and Tolman, H. L. (2012). Global ocean surface
(1964)003<0396:ATFMDI>2.0.CO;2 wave simulation using a coupled atmosphere–wave model. J. Climate 25, 6233–6252.
Benoit,M. (1994).Extensive comparisonofdirectionalwave analysismethodsfrom gauge array doi: 10.1175/JCLI-D-11-00621.1
data. In Proc. 2nd Int. Symp. on Ocean Wave Measurement and Analysis (ASCE), 740–754. Gopalakishnan, S., Upadhayay, S., Jung, Y., Marks, F., Tallapragada, V., Mehra, A.,
Benoit, M. (1992). Practical comparative performance survey of methods used for et al. (2021). 2020 HFIP R&D activities summary: Recent results and operational
estimating directional wave spectra from heave-pitch-roll data. Coast. Eng. 1992, 62–75. implementation. doi: 10.25923/718e-6232
doi: 10.1061/9780872629332.005 Hall, C., Jensen, R. E., and Wang, D. W. (2022). Performance evaluation of the newly
Byrne, D., Horsburgh, K., and Williams, J. (2023). Variational data assimilation of operational NDBC 2.1-m hull. J. Atmospheric Oceanic Technol. 39, 861–880.
sea surface height into a regional storm surge model: Benefits and limitations. J. Geo. doi: 10.1175/JTECH-D-21-0172.1
Res. Ocean 16 (1), 1–14. doi: 10.1080/1755876X.2021.1884405 Heming, J. T., Prates, F., Bender, M. A., Bowyer, R., Cangialosi, J., Caroff, P., et al.
Campbell, T., Allard R., R., Preller, R., Smedstad, L., Wallcraft, A., Chen, S., et al. (2019). Review of recent progress in tropical cyclone track forecasting and expression of
(2010). Integrated modeling of the battlespace environment. Computing Sci. Eng. 12, uncertainties. Trop. Cyclone Res. Rev. 8 (4), 181–218. doi: 10.1016/j.tcrr.2020.01.001
36–45. doi: 10.1109/MCSE.2010.78 Holbach, H. M., Bousquet, O., Bucci, L., Chang, P., Cione, J., Ditchek, S., et al. (2023).
Cangialosi, J. P., Blake, E., DeMaria, M., Penny, A., Latto, A., Rappaport, E., et al. Recent advancements in aircraft and in situ observations of tropical cyclones. Trop.
(2020). Recent progress in tropical cyclone intensity forecasting at the National Cyclone Res. Rev. 12 (2), 81–99. doi: 10.1016/j.tcrr.2023.06.001
Hurricane Center. Weather Forecasting 35, 1913–1922. doi: 10.1175/WAF-D-20- Holland, G. J., Belanger, J. I., and Fritz, A. (2010). A revised model for radial profiles of
0059.1 hurricane winds. Monthly weather Rev. 138, 4393–4401. doi: 10.1175/2010MWR3317.1
Centurioni, L., Braasch, L., Di Lauro, E., Contestabile, P., De Leo, F., Casotti, R., et al. Houghton, I. A., Penny, S. G., Hegermiller, C., Cesaretti, M., Teicheira, C., and Smit,
(2017). A new strategic wave measurement station off Naples port main breakwater. P. B. (2023). Ensemble-based data assimilation of significant wave height from Sofar
Coast. Eng. Proc. 1, 36. doi: 10.9753/icce.v35.waves.36 Spotters and satellite altimeters with a global operational wave model. Ocean Model.
Charnock, H. (1955). Wind stress on a water surface. Quart. J. R. Meteor. Soc 81, 183, 102200. doi: 10.1016/j.ocemod.2023.102200
639–640. doi: 10.1002/qj.49708135027 Janssen, P. A. (1991). Quasi-linear theory of wind-wave generation applied to wave
Chen, S., Cummings, J., Doyle, J. D., Hodur, R. M., Holt, T., Liou, C. S., et al. (2003). forecasting. J. Phys. oceanography 21, 1631–1642. doi: 10.1175/1520-0485(1991)021%
COAMPS version 3 model description. (Washington DC, U.S.A.: Naval Research 3C1631:QLTOWW%3E2.0.CO;2
Laboratory), 143, NRL Report: NRL/PU/7500-04-448. Joyce, B. R., Gonzalez-Lopez, J., van der Westhuysen, A. J., Yang, D., Pringle, W. J.,
Chen, S., Cummings, J. A., Schmidt, J. M., Sanabia, E. R., and Jayne, S. R. (2017). Westerink, J. J., et al. (2019). US IOOS coastal and ocean modeling testbed: Hurricane-
Targeted ocean sampling guidance for tropical cyclones. J. Geo. Res. Ocean 122 (5), induced winds, waves, and surge for deep ocean, reef-fringed islands in the Caribbean.
3505–3518. doi: 10.1002/2017JC012727 J. Geophysical Research: Oceans 124, 2876–2907. doi: 10.1029/2018JC014687
Chen, S., Gaberšek, S., Doyle, J. D., Cook, J., Chu, P., Allard, P. R. A., et al. (2015). Knapp, K. R., Diamond, H. J., Kossin, J. P., Kruk, M. C., and Schreck, C. J. (2018).
Validation test report for the Coupled Ocean/Atmosphere Mesoscale Prediction System International best track archive for climate stewardship (IBTrACS) project, version 4.
(COAMPS) version 5.0: Air/Wave component validation (Washington DC, U.S.A.: [indicate subset used] (NOAA National Centers for Environmental Information).
Naval Research Laboratory), 68. NRL/MR/7322-10-9526. doi: 10.25921/82ty-9e16
Cummings, J. A., and Smedstad, O. M. (2013). “Variational data assimilation for the Komaromi, W. A., Reinecke, P. A., Doyle, J. D., and Moskaitis, J. R. (2021). The naval
global ocean,” in Data assimilation for atmospheric, oceanic and hydrologic applications, research laboratory’s coupled ocean–atmosphere mesoscale prediction system-tropical

Frontiers in Marine Science 13 frontiersin.org

Chen et al. 10.3389/fmars.2024.1332883

cyclone ensemble (COAMPS-TC ensemble). Weather Forecasting 36 (2), 499–517. Saffir, H. S. (1973). Hurricane wind and storm surge Vol. 65 (Virginia, U.S.A: The
doi: 10.1175/WAF-D-20-0038.1 Military Engineer), 4–5.
Komen, G. J., Cavaleri, L., Donelan, M., Hasselmann, K., Hasselmann, S., Janssen, P., Sanabia, E. R., Barrett, B., Black, P. G., Chen, S., and Cummings, J. A. (2013). Real-
et al. (1994). Dynamics and modelling of ocean waves (Cambridge, UK: Cambridge Time upper-ocean temperature observations from aircraft during operational
University Press), 554. doi: 10.1017/CBO9780511628955 hurricane reconnaissance missions: AXBT demonstration project year one results.
Kuik, A. J., van Vledder, G., and Holthuijsen, L. H. (1988). A method for the routine Weather Forecasting 28, 1404–1422. doi: 10.1175/WAF-D-12-00107.1
analysis of pitch-and-roll buoy wave data. J. Phys. Oceanogr. 18, 1020–1034. Sanabia, E. R., and Jayne, S. R. (2020). Ocean observations under two major
doi: 10.1175/1520-0485(1988)018,1020:AMFTRA.2.0.CO;2 hurricanes: Evolution of the response across the storm wakes. AGU Adv. 1,
Lionello, P., Gunther, H., and Janssen, P. A. (1992). Assimilation of altimeter data in e2019AV000161. doi: 10.1029/2019AV000161
a global third-generation wave model. J. Geophys. Res. Ocean 97 (C9), 14453–14474. Saulter, A. N., Bunney, C., King, R. R., and Waters, J. (2020). An application of
doi: 10.1029/92jc01055 NEMOVAR for regional wave model data assimilation. Front. Mar. Sci. 7. doi: 10.3389/
Longuet-Higgins, M. S., Cartwright, D. E., and Smith, N. D. (1963). “Observations of the fmars.2020.579834
directional spectrum of sea waves using the motions of a floating buoy,” in Ocean wave Schuler, D. L., Lee, J. S., Kasilingam, D., and Pottier, E. (2004). Measurement of ocean
spectra (New Jersey, U.S.A.: Prentice-Hall), 111–136. Available at: http://www.jstor.org/ surface slopes and wave spectra using polarimetric SAR image data. Remote Sens.
stable/2414167. Environ. 91, 198–211. doi: 10.1016/j.rse.2004.03.008
Lygre, A., and Krogstad, H. E. (1986). Maximum entropy estimation of the Seemanth, M., Remya, P. G., Bhowmick, S. A., Sharma, R., Nair, T. B., Kumar, R.,
directional distribution in ocean wave spectra. J. Phys. Oceanography 16, 2052–2060. et al. (2021). Implementation of altimeter data assimilation on a regional wave
doi: 10.1175/1520-0485(1986)016<2052:MEEOTD>2.0.CO;2 forecasting system and its impact on wave and swell surge forecast in the Indian
Marchok, T. (2021). Important factors in the tracking of tropical cyclones in Ocean. Ocean Eng. 237, 109585. doi: 10.1016/j.oceaneng.2021.109585
operational models. J. Appl. Meteorology Climatology 60, 1265–1284. doi: 10.1175/ Simpson, R. H. (1974). The hurricane disaster—Potential scale. Weatherwise 27,
JAMC-D-20-0175.1 169–186. doi: 10.1080/00431672.1974.9931702
Martin, P. J., Edwards, K. L., Veeramony, J., Blain, C. A., and Campbell, T. J. (2020). Smit, P. B., Houghton, I. A., Jordanova, K., Portwood, T., Shapiro, E., Clark, D., et al.
VTR for implementation of wave-ocean coupling in COAMPS in both the nearshore and (2021). Assimilation of significant wave height from distributed ocean wave sensors.
offshore ocean. (Washington DC, U.S.A.: Naval Research Laboratory), 67. NRL Ocean Model. 159, 101738. doi: 10.1016/j.ocemod.2020.101738
Memorandum Report NRL/MR/7322–20-10176. Smith, T. A., Chen, S., Campbell, T., Martin, P., Rogers, W. E., Gaberšek, S., et al.
Martin, P. J., Rogers, E., Allard, R. A., Dykes, J. D., and Hogan, P. J. (2013). Tests of (2013). Ocean-wave coupled modeling in COAMPS-TC: A study of hurricane ivan,
parameterized langmuir-circulation mixing in the ocean’s surface mixed layer. (2004). Ocean Modelling 69, 181–194. doi: 10.1016/j.ocemod.2013.06.003
(Washington DC, U.S.A.: Naval Research Laboratory), 47. NRL Memorandum
Theurich, G., DeLuca, C., Campbell, T., Liu, F., Saint, K., Vertenstein, M., et al. (2016). The
Report NRL/MR/7320–13-9444, Naval Research Laboratory, SSC, MS 39529.
earth system prediction suite: Toward a coordinated US modeling capability. Bull. Am.
Metzger, E. J. H. E., Hurlburt, X., Xu, J. F., Shriver, A. L., Gordon, J., Sprintall, R. D., Meteorological Soc. 97, 1229–1247. doi: 10.1175/BAMS-D-14-00164.1
et al. (2010). Simulated and observed circulation in the Indonesian Seas: 1/12 global
HYCOM and the INSTANT observations. Dyn. Atmos. Ocean. 50, 275-300. doi: Thomson, J., Moulton, M., de Klerk, A., Talbert, J., Guerra, M., Kastner, S., et al.
10.1016/j.dynatmoce.2010.04.002 (2019). “A new version of the SWIFT platform for waves, currents, and turbulence in
the ocean surface layer,” in 2019 IEEE/OES twelfth current, waves and turbulence
National Hurricane Center. (2017). Costliest U.S. tropical cyclone tables update. measurement (CWTM) (New Jersey, U.S.A.: IEEE), 1–7. doi: 10.1109/
Available at: https://www.nhc.noaa.gov/news/UpdatedCostliest.pdf. CWTM43797.2019.8955299
NDBC. (2009). Handbook of automated data quality control. Available at: https://www. Tran, N., Vandemark, D., Labroue, S., Feng, H., Chapron, B., Tolman, H. L., et al.
ndbc.noaa.gov/publications/NDBCHandbookofAutomatedDataQualityControl2009.pdf. (2010). Sea state biasin altimeter sea level estimates determined by combining wave
Nichols, C. R., and Wright, L. D. (2020). The evolution and outcomes of a model and satellite data. J. Geophys. Res. 115, C03020. doi: 10.1029/2009JC005534
collaborative testbed for predicting coastal threats. J. Mar. Sci. Eng. 8, 612. Valle-Levinson, A., Olabarrieta, M., and Heilman, L. (2020). Compound flooding in
doi: 10.3390/jmse8080612 Houston-Galveston Bay during Hurricane Harvey. Sci. Total Environ. 747, 141272.
Nowotarski, C. J., Spotts, J., Edwards, R., Overpeck, S., and Woodall, G. R. (2021). doi: 10.1016/j.scitotenv.2020.141272
Tornadoes in hurricane harvey. Weather Forecasting 36, 1589–1609. doi: 10.1175/ Veeramony, J., Walker, D., and Hsu, L. (2010). A variational data assimilation system
WAF-D-20-0196.1 for nearshore applications of SWAN. Ocean Model. 35, 206–214. doi: 10.1016/
Oltman-Shay, J., and Guza, R. T. (1984). A data-adaptive ocean wave directional- j.ocemod.2010.07.008
spectrum estimator for pitch and roll type measurements. J. Phys. Oceanography 14, Voorrips, A. C., Makin, V. K., and Hasselmann, S. (1997). Assimilation of wave
1800–1810. doi: 10.1175/1520-0485(1984)014<1800:ADAOWD>2.0.CO;2 spectra from pitch-and-roll buoys in a north sea wave model. J. Geophys. Res. C: Oceans
Orzech, M., Veeramony, J., and Flampouris, S. (2014). Optimizing spectral wave 102 (C3), 5829–5849. doi: 10.1029/96JC03242. ISSN: 01480227
estimates with adjoint-based sensitivity maps. Ocean Dynamics 64, 487–505. Walsh, E. J., Fairall, C. W., and PopStefanija, I. (2021). In the eye of the storm. J. Phys.
doi: 10.1007/s10236-014-0700-2 Oceanography 51, 1835–1842. doi: 10.1175/JPO-D-20-0219.1
Potter, H., DiMarco, S. F., and Knap, A. H. (2019). Tropical cyclone heat potential Wittmann, P. A., and Cummings, J. A. (2004). “November. Assimilation of altimeter
and the rapid intensification of Hurricane Harvey in the Texas Bight. J. Geophysical wave measurements into WAVEWATCH III,” in 8th international workshop on wave
Research: Oceans 124, 2440–2451. doi: 10.1029/2018JC014776 hindcasting and forecasting(Virginia, U.S.A.: U.S. Defense Technical Information
Riley, R., Hall, C., Stewart, R., DiNapoli, S., and Wang, D. W. (2019). “NDBC OWL wave Center), 14–19. Available at: https://apps.dtic.mil/sti/pdfs/ADA436548.pdf.
system development,” in 2019 IEEE/OES twelfth current, waves and turbulence measurement Wurman, J., and Kosiba, K. (2018). The role of small-scale vortices in enhancing
(CWTM) (Virginia, U.S.A.: IEEE), 1–6. doi: 10.1109/CWTM43797.2019.8955248 surface winds and damage in Hurricane Harvey, (2017). Monthly Weather Rev. 146,
713–722. doi: 10.1175/MWR-D-17-0327.1
Romero-Arteaga, A., Ruiz de Alegrı́a-Arzaburu, A., and Esquivel-Trava, B. (2022).
Spatial variability of surface waves and nearshore currents induced by hurricane harvey WW3DG, T.W.I.D.G (2019). User manual and system documentation of Wavewatch III
along the southern texas coast. J. Mar. Sci. Eng. 10, 1722. doi: 10.3390/jmse10111722 version 6.07. Available at: https://polar.ncep.noaa.gov/mmab/papers/tn276/MMAB_276.pdf.

Frontiers in Marine Science 14 frontiersin.org