Search

On 12 November 2019, an exceptional flood event occurred in Venice, second only to the one that had occurred on 4 November 1966. The event caused the flooding of about 90% of the streets in the historic centre and the sea level reached a value of 189 cm compared to the local mean-sea-level datum. Subsequent analysis of the meteorological event highlighted the contributions at different temporal and spatial scales. A sub-synoptic cyclone, centred in the Tyrrhenian Sea, caused a strong Sirocco wind along the entire Adriatic basin, with a fairly typical atmospheric configuration. However, embedded in the first cyclone, a second meso-beta scale cyclone developed and moved in the north-westward direction over the Adriatic Sea along the Italian coast. This cyclone had a speed of about 12 m/s, very close to the speed of the shallow water waves for the depth of the northern Adriatic basin. The perturbation then triggered a Proudman resonance, as confirmed by the numerical simulations, and caused a meteotsunami-like wave that affected the north-western coasts of the Adriatic Sea. Through model simulations, we have estimated that the mesoscale cyclone contributed about 40 cm, of which about 40% may be attributed to the air-pressure forcing, amplified through the Proudman resonance, and the rest to the wind forcing influencing both the open Adriatic Sea and the shallow Venetian lagoon. Finally, we have also analysed the propagation and transformation of the perturbation upon its entrance into the Venetian lagoon. This work is part of the COST action CA19109 MEDCYCLONES (European Network for Mediterranean Cyclones in weather and climate) and the Interreg Italy-Croatia STREAM project (Strategic development of flood management, project ID 10249186).

In this research we present a modelling study of storm surges influenced by pre‐existing seiches in the Adriatic Sea. The natural oscillation modes ‐ seiches ‐ of the Adriatic Sea can be easily excited by wind, due to the particular marine bathymetry and coastal orography. Moreover, such oscillations can interact with newer wind‐induced storm surges or with other sea level components. This paper first considers modelling issues in the seiche reproduction, using a simplified approach, and then analyses the reproduction with the model of two extreme storm surge events, influenced by pre‐existing seiches. In order to investigate the impact of assimilating sea level data, these events are simulated without and with a data assimilation system, based on an Ensemble Kalman Filter. Results show that, while the formulation of the bottom shear stress does not influence the seiche period, the seiche decay time is correctly estimated with a hybrid linear‐quadratic formulation. We found also that non‐linear interactions of bottom currents, caused by different sea level components, can modify the decay time. However, the two events are poorly reproduced by the model, even with a correct modelling set‐up, because of wind deficiencies. Therefore we considered the issue and found that the assimilation of tide‐gauge residual sea level has a strong positive impact, lasting several days, despite the errors in the atmospheric forcing. This is mainly due to the persistence of the seiche oscillations in the Adriatic Sea.

Full comprehension of the dynamics of hazardous sea levels is indispensable for assessing and managing coastal flood risk, especially under a changing climate. The 12 November 2019 devastating flood in the historical city of Venice (Italy) stimulated new investigations of the coastal flooding problem from different perspectives and timescales. Here Venice is used as a paradigm for coastal flood risk, due to the complexity of its flood dynamics facing those of many other locations worldwide. Spectral decomposition was applied to the long-term 1872–2019 sea-level time series in order to investigate the relative importance of different drivers of coastal flooding and their temporal changes. Moreover, a multivariate analysis via copulas provided statistical models indispensable for correctly understanding and reproducing the interactions between the variables at play. While storm surges are the main drivers of the most extreme events, tides and long-term forcings associated with planetary atmospheric waves and seasonal to inter-annual oscillations are predominant in determining recurrent nuisance flooding. The non-stationary analysis revealed a positive trend in the intensity of the non-tidal contribution to extreme sea levels in the last three decades, which, along with relative sea-level rise, contributed to an increase in the frequency of floods in Venice.

The North Ionian Gyre (NIG) displays prominent inversions on decadal scales. We investigate the role of internal forcing, induced by changes of the horizontal pressure gradient due to the varying density of the Adriatic Deep Water (AdDW), that spreads into the deep layers of the Northern Ionian Sea. In turn, the AdDW density fluctuates according to the circulation of the NIG through a feedback mechanism named Bimodal Oscillating System. We set up laboratory experiments with a two-layer ambient fluid in a circular rotating tank, where densities of 1000/1015 kg m−3 characterise the upper/lower layer, respectively. From the potential vorticity evolution during the dense water outflow from a marginal sea, we analyse the response of the open-sea circulation to the along-slope dense water flow. In addition, we show some features of the cyclonic/anticyclonic eddies that form in the upper layer over the slope area. We illustrate the outcome of the experiments of varying density and varying discharge rates associated with the dense water injection. When the density is high, 1020 kg m−3, and the discharge is large, the kinetic energy of the mean flow is stronger than the eddy kinetic energy. On the other hand, when the density is smaller, 1010 kg m−3, and the discharge is reduced, vortices are more energetic than the mean flow, that is, the eddy kinetic energy is larger than the kinetic energy of the mean flow. In general, over the slope, following the onset of the dense water injection, the cyclonic vorticity associated with a current shear develops in the upper layer. The vorticity behaves in a two-layer fashion, thus becoming anticyclonic in the lower layer of the slope area. Concurrently, over the deep flat-bottom portion of the basin, a large-scale anticyclonic gyre forms in the upper layer extending partly toward a sloping rim. Density record shows the rise of the pycnocline due to the dense water sinking toward the flat-bottom portion of the tank. We show that the rate of increase of the anticyclonic potential vorticity is proportional to the rate of the rise of the interface, namely, to the rate of decrease of the upper layer thickness (i.e., the upper layer squeezing). The comparison of laboratory experiments with the Ionian Sea is made for a situation when the sudden switch from the cyclonic to the anticyclonic basin-wide circulation took place following the extremely dense Adriatic water overflow after the harsh winter in 2012. We show how similar are the temporal evolution and the vertical structure in both laboratory and oceanic conditions. The demonstrated similarity further supports the assertion that the wind-stress curl over the Ionian Sea is not of paramount importance in generating basin-wide circulation inversions, as compared to the internal forcing.

Extreme sea levels at the coast result from the combination of astronomical tides with atmospherically forced fluctuations at multiple time scales. Seiches, river floods, waves, inter-annual and inter-decadal dynamics and relative sea-level rise can also contribute to the total sea level. While tides are usually well described and predicted, the effect of the different atmospheric contributions to the sea level and their trends are still not well understood. Meso-scale atmospheric disturbances, synoptic-scale phenomena and planetary atmospheric waves (PAW) act at different temporal and spatial scales and thus generate sea-level disturbances at different frequencies. In this study, we analyze the 1872-2019 sea-level time series in Venice (northern Adriatic Sea, Italy) to investigate the relative role of the different driving factors in the extreme sea levels distribution. The adopted approach consists in 1) isolating the different contributions to the sea level by applying least-squares fitting and Fourier decomposition; 2) performing a multivariate statistical analysis which enables the dependencies among driving factors and their joint probability of occurrence to be described; 3) analyzing temporal changes in extreme sea levels and extrapolating possible future tendencies. The results highlight the fact that the most extreme sea levels are mainly dominated by the non-tidal residual, while the tide plays a secondary role. The non-tidal residual of the extreme sea levels is attributed mostly to PAW surge and storm surge, with the latter component becoming dominant for the most extreme events. The results of temporal evolution analysis confirm previous studies according to which the relative sea-level rise is the major driver of the increase in the frequency of floods in Venice over the last century. However, also long term variability in the storm activity impacted the frequency and intensity of extreme sea levels and have contributed to an increase of floods in Venice during the fall and winter months of the last three decades.

Strong upwelling driven by the NNW winds was detected off the eastern middle Adriatic coast in May 2017. High resolution CTD data revealed thermocline doming by about 20 m at approximately 20 km from the coast. Main characteristics of the upwelling event are reproduced in the realistic ROMS model simulation. Adriatic scale ROMS model having 2.5 km horizontal resolution, forced by the air-sea fluxes calculated using surface fields from operational weather forecast model ALADIN-HR (Tudor et al., 2013; Termonia et al., 2018), river discharges, tides and water mass exchange through the Strait of Otranto, reproduces cold water dome and two-layer offshore flow in accordance with CTD and shipborne ADCP measurements. Significant improvement in the upwelling simulations is obtained using increased drag coefficient. The location of upwelling is correctly modelled, although with somewhat lower upper layer temperatures if compared with measurements. Moreover, the surface cyclonic circulation indicated by ADCP measurements along the cross-Adriatic transect is also evident in the model results. In order to improve understanding of the upwelling mechanism, several schematized numerical experiments are conducted. Wind fields from dynamical adaptation (Zagar and Rakovec, 1999; Ivatek-Sahdan and Tudor, 2004) of ALADIN-HR8 (8 km horizontal grid spacing) wind forecast to 2 km grid, are decomposed by the Natural Helmholtz-Hodge Decomposition (HHD) into divergence-free (incompressible), rotation-free (irrotational), and harmonic (translational) component (Bhatia et al., 2014). The components thus obtained and their combinations are used for calculation of the wind stress instead of the total wind field. Simulations with decomposed wind stress are conducted in the Adriatic domains with both flat bottom and realistic topography. Schematized simulations reveal that the positive rotational wind component is responsible for the rising of thermocline through Ekman pumping and it is more pronounced in the flat bottom basin. In the simulations with divergent wind component, the thermocline doming disappears and only coastal upwelling is reproduced. Additional idealised simulations with homogeneous NW wind stress are performed assuming both two-layer and uniform initial density field.

The North Ionian Gyre (NIG) displays prominent inversions on decadal scales. We investigate the role of internal forcing induced by changes in the horizontal pressure gradient due to the varying density of Adriatic Deep Water (AdDW), which spreads into the deep layers of the northern Ionian Sea. In turn, the AdDW density fluctuates according to the circulation of the NIG through a feedback mechanism known as the bimodal oscillating system. We set up laboratory experiments with a two-layer ambient fluid in a circular rotating tank, where densities of 1000 and 1015 kg m−3 characterize the upper and lower layers, respectively. From the potential vorticity evolution during the dense-water outflow from a marginal sea, we analyze the response of the open-sea circulation to the along-slope dense-water flow. In addition, we show some features of the cyclonic and anticyclonic eddies that form in the upper layer over the slope area. We illustrate the outcome of the experiments of varying density and varying discharge rates associated with dense-water injection. When the density is high (1020 kg m−3) and the discharge is large, the kinetic energy of the mean flow is stronger than the eddy kinetic energy. Conversely, when the density is lower (1010 kg m−3) and the discharge is reduced, vortices are more energetic than the mean flow – that is, the eddy kinetic energy is larger than the kinetic energy of the mean flow. In general, over the slope, following the onset of dense-water injection, the cyclonic vorticity associated with current shear develops in the upper layer. The vorticity behaves in a two-layer fashion, thereby becoming anticyclonic in the lower layer of the slope area. Concurrently, over the deep flat-bottom portion of the basin, a large-scale anticyclonic gyre forms in the upper layer extending partly toward a sloping rim. The density record shows the rise of the pycnocline due to the dense-water sinking toward the flat-bottom portion of the tank. We show that the rate of increase in the anticyclonic potential vorticity is proportional to the rate of the rise of the interface, namely to the rate of decrease in the upper-layer thickness (i.e., the upper-layer squeezing). The comparison of laboratory experiments with the Ionian Sea is made for a situation when the sudden switch from cyclonic to anticyclonic basin-wide circulation took place following extremely dense Adriatic water overflow after the harsh winter in 2012. We show how similar the temporal evolution and the vertical structure are in both laboratory and oceanic conditions. The demonstrated similarity further supports the assertion that the wind-stress curl over the Ionian Sea is not of paramount importance in generating basin-wide circulation inversions compared with the internal forcing.

This paper reviews the state of the art in storm surge forecasting and its particular application in the northern Adriatic Sea. The city of Venice already depends on operational storm surge forecasting systems to warn the population and economy of imminent flood threats, as well as help to protect the extensive cultural heritage. This will be more important in the future, with the new mobile barriers called MOSE (MOdulo Sperimentale Elettromeccanico, Experimental Electromechanical Module) that will be completed by 2021. The barriers will depend on accurate storm surge forecasting to control their operation. In this paper, the physics behind the flooding of Venice is discussed, and the state of the art of storm surge forecasting in Europe is reviewed. The challenges for the surge forecasting systems are analyzed, especially in view of uncertainty. This includes consideration of selected historic extreme events that were particularly difficult to forecast. Four potential improvements are identified: (1) improve meteorological forecasts, (2) develop ensemble forecasting, (3) assimilation of water level measurements and (4) develop a multimodel approach.

Floods in the Venice city centre result from the superposition of several factors: astronomical tides; seiches; and atmospherically forced fluctuations, which include storm surges, meteotsunamis, and surges caused by atmospheric planetary waves. All these factors can contribute to positive water height anomalies individually and can increase the probability of extreme events when they act constructively. The largest extreme water heights are mostly caused by the storm surges produced by the sirocco winds, leading to a characteristic seasonal cycle, with the largest and most frequent events occurring from November to March. Storm surges can be produced by cyclones whose centres are located either north or south of the Alps. Historically, the most intense events have been produced by cyclogenesis in the western Mediterranean, to the west of the main cyclogenetic area of the Mediterranean region in the Gulf of Genoa. Only a small fraction of the inter-annual variability in extreme water heights is described by fluctuations in the dominant patterns of atmospheric circulation variability over the Euro-Atlantic sector. Therefore, decadal fluctuations in water height extremes remain largely unexplained. In particular, the effect of the 11-year solar cycle does not appear to be steadily present if more than 100 years of observations are considered. The historic increase in the frequency of floods since the mid-19th century is explained by relative mean sea level rise. Analogously, future regional relative mean sea level rise will be the most important driver of increasing duration and intensity of Venice floods through this century, overcompensating for the small projected decrease in marine storminess. The future increase in extreme water heights covers a wide range, largely reflecting the highly uncertain mass contributions to future mean sea level rise from the melting of Antarctica and Greenland ice sheets, especially towards the end of the century. For a high-emission scenario (RCP8.5), the magnitude of 1-in-100-year water height values at the northern Adriatic coast is projected to increase by 26–35 cm by 2050 and by 53–171 cm by 2100 with respect to the present value and is subject to continued increase thereafter. For a moderate-emission scenario (RCP4.5), these values are 12–17 cm by 2050 and 24–56 cm by 2100. Local subsidence (which is not included in these estimates) will further contribute to the future increase in extreme water heights. This analysis shows the need for adaptive long-term planning of coastal defences using flexible solutions that are appropriate across the large range of plausible future water height extremes.

Meteotsunamis (localized ocean waves with periods similar to those of tsunamis but caused by meteorological phenomena) have been observed around the globe and are frequently observed in the Adriatic Sea. Numerous studies have focused on numerical modeling of meteotsunamis, but only a few have modeled flooding and drying of the coastal areas that play an important role in risk assessment. In this study, we model four historic meteotsunami events (Vela Luka Bay, June 21, 1978 ; Široka Bay, August 22, 2007 ; Mali Lošinj Bay August 15, 2008 ; Stari Grad Bay February 19, 2010) that occurred in the Adriatic Sea, using ADICRC, a flooding and drying capable ocean numerical model. Comparison of those results with similar simulations that do not use the flooding and drying algorithm was made to determine differences in modeled wave height. Three of the modeled events (Vela Luka, Široka and Mali Lošinj) are more accurately depicted if including the flooding and drying algorithm, suggesting that extreme events can be more realistically modeled than with the more commonly used cut-off depth (i.e., specifying minimum depth larger than expected maximum wave height). Modeling results for the fourth event (Stari Grad) confirm a previous assumption that flooding occurred due to the superposition of a storm surge and a meteotsunami, rather than a meteotsunami alone.

Floods in the Venice city centre result from the superposition of several factors: astronomical tides, seiches and atmospherically forced fluctuations, which include storm surges, meteotsunamis, and surges caused by planetary waves. All these factors can contribute to positive sea-level anomalies individually and can also result in extreme sea-level events when they act constructively. The largest extreme sea level events have been mostly caused by storm surges produced by the Sirocco winds. This leads to a characteristic seasonal cycle, with the largest and most frequent events occurring from November to March. Storm surges can be produced by cyclones whose centers are located either north or south of the Alps. The most intense historical events have been produced by cyclogenesis in the western Mediterranean, to the west of the main cyclogenetic area of the Mediterranean region in the Gulf of Genoa. Only a small fraction of the interannual variability of extreme sea levels is described by fluctuations in the dominant patterns of atmospheric circulation variability over the Euro-Atlantic sector. Therefore, decadal fluctuations of sea-level extremes remain largely unexplained. In particular, the effect of the 11-year solar cycle appears to be small, non-stationary or masked by other factors. The historic increase in the frequency of extreme sea levels since the mid 19th Century is explained by relative sea level rise, with no long term trend in the intensity of the atmospheric forcing. Analogously, future regional relative mean sea level rise will be the most important driver of increasing duration and intensity of Venice floods through this century, overwhelming the small decrease in marine storminess projected during the 21 century. Consequently, the future increase of extreme sea levels covers a large range, partly reflecting the highly uncertain mass contributions to future mean sea level rise from the melting of Antarctica and Greenland ice-sheets, especially towards the end of the century. In conclusion, for a high emission scenario the magnitude of 1-in-100 year sea level events at the North Adriatic coast is projected to increase up to 65 % and 160 % in 2050 and 2100, respectively, with respect to the present value, and subject to continued increase thereafter. Local subsidence can further contribute to the future increase of extreme sea levels. This analysis shows the need for adaptive planning of coastal defenses with solutions that can be adopted to face the large range of plausible future sea-level extremes.

The aim of this work was to study Chrystal and Proudman resonances in a simple closed basin and to explore and compare how well the two resonant mechanisms are reproduced with different, nowadays widely used, numerical ocean models. The test case was based on air pressure disturbances of two commonly used shapes (a sinusoidal and a boxcar), having various wave lengths, and propagating at different speeds. Our test domain was a closed rectangular basin, 300 km long with a uniform depth of 50 m, with the theoretical analytical solution available for benchmark. In total, 2250 simulations were performed for each of the three different numerical models: ADCIRC, SCHISM and ROMS. During each of the simulations, we recorded water level anomalies and computed the integral of the energy density spectrum for a number of points distributed along the basin. We have successfully documented the transition from Proudman to Chrystal resonance that occurs for a sinusoidal air pressure disturbance having a wavelength between one and two basin lengths. An inter-model comparison of the results shows that different models represent the two resonant phenomena in a slightly different way. For Chrystal resonance, all the models showed similar behavior ; however, ADCIRC model providing slightly higher values of the mean resonant period than the other two models. In the case of Proudman resonance, the most consistent results, closest to the analytical solution, were obtained using ROMS model, which reproduced the mean resonant speed equal to 22.00 m/s— i.e., close to the theoretical value of 22.15 m/s. ADCIRC and SCHISM models showed small deviations from that value, with the mean speed being slightly lower— 21.97 m/s (ADCIRC) and 21.93 m/s (SCHISM). The findings may seem small but could play an important role when resonance is a crucial process producing enhancing effects by two orders of magnitude (i.e., meteotsunamis).

The paper provides a comprehensive review of all aspects of Adriatic Sea level research covered by the literature. It discusses changes occurring over millennial timescales and documented by a variety of natural and man-made proxies and post- glacial rebound models ; mean sea level changes occurring over centennial to annual timescales and measured by modern instruments ; and daily and higher-frequency changes (with periods ranging from minutes to a day) that are contributing to sea level extremes and are relevant for present-day flooding of coastal areas. Special tribute is paid to the historic sea level studies that shaped modern sea level research in the Adriatic, followed by a discussion of existing in situ and remote sensing observing systems operating in the Adriatic area, operational forecasting systems for Adriatic storm surges, as well as warning systems for tsunamis and meteotsunamis. Projections and predictions of sea level and related hazards are also included in the review. Based on this review, open issues and research gaps in the Adriatic Sea level studies are identified, as well as the additional research efforts needed to fill the gaps. The Adriatic Sea, thus, remains a laboratory for coastal sea level studies for semi-enclosed, coastal and marginal seas in the world ocean.

This work assesses the impact of assimilating the sea level data, with an Ensemble Kalman Filter, on storm surge and seiche modelling. The study area is the Adriatic Sea, where seiches are always present after a storm surge, and often overlap on a new storm surge with a possible amplification of the total sea level. Due to errors in the wind and pressure forcing, the forecast of such extreme events is rather challenging in the Adriatic Sea, and a wrong reproduction of such pre-existing seiches reflects on a bad sea-level forecast. Here we show, by two case studies, that the assimilation of sea-level data along the coasts of the Adriatic basin is able to correct the initial state of the hydrodynamic model. Since the initial state is particularly important in the case of pre-existing seiches, the reduction of the initial error propagates several days into the forecast. The two cases here presented were between the most extreme storm surge events in the last years and they both included the pre-existing seiches. The initial forecast was very poor, due to the fact that the wind was underestimated. The assimilation of 3-day long hourly sea level data at eleven stations distributed along the Adriatic coasts produces a better forecast in both cases. Moreover, the ensemble spread allows the uncertainty of the forecast to be estimated, even if the estimate should be calibrated over time in order to be more reliable.

On 12 November 2019, an exceptional flood event occurred in Venice, second only to the one that occurred on 4 November 1966. The maximum recorded sea level value of 189 cm above local datum resulted in the flooding of more than 85% of the pedestrian surface of the historical city. Moreover, with four extremely high tides since 11 November 2019, this has been the worst week for flooding in Venice ever since 1872, when official statistics were first produced. The event that struck Venice and the northern Adriatic Sea on 12 November 2019, although having certain conditions seemingly typical of the events that cause exceptional high waters, also had some peculiar characteristics not observed before and therefore it requires an in-depth analysis. Several factors made this event exceptional: an in-phase timing of the peak of the storm surge and the astronomical tide; an anomalously high monthly mean sea level in the Adriatic Sea induced by a steady low-pressure and wind systems over the Mediterranean Sea associated with large-scale low-frequency atmospheric dynamics; a deep low-pressure system over the central-southern Tyrrhenian Sea that generated strong sirocco (south-easterly) winds along the main axis of the Adriatic Sea pushing the waters towards north; a fast-moving local depression - and the associated wind perturbation - travelling in the north-westward direction along the Italian coast that may have forced long ocean waves (e.g., edge wave); and very strong winds (100 km h-1 on average, with gusts reaching 110 km h-1) over the Lagoon of Venice which led to a further rise in water levels and damage to the historic city. In this study, a large set of available observations and the high-resolution numerical simulations are used to quantify the influence of these drivers on the peak flood event and to investigate the peculiar weather and sea conditions over the Mediterranean Sea during the Venice floods of November 2019.

Under the emerging features of interannual-to-decadal ocean variability, the periodical reversals of the North Ionian Gyre (NIG), driven mostly by the mechanism named Adriatic-Ionian Bimodal Oscillating System (BiOS), are known as impacting on marine physics and biogeochemistry and potentially influencing short-term regional climate predictability in the Eastern Mediterranean. Whilst it has been suggested that local wind forcing cannot explain such variability, aspects of the alternative hypothesis indicating that NIG reversals mainly arises from an internal ocean feedback mechanism alone remain largely debated. Here we demonstrate, using the results of physical experiments, performed in the world’s largest rotating tank and numerical simulations, that the main observed feature of BiOS, i.e., the switch of polarity of the near-surface circulation in the NIG, can be induced by a mere injection of dense water on a sloping bottom. Hence, BiOS is a truly oceanic mode of variability and abrupt polarity changes in circulation can arise solely from extreme dense water formation events.

Regional Ocean Modeling System (ROMS) is used to reproduce and analyse upwelling detected in the middle Adriatic Sea during May 2017. The ROMS domain covers the entire Adriatic, with a rectangular grid having horizontal resolution of 2.5 km and 22 unequally spaced s levels along the vertical. Surface momentum, heat and water fluxes in the ROMS simulations are calculated using atmospheric fields from the operational ALADIN model (Tudor et al., 2013; Termonia et al., 2018), having a horizontal resolution of 8 km for scalar fields (air pressure, air temperature, relative humidity, cloudiness, precipitation, and shortwave radiation) and 2 km resolution for wind fields. The ROMS model, in addition to the atmospheric agents, is forced by river inflows, tides and water mass exchange through the Strait of Otranto. Along the Adriatic coast, 41 rivers are discharging into the sea and their climatological flow rates (Raicich, 1994) are used in the simulations. Tidal forcing is applied on the open boundary taking into account seven tidal harmonics (M2, S2, N2, K2, K1, O1 and P1) crucial for the Adriatic dynamics. The open boundary conditions for the free surface, temperature, salinity, and velocity are taken from the wider Adriatic AREG model operationally run under the Adriatic Forecasting System (Oddo et al., 2006). Baseline ROMS simulation is run for the period from 1 August 2016 to 31 December 2018 and its quality is assessed with available CTD and HF radar measurements, satellite sea surface temperatures (SST) and data collected during May 2017 and June 2018 cruises by the yo-yo CTD profiler and shipborne ADCP. Numerical experiments focus on May 2017 when upwelling induced by prevailing NW wind was recorded. Baseline experiment qualitatively reproduces the upwelling but several sensitivity experiments are needed to increase agreement between model and measurements. Various intensities of horizontal viscosity and diffusivity and drag coefficient are tested in sensitivity studies. Moreover, a third order upwind advection scheme is tested as is the behaviour of solar shortwave radiation along the water column. Significant improvement in the model results is obtained using increased drag coefficient. Circulation pattern recorded by shipborne ADCP with inflowing currents in the first 10 km from the eastern middle Adriatic coast and wind-controlled two-layer flow further offshore is also reproduced by the ROMS model. Upwelling was clearly documented in the SST satellite images of 28 and 29 May 2017 by patches of cold water close to the eastern coast. Similar structure is reproduced by ROMS, although the modelled SST underestimates the measured values by approximately 1 °C. The area of upwelling is correctly located as is the cyclonic circulation indicated by ADCP measurements along the transect.

In this study, the fine-scale responses of a stratified oligotrophic karstic lake (Kozjak Lake, Plitvice Lakes, Croatia; the lake fetch is 2.3 km, and the maximum depth is 46 m) to atmospheric forcing on the lake surface are investigated. Lake temperatures measured at a resolution of 2 min at 15 depths ranging from 0.2 to 43 m, which were observed during the 6 July–5 November 2018 period, were analyzed. The results show thermocline deepening from 10 m at the beginning of the observation period to 16 m at the end of the observation period, where the latter depth corresponds to approximately one-third of the lake depth. The pycnocline followed the same pattern, except that the deepening occurred throughout the entire period approximately 1 m above the thermocline. On average, thermocline deepening was 3–4 cm d−1, while the maximum deepening (12.5 cm d−1) coincided with the occurrence of internal seiches. Furthermore, the results indicate three different types of forcings on the lake surface; two of these forcings have diurnal periodicity – (1) continuous heat fluxes and (2) occasional periodic stronger winds – whereas forcing (3) corresponds to occasional nonperiodic stronger winds with steady along-basin directions. Continuous heat fluxes (1) produced forced diurnal oscillations in the lake temperature within the first 5 m of the lake throughout the entire observation period. Noncontinuous periodic stronger winds (2) resulted in occasional forced diurnal oscillations in the lake temperatures at depths from approximately 7 to 20 m. Occasional strong and steady along-basin winds (3) triggered both baroclinic internal seiches with a principal period of 8.0 h and barotropic surface seiches with a principal period of 9 min. Lake currents produced by the surface seiches under realistic-topography conditions generated baroclinic oscillations of the thermocline region (at depths from 9 to 17 m) with periods corresponding to the period of surface seiches (≈ 9 min), which, to the best of our knowledge, has not been reported in previous lake studies.

On 12 November 2019, an exceptional flood event took place in Venice, second only to the one that occurred on 4 November 1966. Moreover, with four extremely high tides since 11 November 2019, this was the worst week for flooding in Venice since the beginning of sea level records (1872). The event that struck Venice and the northern Adriatic Sea on 12 November 2019, although having certain conditions seemingly typical of the events causing exceptional high waters, had some peculiar characteristics not observed before, which deserved an in-depth analysis. Several factors made this event exceptional: the in-phase timing between the peak of the storm surge and the astronomical tide; a deep low-pressure cyclone over the central-southern Tyrrhenian Sea that generated strong Sirocco (south-easterly) winds along the main axis of the Adriatic Sea, pushing waters to the north; a fast-moving local depression – and the associated wind perturbation – travelling in the north-westward direction over the Adriatic Sea along the Italian coast, generating a meteotsunami; very strong winds (28 m s − 1 on average with 31 m s − 1 gusts) over the Lagoon of Venice, which led to a rise in water levels and damages to the historic city; and an anomalously high monthly mean sea level in the Adriatic Sea, induced by a standing low-pressure and wind systems over the Mediterranean Sea, that was associated with large-scale low-frequency atmospheric dynamics. In this study, the large set of available observations and high-resolution numerical simulations have been used to quantify the contribution of the mentioned drivers to the peak of the flood event and to investigate the peculiar weather and sea conditions over the Mediterranean Sea during the Venice floods of November 2019.

In the recent past, two intensive episodes of high sea level occurred in the Adriatic, which affected particularly the eastern coast of the basin. The events happened on 1 December 2008 and on 1 November 2012 when the eastern-coast station Bakar recorded the two highest water levels in its operational history. Interestingly, the episodes were of milder intensity in Venice, situated at the opposite coast. The question is: why these events were so pronounced along the eastern shore and was there specific meteorological background favouring stronger flooding of the eastern Adriatic coast? The study was carried out using hourly sea-level series along both sides of the basin and ECMWF reanalysis of air pressure and wind fields. Sea level was attributed to four processes: low-frequency variability, tide, pre-existing basin-wide seiche and storm-surge activity and respective contributions to high sea-level formations were evaluated. The effects of meteorological forcings over the basin were analysed by calculating along- and across-basin wind stress with respect to local bathymetry. The two marine storms were outcome of fine interplay between the four mentioned processes. During the 2008 flood, storm-surge activity was dominant while the other components reinforced the maximum. In the 2012 episode, storm surge was intensive at most stations, but relative timing of all the components was more complex. Results also revealed one common feature that was crucial for both floods being stronger along the eastern Adriatic coast. Specifically, presence of Sirocco-wind shear induced higher-than-usual water level rise along the eastern coast while transverse component of wind had a relatively small effect. This means that, at least in these two extreme episodes, transverse sea- level slope in the Adriatic was more sensitive to shear of longitudinal component of wind than to wind’s direction.

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Geodesy in Croatia, 2015–2018 . . . . . . . . . . . . . . . . . . . . . . . . . . 173 Geomagnetism and aeronomy in Croatia, 2015–2018 . . . . . . . . ... 181 Hydrology and physical limnology in Croatia, 2015–2018 . . . . . ... 185 Meteorology in Croatia, 2015–2018 . . . . . . . . . . . . . . . . . . . . . . . . 195 Physical oceanography in Croatia, 2015–2018 . . . . . . . . . . . . . . . . 203 Seismology in Croatia, 2015–2018 . . . . . . . . . . . . . . . . . . . . . . . . . 217

Meteotsunamis, sea level oscillations of meteorological origin with periods similar to those of tsunamis, can, in extreme cases, reach significant wave heights and cause large damage to property and even a loss of life. In this study we have taken a historic meteotsunami in the Adriatic Sea with a wave height of 6 m and attempted to improve upon previous numerical simulations in order to model it as accurately as possible. The great Adriatic meteotsunami of 21 June 1978 was most pronounced in Vela Luka Bay on Korcula Island where sea level oscillations alternately flooded the town and dried out the harbour. Previous data analysis and numerical modelling have shown that the cause of the flood was a travelling air pressure disturbance that caused Proudman resonance in the open sea, with the waves being further amplified by the basin's funnel shape, reflection at the bay head and constructive superposition with the waves coming in from the open sea. To model the sea level oscillations we used a finite element model with a flooding and drying option included, with the entire Adriatic Sea as the modelling domain. A synthetic air pressure disturbance propagating from the Italian coast and over the island at varying speeds and directions was used as the only forcing field in the model. This new setup resulted in a successful reproduction of harbour flooding and drying and an increase in modelled wave heights.

Three variants of the semiempirical method for sea level projection are considered. They differ in assuming that the response of sea level to temperature forcing is equilibrium, inertial, or a combination of the two. All variants produce a successful regression of the temperature and sea level data, albeit with controlling parameters that differ among the cases. The related response times vary considerably, with a realistic value (~50 yr) obtained only if both the equilibrium and the inertial dynamics are taken into account. A comparison of sea levels projected by using the three variants shows that the time series are similar through the middle of the twenty-first century but they radically diverge by the end of the twenty-third century. This result is interpreted with the aid of the underlying transfer functions. It suggests that one should be cautious when using the semiempirical method to project sea level beyond the twenty-first century.

It has been known for a while that there are two sites of wintertime dense water formation in the North Adriatic – one in the open sea and the other in the Croatian coastal sea. Recently, it has been established that dense water is transported between the two basins, with both directions of the transport being possible. Here, a simple two-box model is developed in order to interpret the finding. The model allows for surface heat loss from the two basins and for an advective exchange of heat between the basins. Explicit solution is obtained, not only for the original, nonlinear problem but also for a simplified, linearized problem, when the initial temperature difference between the two basins vanishes. Moreover, the effect of the initial temperature difference is explored with the linearized model. The solutions point to a continuous temperature decrease in the two basins, with the temperature differences tending to limiting values. The temporal variability is controlled by the initial temperature differences, surface heat fluxes and basin dimensions and it suggests that the sum of surface heat loss and advective heat gain in one basin tends to become equal to the sum of surface and advective heat losses in the other basin. The solutions also indicate that the sign of the temperature difference between the two basins could be positive or negative, implying that the cold, dense water could be transported either way. Finally, an index, incorporating the initial temperatures, the surface heat fluxes and the basin depths, is proposed with the aim of quantifying relative importance of the two North Adriatic sites of dense water formation for each particular winter., Već je neko vrijeme poznato da se gusta voda stvara zimi u Sjevernom Jadranu u dva područja – na otvorenom moru i u hrvatskom obalnom moru. Nedavno je utvrđeno da dolazi i do transporta guste vode između ta dva područja, pri čemu su moguća oba smjera transporta. U ovom je članku razvijen jednostavan model dvije kutije radi interpretacije tih nalaza. Model uvažava gubitak topline s površine dvaju područja te advektivnu razmjenu topline između područja. Dobivena su eksplicitna rješenja, ne samo za izvorni, nelinearni problem nego i za pojednostavljeni, linearizirani problem, za slučaj kad iščezava početna razlika temperatura između dva bazena. Osim toga, razmotren je i utjecaj početne razlike temperatura pomoću lineariziranog modela. Rješenja ukazuju na kontinuirano smanjenje temperature u dva područja, pri čemu razlika temperatura teži graničnoj vrijednosti. Takva je vremenska promjenjivost određena početnom razlikom temperatura, površinskim protocima topline i dimenzijama bazena te sugerira da se suma površinskog gubitka topline i advektivnog primitka topline u jednom bazenu s vremenom približava sumi površinskog i advektivnog gubitka topline u drugom bazenu. Rješenja također pokazuju da predznak temperaturne razlike između dva bazena može biti pozitivan ili negativan, što znači da hladna, gusta voda može biti transportirana u oba smjera. Naposljetku, predložen je indeks koji uključuje početne temperature, površinske protoke topline i dubine bazena i koji bi trebao omogućiti da se kvantificira relativna važnost dvaju sjevernojadranskih područja stvaranja guste vode za svaku pojedinu zimu.

Continental Shelf Waves (CSWs) are oscillatory phenomena migrating along the continental margins, controlled by the interplay of rotation and bathymetric gradients. Here we combine observational data from five moored current meters and high-resolution hydrodynamic model fields for describing the generation and propagation of CSWs along the Southern Adriatic Margin (SAM, eastern Mediterranean Sea), where the possibility of their occurrence has been theoretically hypothesised but not experimentally observed up to now. Results show that in spring 2012 a train of CSWs with 35–87 km wavelength and 2–4 day period was generated on the northern sectors of the SAM and propagated southwards along its western slope. Along their path, CSWs modify their apparent frequency and oscillation mode as an effect of the background current and scattering caused by changes in the continental margin morphology. This signal appears as a persistent feature triggered by the inflow of a dense water vein formed in the northern Adriatic Sea, propagating upwelling and downwelling patterns along broad sectors of the continental slope. CSWs thus appear as an additional remote-controlled mechanism for cross-shelf exchange of water, sediment and nutrients in the SAM, besides the well-acknowledged dense water downflow along preferential pathways driven by local topographic constraints.

Regional sea levels in the Mediterranean sub-basins, the Black Sea and the Atlantic close to Gibraltar between 1930 and 2015, are constructed, based on high-quality tide-gauge data in the wider Mediterranean area, to identify long-term trends against decadal and multidecadal changes. Regional sea-level variability induced by direct atmospheric forcing and steric changes is determined, respectively, from air pressure and temperature and salinity data. Vertical land movements due to glacial isostatic adjustment are also taken into account. Focusing on linear trend in the period 1950–1990, the individual contributions to the trend are calculated and sea-level budget is examined within each region, according to proposed physical model. The trends with their uncertainty intervals are determined using Bayesian statistics. In the Atlantic off Gibraltar and in the Black Sea, the regional sea- level trends were close to the global values ; in the Mediterranean, they were close to zero. Sea-level rise in the Atlantic was supported by regional atmospheric loading and thermohaline changes, while the trend underlying the residual part of sea-level variability was comparable to the global mass contribution. Throughout the Mediterranean and in the Black Sea, atmospheric forcing and steric effects induced lowering of sea level. In the Mediterranean, and partly in the Black Sea, these regional effects compensated the effect of global mass increase. It is concluded that over the 1950–1990 interval, the sea-level budget is closed within the, rather wide, credible limits, which are obtained when autocorrelation of the linear-fit residuals is taken into account.

The paper investigates wintertime dynamics of the coastal northeastern Adriatic Sea, and is based on numerical modelling and in situ data collected through field campaigns executed during the winter and spring of 2015. The data have been collected by a variety of instruments and platforms (ADCPs, CTDs, glider, profiling float), and have been accompanied with a one-way coupled ALADIN/ROMS modelling effort. Research focus has been put on dense water formation (DWF), thermal changes and circulation, and water exchange between the coastal and open Adriatic. According to both observations and modelling results, dense waters are formed in the northeastern coastal Adriatic during cold bora outbreaks, even during milder-than-average winters (as was the winter of 2015). However, dense water formed in this coastal region has, due to lower salinities, lower densities than dense water formed at the open Adriatic. Since the sea is deeper in the coastal area than at the open Adriatic, dense waters from the open Adriatic occasionally enter the coastal area near the bottom of the connecting passages, while the surface flow is mostly outward from the coastal area. Median residence time of the coastal area is estimated to about 1–2 months, indicating that the coastal area may be relatively quickly renewed by the open Adriatic waters. The model significantly underestimates currents and transports in connecting channels, which may be a result of a too coarse resolution of atmospheric forcing, misrepresentation of bathymetry or absence of the air-sea feedback in the model. Obtained data represents a comprehensive marine dataset, pointing to a number of interesting phenomena to be investigated in the future.

A pool of less saline surface waters was observed in late June 2006 at the northern edge of the South Adriatic Pit (SAP). Three possible sources were considered: (1) Albanian rivers, (2) local Croatian rivers, or (3) relatively fresh West Adriatic Current (WAC) waters. Available CTD and ADCP data, together with satellite images, indicate that WAC waters are the most likely source. This requires an excursion of WAC water across the width of the Adriatic and is especially surprising as low winds and stable atmospheric conditions prevailed in mid/late June. However, quite strong NNW winds occurred during the first 12 days of June, with peak winds close to the western shore. These winds were the result of the translation of the Azorean high to the British Isles, producing strong pressure gradients over the Adriatic. The winds enhanced the WAC during early June 2006, preconditioning a cross-basin eddy circulation that appeared during the wind relaxation and calm conditions. As the unusually calm conditions persisted for more than two weeks, the WAC eddies and filaments grew freely and had enough time to reach middle east-Adriatic waters. Navy Coastal Ocean Model (NCOM) simulations, using high-resolution Adriatic bathymetry and realistic atmospheric forcing, show that such excursions are plausible and can occur when eddies and instabilities push WAC waters across the hyperbolic flow point separating the WAC and Eastern Adriatic Current near the Palagruža Sill. During the latter half of June 2006, NCOM simulations show that the hyperbolic point was particularly well formed as an anticyclonic WAC, a cyclonic SAP rim flow, an anticyclonic cell southeast of Lastovo Island, and a cyclonic cell over the centre of the Palagruža Sill all bordered on each other. A simplified channel model suggests that the presence of the escarpment is a critical factor for producing cross-basin exchange of the coastal current following the relaxation of strong winds with a cyclonic wind stress curl. However, the introduction of the Gargano Peninsula in the simulations was critical to the production of mesoscale eddies in the exchange flow, and such eddies qualitatively agree with the convoluted structures observed in satellite images.

The diurnal evolution of the sea/land breeze hodographs over the northeastern Adriatic coast has been examined for several episodes, each lasting a few days. A limited set of observations, as well as the results of a three-dimensional nonhydrostatic mesoscale model are studied by applying a rotary-component method. This revealed a spatial distribution of the clockwise versus anti-clockwise rotation of the wind vectors for the sea/land breeze events. An application of the rotary-component method to the results of additional numerical sensitivity tests showed that the coastal topography height considerably influences both the shape of the wind hodographs and the sense of wind rotation. The method also displayed which areas are under a strong influence of the 24-h topographic forcing and which areas show a large wind consistency. A further analysis of the 24-h modeled wind hodographs at several selected stations has been performed, thus revealing the hodograph shapes and the reasons for them.

This article delivers a short history of the early quantitative documentation of a rotor-type circulation in the bora-type flow on the northern Adriatic by Andrija Mohorovičić, an all-around geophysicist and the father of Croatian geophysical research who is widely known as the discoverer of discontinuity between the Earth's crust and mantle. This historical work presents an overview of Mohorovičić's research technique and rotor-related contributions, together with a short account of other observations of rotors contemporary to Mohorovičić as well as those from the 1920s and 1930s, considered to be seminal work on the subject on atmospheric rotors to date. In the year that marks the 150th anniversary of Mohorovičićs birth, his early meteorological observations remain germane for atmospheric rotor research, which is currently experiencing a renaissance with the Terrain-Induced Rotor Experiment (T-REX), a recently completed international field campaign and an ongoing research effort focused on atmospheric terrain-induced rotors.

The Bora wind field is characterised by strong vorticity and divergence. Several numerical experiments, in which an oceanographic model was forced with northeasterly winds having climatological alongshore variability, were performed in order to study the influence of spatial variability in the bora wind field on the surface currents in the northern Adriatic. Numerical model results showed that during bora episodes with lower speeds and fast offshore decay surface currents along transect Rovinj - Po River are predominantly in the downwind direction. On the other hand, during bora episodes with strong intensity and slow offshore decay, a cyclonic gyre due to the pronounced bora alongshore variability is formed in the northernmost part of the Adriatic Sea and the studied transect is influenced by the counter currents. Moreover, bora having a high speed and a short offshore range produces the same effect in the eastern part of the Rovinj - Po River transect as low-speed bora characterised by slow offshore decay. Eulerian current measurements performed in the northern Adriatic during bora episodes characterised by different synoptic conditions supported the numerical model findings. Surface currents during the bora episode of 8-11 February 1984 were directed downwind, whereas during the episode of 12-19 February 1984 they were directed upwind. The first episode was characterised by a deep bora layer with cyclonic activity over the western Mediterranean and Genoa Bay, whereas the second one was accompanied by temperature inversion and a southwesterly tropospheric wind above a shallow bora layer. According to the hydraulic theory developed by Smith(1985), an observed descent of isentropes during the second bora episode led to the stronger acceleration in the bora layer and its larger offshore extent. Different offshore bora decays during studied events were confirmed by a comparison of the wind data originating from the meteorological stations positioned on the opposite Adriatic coasts.

This study provides a new insight into the behaviour of the bora wind gusts. Wind speed and direction measured with a 1 s sampling interval between 1 December 2001 and 31 January 2002 at Senj (east Adriatic) provided a sufficiently large database for a study of the gusts behaviour. The performed spectral analysis confirmed the occurrence of pulsations in the bora flow. Moreover, a new type of dynamics, involving the onset, cessation and reappearance of the pulsations within a single episode, has been observed in several cases. Suggested mechanisms responsible for generating these phenomena have been determined from comparisons of surface wind data with the upstream tropospheric thermodynamical structure derived from Zagreb radiosonde data. In particular, it has been shown that the appearance of an upper-tropospheric jet stream results in cessation of the pulsations, and the decrease in the jet stream supports quasi-periodic gust behaviour. As the pulsations are generated by wave breaking, the jet stream appearance has been related to the disappearance of the wave-breaking region in the lee of the mountain, which is in accordance with previous studies of downslope windstorms. Copyright © 2004 Royal Meteorological Society.

A bibliography of papers, reports and books authored by Stjepan Mohorovičić is compiled and is presented in this paper. The bibliography con- tains 440 entries grouped into five categories: scientific papers (89 entries), professional and popular papers (87 entries), data reports (69 entries), literature reports (189 entries), and monographs and books (6 entries). A significant frac- tion of the papers and reports were published in international journals. The list reveals that Stjepan Mohorovičić was one of the most versatile and productive scientists working in Croatia in the twentieth century, whereas a brief overview of the impact of his publications, also presented in this paper, shows that he was one of the most creative Croatian scientists as well. It is stressed that Stjepan Mohorovičić’s output merits an in- depth evaluation and the bibliography is of- fered as the starting point for such an endeavor.

A catalogue of waves, which belong to the same frequency band as tsunamis but are of meteorological origin, is compiled for the Croatian part of the Adriatic Sea. The catalogue presently includes 21 flooding events observed between the years 1931 and 2010. The areas most often affected were Vela Luka on the island of Korčula and Stari Grad on the island of Hvar. A majority of the events occurred in the warm part of the year. The events tended to start either early in the morning or late in the afternoon, to last between 1 and 6 hours, and to be dominated by sea-level oscillations of 10-40 min periods. The largest trough-to-crest height of 6 m was observed in Vela Luka on 21 June 1978., Za hrvatski dio Jadranskog mora sastavljen je katalog valova, koji pripadaju istom frekvencijskom području kao tsunamiji, ali su meteorološkog porijekla. Katalog trenutačno uključuje 21 slučaj poplava, koje su opažene između 1931. i 2010. godine. Najčešće je poplavljivana Vela Luka na otoku Korčuli, te Stari Grad na otoku Hvaru. Većina se slučajeva dogodila u toplom dijelu godine. Epizode su pretežno počinjale rano u jutro ili kasno poslije podne, trajale su između 1 i 6 sati, a obilježilo ih je osciliranje morske razine s periodima od 10-40 minuta. Najveća valna visina u iznosu od 6 m opažena je u Veloj luci dana 21. lipnja 1978. godine.

This paper addresses an extraordinary storm surge in the Northern Adriatic that was more pronounced on the eastern than on the western shore. On 1 December 2008, Adriatic monitoring stations detected exceptionally high sea levels ; the oldest Croatian tide gauge station recorded the highest water level in its operating history at the time. Apart from the Northern Adriatic, large portion of the Dalmatian Coast was also exposed to high water levels, while Venice experienced a less- dramatic event. This marine storm was different from the capital storm of 4 November 1966 during which the surge had the highest impact ahead of Venice and along the north-western coastline. The 2008 event is studied here in detail, and the mechanisms that resulted in the different flooding of the two shores are identified. The study is based on hourly sea level, air pressure and wind data measured along both basin sides together with ECMWF reanalysis fields. Four components of sea-level evolution are identified: the storm surge, tide, Adriatic seiche and, low-frequency variability. The 2008 event was the outcome of a fine interplay between the first three components, which were all superimposed on the raised sea level due to low-frequency variability. The marine storm differed from the 1966 storm in the atmospheric forcing and relative timing of all contributing processes. The 2008 flooding of the eastern coast was mainly due to the Sirocco-wind shear, whereas the 1966 flooding of the western coast was due to the combined effect of almost uniform Sirocco and bottom slope.

The paper examines dense water formation in the South Adriatic by analysing temperature, salinity and dissolved oxygen data collected during six along-Adriatic cruises carried out from the Jabuka Pit to the Otranto Strait. The least-squares tracer analysis method is applied to distinguish the fractions of the four characteristic deep, dense water masses of the Adriatic Sea. Two types of dense water generation are considered: (1) shelf type occurring during cold and dry winters in the North Adriatic and resulting in the North Adriatic dense water (NAdDW) which partially transforms to the Middle Adriatic dense water (MAdDW) on its way along the Italian shelf to the Bari Canyon, where it interacts with topography, violently mixes and sinks; and (2) deep-convection type occurring in the centre of the South Adriatic Pit, generating dense water which reaches down to 800 m, fills the bottom and enters the Ionian Sea through the Otranto Strait. Both processes are visible in the July 1976 data, as the previous winter was cold (February) and relatively dry over the whole Adriatic basin. Both types of processes can influence the bottom layer of the Otranto Strait, where the fraction of shelf-type dense water generated in the North and Middle Adriatic was occasionally observed along with the fraction of water generated by deep convection in the South Adriatic. During the non-generative period, the cyclonic South Adriatic Gyre moved more to the northwest in general, as the Levantine intermediate water (LIW) occupied southeastern part of the South Adriatic Pit. However, such behaviour was not observed in January 1980, when LIW filled much of the Adriatic, thus diminishing density gradients in the South Adriatic. By applying a simple box model of water masses, it is estimated that the decay time of the Adriatic water masses equals 26 months, whereas transport of the Adriatic deep water (ADW) in the Otranto Strait amounted to 0.34 Sv during summer 1975. In the winter 1974/75 the ratio between the NAdDW+MAdDW and ADW transports in the Otranto Strait equalled 0.04, whereas during the winter 1975/76 it rose to 0.06. Conversion rates of NAdDW+MAdDW and LIW into ADW at that time are estimated at about 0.02 and 0.4–0.5 Sv, respectively.

The paper deals mainly with the temperature and salinity data collected during 30 cruises along the South Adriatic hydrographic section Bari-Dubrovnik but also uses data collected along the sections Palagruža Sill, Brindisi-Bar and Otranto between 1967 and 1990, most of which have never been described before. The analysis reveals the fractions of Middle Adriatic Deep Water (MADW), South Adriatic Deep Water (SADW) and Modified Levantine Intermediate Water (MLIW) computed by least-squares tracer analysis. The major feature found by analysis of the time series is the shift of water properties in the early 1980s, giving an increase in density (about 0.1 kg m −3 on average), due to exceptionally cold and dry winters over the whole Mediterranean. At the section Bari-Dubrovnik the fraction of SADW is the highest and the fractions of MADW and MLIW are the lowest during winter, because of the deep convection occurring in the South Adriatic Pit in winter, usually reaching the depth of about 600–800 m. Subsurface southeastward outflow of dense water usually occurs during spring, 4 months after the weakening of the Po River discharge and 2 months after the high surface buoyancy losses in the North Adriatic, respectively. Surface buoyancy losses occurring 6 and 8 months prior to the appearance of the subsurface outflow are in general favourable events for generation of dense water in the North Adriatic. The increase of MADW fraction at the bottom of the South Adriatic Pit lags for up to 2 years behind dense water formation in the North Adriatic. The case study of November 1989 reveals the renewal of bottom water in the South Adriatic Pit due to intrusion of MLIW, whereas in April 1987 strong mixing and sinking of extremely dense water, originating in the North Adriatic, was observed in the area of the Bari Canyon on the Italian shelf break. Near-bottom outflow of the Adriatic deep water towards the Ionian Sea was observed in the Otranto Strait in both cases. The case study of March 1975 revealed the maximum of SADW fraction.

In this paper we demonstrate the existence of a descending surface current which usually appears in August along the Istrian coast line and runs counter to the general Adriatic-wide cyclonic flow. This is achieved by computing the relative (with respect to the 30-dbar surface) geostrophic currents as well as absolute currents and net transports from hydrographic data collected monthly to seasonally at six stations of the Po-Rovinj profile in the northern Adriatic. We name it the Istrian Coastal Countercurrent (ICCC). Its mean relative speed is high, 7 cm s−1(relative currents at the profile on average stay below 5 cm s−1in winter and below 8 cm s−1in summer), inducing a net outflow near the Istrian coast. The ICCC was strong in 1968, 1977, 1983, 1984, 1988, 1989, 1991, 1992, 1996 and 1997, weak or absent in 1971–1973, 1982, 1986, 1993–1995. Whenever the ICCC was well pronounced, specific dynamic and hydrographic features—pointing to a possible existence of an anticyclonic gyre off Rovinj—were observed. The existence of the ICCC is confirmed by direct current measurements, which show that the descending coastal current was intense in 1977 and 1983, and that it did not develop in 1976, 1978 and 1987. The intensity of the ICCC in August over the 1966–1992 interval was significantly correlated to the air–sea heat flux observed with the −7 months time lag in combination with the Po River discharge rate with the −1 month time lag and to the sea surface temperature recorded in the area with the −5 months time lag. On the correlation basis a weak or non-existent ICCC is hindcasted for 1976, 1978 and 1987. It seems that high intensity of the ICCC coincides with oceanographic conditions which favour near anoxia or mucilage events, observed in northern Adriatic in 1977, 1988, 1989, 1991 and 1997.

Two campaigns, comprising the measurements of temperature, salinity, currents and sea level, were performed in the Zadar and Pasman Channels during summer 1994. Basic dynamics, including tidal, wind-driven and residual circulation, are briefly considered here, but the paper concentrates on the surface seiche and internal Kelvin wave analyses and modelling. Surface seiches are found to be excited by strong meteorological disturbances in the late summer of 1994, with a period of approximately 2·2 h and a sea-level and current amplitude of approximately 10 cm and 9 cm s−1, respectively. Internal Kelvin waves, having a period of approximately 4 days and maximum surface currents of approximately 8 cm s−1, appeared during July 1994. The energy loss at the head of the basin, occurring due to the basin bathymetry, is analytically modelled by imposing the radiation condition there, both for the surface seiches and internal Kelvin waves, while the nodal line is enforced at the basin mouth. The modelled periods of surface seiches and internal Kelvin waves are 2·1 h and 94 h, respectively. The currents and sea levels agree fairly well with the observed values. Furthermore, the Defant numerical procedure, which allows for a variable basin bathymetry, gives slightly lower periods (1·9 h for the surface seiches and 91 h for the internal Kelvin waves) and moves the maximum current amplitudes towards the head of the basin in agreement with the observations.

Analysis of oceanographic data collected during 2006 in the eastern Adriatic Sea indicated the presence of large internal waves (with a maximum range of about 30 m) at the diurnal frequency around the island of Lastovo. The amplitude ratio and phase difference between diurnal surface tides and diurnal isotherm fluctuations changed considerably between pronounced internal wave episodes, depending on stratification properties, thus suggesting possible resonant excitation of internal oscillations. On the contrary, no significant diurnal thermocline fluctuations were observed at two other islands (Bisevo and Susac). Theoretical analysis presented here focused on the trapping of long-period internal waves around a circular island corresponding to Lastovo and confirmed that stratification properties during the summer of 2006 around the island were close to resonant ones. The analysis also showed that Bisevo and Susac are too small to support diurnal near-resonant excitation. Application of a numerical model for the current flow around equivalent circular and elliptical islands in the stratified sea provided more details on resonant excitation. Theoretical and numerical modeling results particularly emphasized the importance of island dimensions, stratification properties (pyconocline depth and density defect) and the periodicity of the forcing. Furthermore, idealized numerical simulations demonstrated that the waves trapped at Lastovo behave as the gravest azimuthal mode of internal Kelvin-like waves, revolving in a clockwise direction around the island, and that the eccentricity of the island has almost no effect on the resonant period.

A salient feature of sea level records from the Adriatic Sea is the frequent occurrence of energetic seiches of period about 21 h. Once excited by a sudden wind event, such seiches often persist for days. They lose energy either to friction within the Adriatic, or by radiation through Otranto Strait into the Mediterranean. The free decay time of the dominant (lowest mode) seiche was determined from envelopes of handpassed sea level residuals from three locations (Bakar, Split and Dubrovnik) along the Croatian coast during twelve seiche episodes between 1963 and 1986 by taking into consideration only time intervals when the envelopes decreased exponentially in time, when the modelled effects of along-basin winds were smaller than the error of estimation of decay time from the envelopes and when across-basin winds were small. The free decay time thus obtained was 3.2±0.5 d. This value is consonant with the observed width of the spectral peak. The decay caused by both bottom friction and radiation was included in a one dimensional variable cross section shallow water model of the Adriatic. Bottom friction is parameterized by the coefficient k appearing in the linearized bottom stress term ρ0u (where u is the along-basin velocity and ρ0 the fluid density). The coefficient k is constrained by values obtained from linearization of the quadratic bottom stress law using estimates of near bottom currents associated with the seiche, with wind driven currents, with tides and with wind waves. Radiation is parameterized by the coefficient f appearing in the open strait boundary condition ζ =auh/c (where ζ is sea level, h is depth and c is phase speed). This parameterization of radiation provides results comparable to allowing the Adriatic to radiate into an unbounded half plane ocean. Repeated runs of the model delineate the dependence of model free seiche decay time on k and a, and these plus the estimates of k allow estimation of a. The principle conclusions of this work are as follows. 1. (1) Exponential decay of seiche amplitude with time does not necessarily guarantee that the observed decay is free of wind influence. 2. (2) Winds blowing across the Adriatic may be of comparable importance to winds blowing along the Adriatic in influencing apparent decay of seiches; across-basin winds are probably coupled to the longitudinal seiche on account of the strong along-basin variability of across-basin winds forced by Croatian coastal orography. 3. (3) The free decay time of the 21.2 h Adriatic seiche is 3.2±0.5 d. 4. (4) A one dimensional shallow water model of the seiche damped by bottom stress represented by Godin's (1988) approximation to the quadratic bottom friction law ρ0CDu|u| using the commonly accepted drag coefficient CD = 0.0015 and quantitative estimates of bottom currents associated with wind driven currents, tides and wind waves, as well as with the seiche itself with no radiation gives a damping time of 9.46 d; radiation sufficient to give the observed damping time must then account for 66% of the energy loss per period. But independent estimates of bottom friction for Adriatic wind driven currents and inertial oscillations, as well as comparisons between quadratic law bottom stress and directly measured bottom stress, all suggest that the quadratic law with CD=0.0015 substantially underestimates the bottom stress. Based on these studies, a more appropriate value of the drag coefficient is at least CD=0. In this case, bottom friction with no radiation leads to a damping time of 4.73 d, radiation sufficient to give the observed damping time then accounts for 32% of the energy loss per period.