Evaluating High-Frequency radar data assimilation impact in coastal ocean operational modelling

The impact of the assimilation of HFR (High-Frequency Radar) observations in a high-resolution regional model is evaluated, focusing on the improvement of the mesoscale dynamics. The study area is the Ibiza Channel, located in the Western Mediterranean Sea. The resulting fields are tested against trajectories from 13 drifters. Six different assimilation experiments are compared to a control run (no assimilation). The experiments consists in assimilating (i) Sea surface temperature, sea level 5 anomaly and Argo profiles (generic observation dataset); the generic observation dataset plus (ii) HFR total velocities and (iii) HFR radial velocities. Moreover, for each dataset two different initialization methods are assessed: a) restarting directly from the analysis after the assimilation or b) using an intermediate initialization step applying a strong nudging towards the analysis fields. The experiments assimilating generic observations plus HFR total velocities with the direct restart provides the best results, improving ::::::: reducing : by 53% the average separation distance between drifters and virtual particles after the 10 first 48 hours of simulation in comparison to the control run. When using the nudging initialization step, the best results are found when assimilating HFR radial velocities, with a reduction of the mean separation distance by around 48%. Results show ::: that ::: the ::::::::: integration :: of ::::: HFR :::::::::: observations :: in ::: the ::: the :::: data :::::::::: assimilation ::::::: system ::::::: enhances ::: the ::::::::: prediction :: of :::::: surface ::::::: currents :::::: inside the capability of the Ensemble Optimal Interpolation data-assimilative system to correct surface currents not only inside but also beyond the HFR coverage area :::: area :::::: covered ::: by :::: both ::::::::: antennas, ::::: while ::: not ::::::::: degrading ::: the ::::::::: correction :::::::: achieved ::::: thanks ::: to 15 :: the ::::::::::: assimilation :: of :::::: generic :::: data ::::::: sources :::::: beyond :: it. The assimilation of radial observations benefits from the smoothing effect associated with the application of the intermediate nudging step.

after analysis, which progressively applies data assimilation increments, to preserve appropriate dynamical balances. This data assimilation approach resulted in an increase of the correlation between model and observations from 0.42 to 0.78. 60 More recently, hourly reconstructed total currents have also been employed using both sequential (Ren et al., 2016;Paduan and Shulman, 2004) and variational data-assimilation schemes such as 4D-Var (Zhang et al., 2010;Wilkin and Hunter, 2013;Yu et al., 2016). However, depending on the model set-up and the oceanic processes of interest, the use of hourly data may not be the most appropriate, as for instance in Kerry et al. (2016), where radial speeds and angles are spatially averaged onto the model grid and a 24 h boxcar-averaging filter is used to remove tides and inertial oscillations that are not resolved by the 65 model. Kerry et al. (2018) show that among all the assimilated observations, HFR were the ones which had the larger impact on the currents and the transport in the Eastern Australian Current.
The use of hourly data in sequential data assimilation schemes is not straightforward, due to the analysis frequency which is generally larger than one hour. An option is to use an extended state vector as in Barth et al. (2008), who employed an ensemble based Kalman Filter (KF) method using hourly radial observations in the West Florida Shelf. For the initialization 70 Barth et al. (2008) implemented a spatial filter and averaged the ensemble fields in an attempt to remove spurious variability before it is introduced into the model. Barth et al. (2011) and Marmain et al. (2014) employed a similar approach, using all radial hourly observations available during the assimilation window and an extended state vector to correct the wind forcing fields and boundary conditions respectively in a similar way to variational methods. While Barth et al. (2011) showed that the correction had a positive impact on the reconstructed winds and the SST in the German Bight, Marmain et al. (2014) 75 found an improvement in surface currents in the North-Western Mediterranean Sea, although with some degradation on the density fields and under surface currents. Stanev et al. (2015) also used hourly radial observations to correct tidal currents in the German Bight. In an operational context and based on a spatio-temporal optimal interpolation (STIO), Stanev et al. (2016) demonstrated that their system had a good skill to correct currents even beyond the HFR covered area.
A comparison of the impact of both time-filtered and unfiltered HFR currents (with respect to a model with and without 80 tides) was done in Shulman and Paduan (2009), showing that the sub-tidal period velocity simulations were similarly improved through the assimilation of either low-pass-filtered surface currents or instantaneous (hourly) surface currents. More recently, Vandenbulcke et al. (2017) using different KF schemes, with an extended state vector, assimilated hourly radial velocities to correct inertial oscillations in a regional model of the Ligurian Sea. They show an important effect on the correction of inertial oscillations during the first 12 hours, when considering all hourly observations in a 48-hour time-window instead of using only 85 the corresponding to one single hour.
In the present study we aim at evaluating the impact in coastal ocean operational modelling of the assimilation of both HFR total and radial velocities, also exploiting different initialization methods after analysis. Our focus is on the correction of mesoscale structures and larger scale circulation, rather than inertial oscillations or tidal currents.
The study area is the Ibiza Channel (IC) (, : Fig. 1), which is the passage between the oriental coast of Spain mainland and 90 the island of Ibiza. It is a crucial area for understanding mixing and transport processes in the Northwestern Mediterranean Sea. Two different water masses interact in the IC: (i) a relatively salty water that has already recirculated in the Western Mediterranean flowing southward along the shelf as the Northern Current, and (ii) a branch of the Modified Atlantic waters transporting fresher waters originally entering through the strait of Gibraltar and flowing northward (Pinot et al., 1994(Pinot et al., , 1995 on its easternmost part. The dynamics, and the ecological and economical importance of the area have raised a specific interest 95 in understanding the relevant ocean processes (Heslop et al., 2012;Balbín et al., 2014;Pinot et al., 2002;Hernández-Carrasco et al., 2018;Vargas-Yáñez et al., 2021). The analysis of repeated observations along a glider endurance line in the Ibiza Channel has revealed a high variability of meridional transports over time scales of days to weeks (Heslop et al., 2012). This high variability due to the interaction of multiple processes with different water masses over a complex topography make the operational forecasting particularly challenging. The anthropogenic pressure in the region makes it necessary to develop 100 accurate tools for Search and Rescue, oil spill forecasting or larval dispersion to efficiently respond to emergencies and protect ecosystems.
Since 2012, the Balearic Island Coastal Observing and forecasting System (SOCIB, Tintoré et al. (2013)) operates a CODAR HFR system that monitors the IC with two antennas measuring hourly surface currents (Tintoré et al., 2020). Lana et al. (2016) validated the IC HFR observations against current-meter, ADCP and surface Lagrangian drifters, showing a good agreement 105 and the absence of significant mean error (hereafter referred as bias). A joint analysis of HFR observations and surface winds in terms of Empirical Orthogonal Functions (EOF) demonstrated that the surface current variability was mainly driven by local winds and mesoscale circulation.
Seven one-month period simulations have been generated to investigate the data assimilation performance of HFR raw radial observations compared to reconstructed totals currents. We have employed three different datasets, and for each of them, 110 two different initialization methods after analysis. Additionally, a free-run simulation without assimilation has been used as control run. An exhaustive assessment has been performed following both Eulerian and Lagrangian approaches, including an independent set of 13 drifters deployed in the area.
The paper is structured as follows: Section 2 describes the data and methods employed, including the DA system and the description of the experiments. Results are presented in Section 3. Finally, the discussion of the results and the conclusions are 115 presented in Sections 4 and 5.
2 Data and Methods 2.1 High Frequency :::::::::::::: High-Frequency Radar The SOCIB HFR system consists in two CODAR SeaSonde stations of the islands of Ibiza and Formentera (named GALF and FORM, respectively), covering the eastern side of the IC. It operates ::: has ::::::: operated : since June 2012, providing real-time 120 high-resolution observations of surface currents (Tintoré et al., 2020;Lana et al., 2015Lana et al., , 2016. Each HFR station emits at a central frequency of 13.5 MHz and a bandwidth of 90 kHz, reaching ranges up to 85 km. Emitted electromagnetic waves are back-scattered by surface waves of exactly half the HFR wavelength. Radial velocities (velocities toward or away from the antenna) are derived from the Doppler shift due to the difference between ideal and measured Bragg frequency (Barrick, 2008).
At the specified operating frequency, measurement depth is approximately 0.9 m (Stewart and Joy, 1974). Radial observations 125 provide the velocity along a bearing, calculated from radio signals backscattered from the ocean surface. Hourly radial velocity maps from both stations are systematically quality controlled and the total velocity vectors are reconstructed by combining the radial velocities with overlapping coverage, on a regular 3 × 3 km grid. Each grid point observation is computed using a unweighted least-square fitting (UWLS) (Lipa and Barrick, 1983), considering all radial observations within a 6km radius.
Total reconstructed observations have a range up to 65 km off the antenna, compared to the 85 km that radials can reach.

Regional model configuration
The Western Mediterranean OPerational system (WMOP, Juza et al. (2016); Mourre et al. (2018)) is a high-resolution regional configuration of the ROMS (Regional Ocean Modelling System) model (Shchepetkin and McWilliams, 2005) for the western 150 Mediterranean Sea. The spatial coverage spans from Gibraltar strait on the West to the Sardinia Channel on the East (6ºW-9ºE, 35ºN-44.5ºN, see Fig. 1) with a horizontal resolution around 2 km and 32 vertical sigma levels (resulting in a vertical resolution between 1 and 2m at the surface). The WMOP system is used to produce daily forecasts of the regional ocean circulation, which is used for a wide range of applications including search-and-rescue and analysis of plastic, parasite or larval dispersion for instance (Calò et al., 2018;Ruiz-Orejón et al., 2019;Cabanellas-Reboredo et al., 2019;Compa et al., 2020;Torrado et al., 2021;  scale Copernicus Forecasting System (CMEMS MED-MFC), with a 1/16º horizontal resolution (Simoncelli S., 2017). The atmospheric forcing is provided every 3 hours at 1/20 • resolution by the Spanish Meteorological Agency (AEMET) through the HIRLAM model (Undén et al., 2002). These fields are used to compute surface turbulent and momentum fluxes through bulk formulae. Atmospheric pressure forcing is neglected to avoid SSH high-frequency variability issues. Inflows from the six major rivers in the region are considered as point sources, using daily climatological values. Tides are not considered in the depict an average southward current west of 0.8E. This current is deviated towards the south-east of 38.7N, and the flow is directed northward in the eastern side of the coverage area, close to Ibiza and Formentera coast. The control run represents this overall pattern, but with a significant overestimation of the mean velocities and a spatial mismatch of the eastward deviation of the flow (this deviation occurs too much to the east in the model).

175
The assimilation scheme employed here is the multimodel local :::: local ::::::::::: multi-model Ensemble Optimal Interpolation (EnOI) employed in Hernández-Lasheras and . It is a form of the EnOI, which has been a widely used scheme , since it represents a cost-effective alternative compared with more complex methods as the Ensemble Kalman Filter or the 4Dvar (Oke For each analysis, the state vector x = (T i,j,k , S i,j,k , u i,j,k , v i,j,k , SSH i,j ) T , contains the model trajectory, i.e., the prognostic model variables at all wet gridpoints i, j, k.
During the analysis step, the state vector x a is updated according to Eq. (1), where x f is the background model state vector,

185
H is the linear observation operator projecting the model state onto the observation space andK is the Kalman gain estimated from the sample covariances (Eq. 2). y is the vector of observations. MatricesP f and R are the error covariance matrices of the model and the observations, respectively. Sea. An independent analysis is performed for each water column of the model domain, considering only the observations within the localization radius.
The state vector equivalents of HFR radials are obtained using the following equation: where u x and u y are the model surface velocity components interpolated at the observation point, and α denotes the angle (anti-clockwise towards the east) pointing from an antenna station to a certain location.

220
A 3-day assimilation cycle is applied with different time windows for each source of observation as explained in the following section. In each analysis (day n) the daily average field is employed as background and two different initialization approaches ( Fig. 3) have been applied to restart the model after the analysis. Sequential assimilation methods are affected by initialization issues, as primitive equation models are sensitive to discontinuous changes in their model fields (Oke et al., 2002). These discontinuities may introduce artificial waves or structures in the model that affect the quality of predictions. Different strategies 225 have been proposed to address this problem (Sandery et al., 2011;Yan et al., 2014).
The : In ::: the :::: first :::::::: approach, ::: the : simulation for day n+1 restarts directly from the results of the analysis. The second approach, which will be referred to as nudging consists in running again the day n applying a very strong nudging (time scale of one day) towards the temperature, salinity and SSH fields provided by the analysis. Notice that the nudging is not applied to the velocity fields. These are adjusted by the model itself according to its dynamics. This procedure reduces the model corrections but

Simulations
Seven simulations of WMOP are used to investigate the impact of both HFR observation ::::::::::: observations and initialization methods (Table 1). The period selected for the simulation experiments covers one month, from September 20 th to October 20 th 235 2014, assimilating different sets of observations every 3 days. During this period a total of 13 satellite-tracked surface drifters (Tintoré et al., 2014) were deployed in the area covered by the HFR and used as independent data for validating the numerical experiments ( Fig. 1). We adopted the operational prediction setup of WMOP, considering only observations before the analysis date. Notice that a "retrospective analysis" framework considering a time window centered on the analysis date could slightly improve the results presented in this paper. However, since our objective is to implement this method for daily predictions, the 240 operational setup has been selected. Satellite SLA (sea level anomalies), SST (sea surface temperature) and T-S (temperature and salinity) Argo profiles, defined as the Generic Observing sources (GO), are assimilated in all these simulations. The SLA consists in along-track L3 multi-satellite reprocessed observations provided by CMEMS. We consider a 3-day window for SLA observations. The SST comes from a L4-GHRSST foundation SST product distributed by JPL-MUR (NASA/JPL, 2015). The foundation SST is the temperature free of diurnal temperature variability, corresponding to the temperature of the surface just 245 before the daily heating by the sun. Since the model daily average contains the signature of the diurnal cycle, this effect needs to be accounted for in the representativity error. This is approximated by computing the variance of the difference between the model SST field at 8 a.m. and the daily average field used as background for each of the grid points. The ultra-high 1 km resolution gridded fields have been smoothed and interpolated to a 10 km grid to limit the number of observations, while still representing the effective scale that this SST product can resolve (Chin et al., 2017). For the T-S Argo profiles we have 250 considered a 5-day time-window, which corresponds to the nominal time of Mediterranean Argo floats cycles. For each profile, values are binned vertically to obtain a single value for each model grid cell. The variance of the data within a bin is used as the vertical representation error, which is added to the horizontal one, assumed to be 0.25 • C 2 and 0.05 2 for temperature and salinity measurements, respectively.
A control run (CR) without data assimilation has been used as benchmark to assess the performance of the different assim-255 ilation experiments. We called GN R the simulation in which we only assimilated GO. Additionally, four other simulations assimilating HFR data together with GO have been generated. In all four cases we assimilate daily averages to remove the impact of inertial oscillations and tides, which are not the focus of this study. Daily averaged fields from the model are used as background for the analysis. T OT simulation employs HFR totals, computed as described before. We called RAD the simulation assimilating all possible daily mean radial observations.

260
Data assimilation experiments have been repeated using both types of initialization for every dataset. Our analysis will first evaluate the impact and trade-offs of the different kind of HFR observations when using the direct restart from the analysis procedure. Then, the impact of the nudging initialization method will be specifically discussed.  previously assimilated. However, they can not be considered as fully independent since the data employed for the validation come from the same platforms that provide the assimilated measurements. 275 Taylor diagrams (Taylor, 2001) are presented here for the evaluation of the simulations. They illustrate the correspondence between model and observations in terms of correlation coefficient, centered root mean square difference (CRMSD) and standard deviation. However, note that the diagram does not represent the mean error between the observations and the model, which has been examined separately. The magnitude of the SST mean error decreases from -0.29 • C to -0.14 • C , representing in all simulations less than the 14% of the total RMSD 1 . The mean error between the CR and the Argo profiles is 0.4 • C and 280 -0.13 for temperature and salinity respectively, representing less than 8% of the RMSD in both cases.
The use of DA results in a significant improvement of both the SLA and SST fields, as shown in Fig. 4. For both data sources the symbols corresponding to each simulation assimilating data overlap, meaning that the validation metrics are very similar for all of them. For the SLA it leads to a significant increase in the correlation, with values from 0.42 to around 0.70, and a 30% reduction in the CRMSD for all the experiments with DA. Notice that the model SLA presents a relatively large mean 285 error, with a value of around -0.07m. Discrepancies are common when comparing models to altimetry due to differences in the mean sea level. This mean error, which persists after DA, accounts for the difference between the mean dynamic topography of the model and observations. This way, the reduction in the RMSD is mostly due to reductions of the CRMSD, which can be observed in the diagrams.
Concerning the SST, we obtain a similar error reduction in terms of centered RMSD, of the order of 30% closer to observa-290 tions when using DA. An increase in correlation is also obtained, from 0.82 to around 0.92 when compared with the CR. We do not observe a significant difference between the simulations using different datasets.
Similar conclusions are obtained when examining the Taylor diagrams focusing on Argo temperature and salinity profiles (Fig. 5). Although the CR simulation shows a very high correlation with observations (0.88 and 0.95 for temperature and salinity respectively), this correlation is further increased for the experiments with DA. A CRMSD reduction of more than 35% The impact of the assimilation on the different fields has been also evaluated considering only observations surrounding the IC area, leading to similar results.

Eulerian assessment of the impact of DA on surface currents 300
To evaluate the DA capabilities in improving : to :::::::: improve the representation of surface currents, we performed an Eulerian analysis in the HFR coverage area. WMOP surface daily mean velocities are compared against HFR totals daily mean fields.
The total observations are derived from the radial data, as described in section 2.1. The ::: We ::::::: compute ::: the daily mean field is only computed :::: only at those points that provide more than 50% of hourly data . The model is then interpolated ::: and :::: then ::::::::: interpolate :: the :::::: model : to HFR observation grid points. As for the SLA, SST and Argo TS profiles, the validation can not be considered 305 here as fully independent , since we are using :::: fully :::::::::: independent ::::: since ::: we ::: use the same observing platform. However, the data used for validation at a given time have not yet been assimilated in the model.
The :: We :::: first ::::::: analyze ::: the : performance in terms of surface currents is first analyzed by using the Taylor diagrams for the velocity components (Fig.6). Observing the :::: The zonal velocity component , it experiences a strong correction with the assimilation of GO. Specifically, the CRMSD suffers a reduction of : is ::::::: reduced :: by : 28% while the correlation increases towards the 310 Figure 5. Same as Figure 4 for Argo temperature (left) and salinity (right) profiles.
observation from 0.28 to 0.44. This performance is further improved by the two experiments using HFR data, with more than a 40% reduction in CRMSD. While T OT experiment exhibits the largest error reduction, RAD provides the best correlation with observations (0.7), compared to 0.63 obtained by T OT .. : Considering the meridional velocity component , we can observe how GN R has a lower correlation and higher CRMSD than the CR. Here, the use of HFR observations is necessary to reduce the difference between model and observations. The 315 correlation slightly increases :::: with ::: the ::::::::: asimilation ::: of :::: HFR, with the best results obtained for T OT (0.47)and RAD (0.43).
DIVAnd is a n-dimensional variational analysis method which is used here to reconstruct hourly 2D vectorial fields from radial observations. It was shown to improve the reconstruction compared to the Open-boundary Modal Analysis (Kaplan and 350 Lekien, 2007).  better reproduced in the model, as previously described by (Révelard et al., 2021). .
The assimilation of HFR data along with GO further increases the skill score. The improvement is particularly significant inside the HFR domain, where most of the trajectories have positive SS. T OT has the best results among the model experiments, with a mean value of 0.41 inside the coverage area, which in comparison to : is ::::: better :::: than : RAD (0.36)is a clear improvement.
The average separation distance is computed according to Equation 6, where n drif = 13 is the number of drifters, n part = 1000 is the number of particles and x d and x v are the positions of the real drifter and the corresponding virtual particle respectively ( Figure 10). For each 5 day trajectory, the mean separation distance is first computed averaging over the number of drifters, providing a single distance as a function of time d(t) for the 13 drifters (Eq. 6). Then, the four values of d(t), one 385 for each of the four simulations starting in consecutive days are averaged. :::::: The mean distance between virtual and real drifters is significantly reduced when DA is applied. The assimilation of GO efficiently helps to reduce the mean separation distance, with a reduction of 31% after 48 hours compared to CR (18.9 versus 27.2 km). Consistent with the previous analysis, the assimilation of HFR total observations along with the GO further increases 390 the performance, leading to the lowest mean separation distance (12.8 km), with a 53% reduction compared to the CR. The use of radial observations also leads to a high reduction of the mean separation distance (48%), which is reduced to 14.3 km after 48hr.

Impact of the nudging restart strategy
Overall, the results in the whole domain comparing to satellite and Argo observations are similar to those obtained for the simulations restarting directly from the analysis. The improvement is slightly lower due to the nudging step, but all data assimilative simulations provide comparable metrics. The reduction of the RMSD compared to the CR is around 8% for the 400 SLA, while for the SST is reduced around 30%. Considering Argo profiles, the reduction of the RMSD is of 35% for all simulations, both for temperature and salinity. Table 4 presents the bias and normalized RMSD for total velocity ::::: RMSD ::: for ::: the :::::: model :::::: surface :::::: current ::::: speed :::: and ::: the ::::: zonal ::: and ::::::::: meridional ::::::::::: components. This has to be compared with Table 2, which shows the results for the previous simulations ::::::::: simulations :::::::: restarting ::::: from ::: the ::::::: analysis. We can observe a slight improvement for the GN R − N simulation when using the 405 nudging initialization in comparison to restarting directly from the analysis, with a reduction of both the bias and the RMSD.
While for the RAD this initialization method also helps to reduce the bias compared to direct restart from the analysis ::: for ::::: RAD , this is not the case when using total observations. The Lagrangian assessment confirms these results, reflecting the usefulness of HFR data to correct surface currents using this initialization method even when the nudging is only performed towards the SSH and TS fields. The SS for the GN R − N 410 simulation ::::: (Table :: 5) : increases significantly inside the coverage area while decreasing outside, with an average value of 0.39, larger than the value of 0.34 obtained with the other approach.

Discussion
The assimilation of high-resolution HFR surface currents observations in a reduced part of the modelling domain could have We have used DIVAnd reconstructed fields as a benchmark for our Lagrangian validation. These hourly fields properly represent the inertial oscillations, compared to other gap-filling techniques (Barth et al., 2021), and we consider it as the best possible high resolution observation ::::::::::: representation : of the surface currents in the area which allows the simulation of Lagrangian 440 trajectories. It is very positive that the skill scores obtained for the HFR DA experiments are very close to that obtained by DIVAnd. While DIVAnd outperforms the capabilities of the WMOP DA system inside the coverage of both HFR antennas, it is the opposite outside this region, demonstrating ::::: which ::::::::::: demonstrates the capacity of the model to improve the represention of the currents beyond the HFR coverage area. The assimilation of GO, in particular SLA, constrains the geostrophic circulation, leading to a better representation of the Balearic current and an increase of the SS in that area. The importance of this constrain While the mean SS for the DIVAnd-derived trajectories inside the area is 0.53, it drops to 0.29 outside of it, being significantly lower than all model-derived trajectories. This behaviour is consistent with Barth et al. (2021) :::::: results, which show that the DIVAnd reconstructed fields outside the area covered by both HFR antennas are much less reliable. Our results demonstrate the utility of dynamical models assimilating high-resolution observations as good alternatives to data-driven short-term forecasting 450 methods, due to their capacity to extend the correction beyond the observation coverage area. They also show the importance of combining HFR and altimeters observations which help to constrain the geostrophic circulation ::: over :: a ::::: wider :::: area.
Two different initialization strategies have been evaluated. While restarting directly from the analysis may introduce some high frequency :::::::::::: high-frequency and spurious waves or instabilities in the system due to inconsistencies between the corrected fields and the model equations, it considers an initial state which is closer to observations. On the other hand, the nudging 455 strategy provides a more conservative framework, in which the model dynamics are better respected but with the drawback that some of the correction achieved with the observations may be lost. In general :::::: Overall, both approaches show similar results leading to a reduction of the RMSD over the whole domain. The :: As :: in ::: the :::: case :: of ::: the ::::: direct :::::: restart :::: from ::: the :::::::: analysis, ::: the use of the nudging strategy also leads to an improvement of the predictions of surface currents when adding ::::::::: assimilating : HFR observations, compared to the simulation that only uses generic data sources. It is important to point out that, in our case, 460 nudging is only applied towards the temperature, salinity and sea surface height fields, but not towards the velocity fields : , :: to :::: avoid :::::: model :::::::::: instabilities. Therefore, the assimilation of the surface currents enables to correct the density fields, which in turn improves the surface velocities due to the model initial adjustments.
The nudging strategy limits the possible shocks and anomalous gradients that may be generated in the analysisand : , :: so :::: that :: the :::::::: solution remains closer to the physical balances. We found that it was not optimal for surface currents prediction when 465 using HFR total velocities but a better choice for radial data. This is probably due to the fact that reconstructed total velocities are already smoothed out through a pre-processing step contrarily to ::: the :::: case :: of : radial data, which are more noisy and then directly benefit from the smoothing effect of the nudging approach. The nudging strategy appears to be a good solution for operational purposes, when the ocurrence :::::::: occurrence : of noisy data tends to be more frequent. It may also be a good choice for systems depending on operational data sources for which HFR antennas, for instance, may not work during certain periods or 470 satellite and Argo data may not be available on time. It could also be less sensitive to potential errors in data in cases where near real-time observations could have large :: be ::::::: affected :: by ::::::::: significant errors.
Data availability.
-Simulations are archived on the SOCIB server and are available upon request to info@socib.es.