A Neural Network Model for K(λ) Retrieval and Application to Global Kpar Monitoring.

Accurate estimation of diffuse attenuation coefficients in the visible wavelengths Kd(λ) from remotely sensed data is particularly challenging in global oceanic and coastal waters. The objectives of the present study are to evaluate the applicability of a semi-analytical Kd(λ) retrieval model (SAKM) and Jamet's neural network model (JNNM), and then develop a new neural network Kd(λ) retrieval model (NNKM). Based on the comparison of Kd(λ) predicted by these models with in situ measurements taken from the global oceanic and coastal waters, all of the NNKM, SAKM, and JNNM models work well in Kd(λ) retrievals, but the NNKM model works more stable and accurate than both SAKM and JNNM models. The near-infrared band-based and shortwave infrared band-based combined model is used to remove the atmospheric effects on MODIS data. The Kd(λ) data was determined from the atmospheric corrected MODIS data using the NNKM, JNNM, and SAKM models. The results show that the NNKM model produces <30% uncertainty in deriving Kd(λ) from global oceanic and coastal waters, which is 4.88-17.18% more accurate than SAKM and JNNM models. Furthermore, we employ an empirical approach to calculate Kpar from the NNKM model-derived diffuse attenuation coefficient at visible bands (443, 488, 555, and 667 nm). The results show that our model presents a satisfactory performance in deriving Kpar from the global oceanic and coastal waters with 20.2% uncertainty. The Kpar are quantified from MODIS data atmospheric correction using our model. Comparing with field measurements, our model produces ~31.0% uncertainty in deriving Kpar from Bohai Sea. Finally, the applicability of our model for general oceanographic studies is briefly illuminated by applying it to climatological monthly mean remote sensing reflectance for time ranging from July, 2002- July 2014 at the global scale. The results indicate that the high Kd(λ) and Kpar values are usually found around the coastal zones in the high latitude regions, while low Kd(λ) and Kpar values are usually found in the open oceans around the low-latitude regions. These results could improve our knowledge about the light field under waters at either the global or basin scales, and be potentially used into general circulation models to estimate the heat flux between atmosphere and ocean.

Regional ocean data assimilation.

This article reviews the past 15 years of developments in regional ocean data assimilation. A variety of scientific, management, and safety-related objectives motivate marine scientists to characterize many ocean environments, including coastal regions. As in weather prediction, the accurate representation of physical, chemical, and/or biological properties in the ocean is challenging. Models and observations alone provide imperfect representations of the ocean state, but together they can offer improved estimates. Variational and sequential methods are among the most widely used in regional ocean systems, and there have been exciting recent advances in ensemble and four-dimensional variational approaches. These techniques are increasingly being tested and adapted for biogeochemical applications.

A three-step model to assess shoreline and offshore susceptibility to oil spills: the South Aegean (Crete) as an analogue for confined marine basins.

This study combines bathymetric, geomorphological, geological data and oil spill predictions to model the impact of oil spills in two accident scenarios from offshore Crete, Eastern Mediterranean. The aim is to present a new three-step method of use by emergency teams and local authorities in the assessment of shoreline and offshore susceptibility to oil spills. The three-step method comprises: (1) real-time analyses of bathymetric, geomorphological, geological and oceanographic data; (2) oil dispersion simulations under known wind and sea current conditions; and (3) the compilation of final hazard maps based on information from (1) and (2) and on shoreline susceptibility data. The results in this paper show that zones of high to very-high susceptibility around the island of Crete are related to: (a) offshore bathymetric features, including the presence of offshore scarps and seamounts; (b) shoreline geology, and (c) the presence near the shore of sedimentary basins filled with unconsolidated deposits of high permeability. Oil spills, under particular weather and oceanographic conditions, may quickly spread and reach the shoreline 5-96 h after the initial accident. As a corollary of this work, we present the South Aegean region around Crete as a valid case-study for confined marine basins, narrow seaways, or interior seas around island groups.

Tracking the Atlantic Multidecadal Oscillation through the last 8,000 years.

Understanding the internal ocean variability and its influence on climate is imperative for society. A key aspect concerns the enigmatic Atlantic Multidecadal Oscillation (AMO), a feature defined by a 60- to 90-year variability in North Atlantic sea-surface temperatures. The nature and origin of the AMO is uncertain, and it remains unknown whether it represents a persistent periodic driver in the climate system, or merely a transient feature. Here, we show that distinct, ∼55- to 70-year oscillations characterized the North Atlantic ocean-atmosphere variability over the past 8,000 years. We test and reject the hypothesis that this climate oscillation was directly forced by periodic changes in solar activity. We therefore conjecture that a quasi-persistent ∼55- to 70-year AMO, linked to internal ocean-atmosphere variability, existed during large parts of the Holocene. Our analyses further suggest that the coupling from the AMO to regional climate conditions was modulated by orbitally induced shifts in large-scale ocean-atmosphere circulation.

Temporal changes of surface wave velocity associated with major Sumatra earthquakes from ambient noise correlation.

Detecting temporal changes of the medium associated with major earthquakes has implications for understanding earthquake genesis. Here we report temporal changes of surface wave velocity over a large area associated with 3 major Sumatra earthquakes in 2004, 2005, and 2007. We use ambient noise correlation to retrieve empirical Green's function (EGF) of surface waves between stations. Because the process is completely repeatable, the technique is powerful in detecting possible temporal change of medium. We find that 1 excellent station pair (PSI in Indonesia and CHTO in Thailand) shows significant time shifts (up to 1.44 s) after the 2004 and 2005 events in the Rayleigh waves at 10-20 s but not in the Love waves, suggesting that the Rayleigh time shifts are not from clock error. The time shifts are frequency dependent with the largest shifts at the period band of 11-16 s. We also observe an unusual excursion approximately 1 month before the 2004 event. We obtain a total of 17 pairs for June, 2007 to June, 2008, which allow us to examine the temporal and spatial variation of the time shifts. We observed strong anomalies (up to 0.68 s) near the epicenter after the 2007 event, but not in the region further away from the source or before the event or 3 months after the event. The observations are interpreted as stress changes and subsequent relaxation in upper-mid crust in the immediate vicinity of the rupture and the broad area near the fault zone.

Mean dynamic topography: inter-comparisons and errors.

Knowledge of the ocean dynamic topography, defined as the height of the sea surface above its rest-state (the geoid), would allow oceanographers to study the absolute circulation of the ocean and determine the associated geostrophic surface currents that help to regulate the Earth's climate. Here a novel approach to computing a mean dynamic topography (MDT), together with an error field, is presented for the northern North Atlantic. The method uses an ensemble of MDTs, each of which has been produced by the assimilation of hydrographic data into a numerical ocean model, to form a composite MDT, and uses the spread within the ensemble as a measure of the error on this MDT. The r.m.s. error for the composite MDT is 3.2 cm, and for the associated geostrophic currents the r.m.s. error is 2.5 cms(-1). Taylor diagrams are used to compare the composite MDT with several MDTs produced by a variety of alternative methods. Of these, the composite MDT is found to agree remarkably well with an MDT based on the GRACE geoid GGM01C. It is shown how the composite MDT and its error field are useful validation products against which other MDTs and their error fields can be compared.

Interannual variability of surface heat fluxes in the Adriatic Sea in the period 1998-2001 and comparison with observations.

Surface heat fluxes of the Adriatic Sea are estimated for the period 1998-2001 through bulk formulae with the goal to assess the uncertainties related to their estimations and to describe their interannual variability. In addition a comparison to observations is conducted. We computed the components of the sea surface heat budget by using two different operational meteorological data sets as inputs: the ECMWF operational analysis and the regional limited area model LAMBO operational forecast. Both results are consistent with previous long-term climatology and short-term analyses present in the literature. In both cases we obtained that the Adriatic Sea loses 26 W/m2 on average, that is consistent with the assessments found in the literature. Then we conducted a comparison with observations of the radiative components of the heat budget collected on offshore platforms and one coastal station. In the case of shortwave radiation, results show a little overestimation on the annual basis. Values obtained in this case are 172 W/m2 when using ECMWF data and 169 W/m2 when using LAMBO data. The use of either Schiano's or Gilman's and Garrett's corrections help to get even closer values. More difficult is to assess the comparison in the case of longwave radiation, with relative errors of an order of 10-20%.

Changes in nutrient concentrations and ratios during mucilage events in the period 1999-2002.

Nutrient and chlorophyll a concentrations and salinity were measured, approximately monthly, from June 1999 to July 2002 at 20 stations along three transects in the northern Adriatic Sea, north of the line Susak Island-Senigallia, with the aim of confirming or rejecting the hypothesis that changes in nutrient ratios may have an important role in the mucilage phenomenon. The data analyses were focused on the two major water types identified in the region: lower salinity (32-37) and oxygenated surface waters (type 1) in which the mucilage phenomenon primarily developed, and high salinity water originating from other parts of the Adriatic (type 4). Marked variability of dissolved inorganic nitrogen (DIN) in type 1 waters was roughly correlated with extreme fluctuations of the Po River flow rate during the investigated period. In contrast, the orthophosphate (PO4) concentration was primarily controlled by phytoplankton assimilation. The nutrient discharges and DIN/PO4 ratios (median 120) in the freshened surface layer were much higher and more variable in the period before the mucilage event in 2001 than in 2000 (median 75), and particularly in 2002 (median 30), although intensity and duration of the 2001 event were the lowest. However, in that period of 2000 and 2002 significant transversal transport of freshened waters occurred, despite the unusually low Po flow rates. In summer, in conditions of low freshwater discharge and the prevailing of semi-enclosed circulation in the region, more efficient DIN assimilation by phytoplankton occurred, probably due to a faster recycling of PO4. However, in 2002 this process appeared to have already started in March. Changes of the orthosilicate (SiO4)/DIN ratio were mainly dependent on DIN concentrations. In the more saline waters (type 4) the nutrient concentrations, particularly DIN, were much lower and no significant relationships were noticed among the studied parameters. Nutrient concentration and ratio changes do not trigger mucilage events, although very probably they have an essential role in combination with several other physical (pulsing freshwater discharge, marked stratification, minimal water exchange) and biological (e.g., increased plankton excretion, limited bacterial degradation) factors.

Meteorological and oceanographic conditions in the northern Adriatic Sea during the period June 1999-July 2002: influence on the mucilage phenomenon.

Mucilage events (formation of very large organic aggregates and gelatinous surface layers) have been documented several times during the past two centuries in the northern Adriatic Sea (NA), while their frequency has significantly increased since 1988. In this work, meteorological and oceanographic conditions in the NA during the period June 1999-July 2002 are described and their relation to the outbreak and fate of the mucilage phenomenon was investigated. Salinity and temperature data were collected during approximately monthly cruises along three transects in the NA. Relevant meteorological situations (air temperature, rainfall, wind) were selected from large-scale ECMWF analyses and from the Local Analysis and Prediction System (LAPS; Emilia Romagna Meteorological Service), while sea conditions (waves) were analysed by means of the Wave Adriatic Model (WAM). Data for air temperature, rainfall, and wind from several meteorological stations in the region were used. Average seasonal cycles of sea temperature and salinity simulated with statistical models, based on historical data collected in the NA since 1972, were used to determine thermal and haline anomalies. The monthly anomaly variability of maximum and minimum air temperatures, rainfall amount and number of rainy days did not appear to be relevant for the mucilage phenomenon outbreak. In contrast, both vertical and horizontal thermohaline gradients in the region were more developed during late spring and summer of 2000 and particularly of 2002, when the mucilage events were of greatest extent in space and time, compared to 2001 (short-lived event) and 1999 (no event). These more pronounced gradients were due to a combination of several unusual conditions: sharp heating of the sea surface in May-June, domination of eastwards transport of freshened waters formed in the Po Delta area, and intrusion of very high salinity intermediate waters originating in the eastern Mediterranean. Moreover, in winter of both 2000 and 2002 very dense and cold water formed and remained in the bottom layer until spring, contributing to increase the stratification degree of the water column. The duration of the mucilage events and their spatial distribution in the region depend strongly on meteorological changes. Recurrent anticyclonic conditions, characterized by low wind and calm sea, favour extended events in time (up 2 months in 2002). In contrast, highly perturbed weather, particularly due to strong "bora" wind, can be determined in sharp decay of the event (e.g. in July 2000).

Trends and interactions of physical and bio-geo-chemical features in the Adriatic Sea as derived from satellite observations.

Time series of satellite data, generated by the AVHRR (1981-1999), CZCS (1979-1985) and SeaWiFS (1998-2002), have been used to assess trends and interactions of physical and bio-geo-chemical features in the Adriatic Sea. The images were processed to estimate Sea Surface Temperature (SST) and Chlorophyll-like Pigment Concentration (CPC). Long-term composites and climatologies were derived, using fixed geographical grids and projections. The AVHRR data show an apparent warming trend, when plotting the sequence of seasonal cycles (monthly mean SST, averaged over the whole basin) against time, due to a steady rise of summer values. Considering 3 regions (north, central and south), split into east and west sections, the northern Adriatic shows high SST fluctuations (possibly associated with the cycle of winter cooling and summer warming, typical of the relatively shallow sub-basin), while the southern Adriatic exhibits a lower variability (possibly influenced by the periodic water incoming from, and outflowing to the Ionian Sea). During summer, an east-west gradient prevails, while during winter only a general north-south gradient can be found. The SeaWiFS-derived CPC values, distributions and trends appear to be consistent with the historical CZCS record. Persistent differences in the quantitative assessment of CPC for coastal waters is due to the use of improved algorithms, less influenced by the presence of dissolved organics and suspended sediments in the water column, for the processing of SeaWiFS data. Apparent incongruities of the space and time patterns in the SeaWiFS record with respect to the reference climatology, obtained by CZCS more than a decade before, occur chiefly when considering the spring bloom in the southern Adriatic and the summer development of the north Adriatic front. The comparison of the long-term times series of satellite data shows that there is a high correlation between patterns in the thermal field and in the colour field. This suggests that different surface waters, identified by the SST index, are also traced by different ecological features, identified by the CPC index. Both indices also show a high correlation with the classical cyclonic circulation scheme of the Adriatic Sea, proposing once again an intimate relationship between the water dynamics and its bio-geo-chemistry.

A numerical study of the interannual variability of the Adriatic Sea (2000-2002).

A free-surface, three-dimensional finite-difference numerical model based on the Princeton Ocean Model (POM) has been implemented in order to simulate the interannual variability of the Adriatic Sea circulation. The implementation makes use of an interactive surface momentum and heat flux computation that utilizes the European Centre for Medium-Range Weather Forecasts (ECMWF) 6-h analyses and the model predicted sea surface temperatures. The model is also nested at its open boundary with a coarse-resolution Mediterranean general circulation model, utilizing the same surface forcing functions. The simulation and analysis period spans 3 years (1 Jan 2000 to 31 Dec 2002) coinciding with the "Mucilage in the Adriatic and the Tyrrhenian" (MAT) Project monitoring activities. Model results for the simulated years show a strong interannual variability of the basin averaged proprieties and circulation patterns, linked to the atmospheric forcing variability and the Po river runoff. In particular, the years 2000 and 2002 are characterized by a weak surface cooling (with respect to the climatological value) and well-marked spring and autumn river runoff maxima. Conversely, 2001 is characterized by stronger wind and heat (autumn cooling) forcings but no river runoff autumn peak, even though the total amount of water inflow during winter and spring is sustained. The circulation is characterized by similar patterns in 2000 and 2002 but very different structures in 2001. During the latter, deep water is not formed in the northern Adriatic. A comparison with the observed data shows that the major model deficiencies are connected to the low salinity of the waters, probably connected to the missed inflow of salty Ionian waters of Aegean origin and to the numerical overestimation of the vertical mixing processes.

Circulation and horizontal fluxes in the northern Adriatic Sea in the period June 1999-July 2002. Part II: nutrients transport.

Nutrient concentrations (orthophosphate, orthosilicate and dissolved inorganic nitrogen-DIN) were measured on three transects in the northern Adriatic approximately monthly in the period June 1999-July 2002. The net nutrient transports across the three transects were estimated from these concentration values and calculated water fluxes [Grilli F, Paschini E, Russo A, Precali R, Supic N. Circulation and horizontal fluxes in the Northern Adriatic Sea in the period June 1999_July 2002. Part I: geostrophic circulation and current measurement. Sci Total Environ 2005-this issue], with the aim to verify their possible role in the mucilage phenomenon. The nutrient transports in the northern Adriatic were very variable, seasonally and from year to year, both in intensity and direction. Some differences were noticed also among the nutrient species. At the northernmost transect Po Delta-Rovinj minimal values and variable directions were recorded, probably due to the prevailing of eastward transversal transports, observed already in late winter, and generally dominant in spring and summer. Northward transport was often measured at the central transect Cesenatico-Cape Kamenjak, particularly in spring 2000 and 2002, and summer 2001, as well as for only DIN in summer 1999. In contrast, southward directions prevailed at the southernmost transect Senigallia-Susak Island. In October significant southward transports occurred at all transects and for all nutrients, particularly strong in 2000, when exceptionally high Po River discharges occurred, and the Western Adriatic Current (WAC) was well developed. However, the nutrient transports in the upper water column were low or northwards at both transect Cesenatico-Cape Kamenjak and Senigallia-Susak Island in June and July 2000, 2001, and 2002, when the mucilage phenomenon developed, but were opposite in 1999, a year without events. The results suggest that high variations of nutrient fluxes and their ratios, in conditions of reduced water dynamics, are essential for the development of the phenomenon, rather than the absolute amounts of the nutrient inputs.

Circulation and horizontal fluxes in the northern Adriatic Sea in the period June 1999-July 2002. Part I: geostrophic circulation and current measurement.

The dramatic increase in the occurrence of massive mucilage events in the northern Adriatic (NA) since their recent conspicuous reappearance in the late 1980s prompted a study of circulation and horizontal fluxes. Three transects with equidistant stations (10 km) were thus monitored monthly between June 1999 and July 2002. The geostrophic method was used to compute currents across the three transects from the CTD data, and dynamic heights provided a picture of the horizontal surface circulation. Currentmeter data records were used to adjust the reference surface and to validate the results for the southernmost and deeper (up to 70 m) transect (Senigallia-Susak Island). Geostrophic currents allowed estimation of monthly water fluxes across the transect. Different circulation regimes in the NA were observed, which may have affected mucilage events. When mucilage was absent (1999) or reduced (2001) in the western sector, the Western Adriatic Current (WAC, carrying water out of the NA) was found to be active, whilst the WAC was very weak or reversed when massive mucilage events occurred (2000 and 2002). Opposite behaviour has been observed for the Istrian Coastal Counter-Current (ICCC, retaining freshwater water in the NA) which was more intense during or after massive mucilage events and did not appear when mucilage was absent. Both WAC weakening and ICCC strengthening indicate a longer residence time of riverine waters in the NA, which favours mucilage development. Conclusively, WAC and ICCC result as key elements in controlling massive mucilage phenomena in the NA.

Field observations of dilution on the Ipanema Beach outfall.

Field observations of the Ipanema Beach, Rio de Janeiro, ocean sewage outfall are presented. Measurements of dilution and other wastefield characteristics were obtained by adding dye tracer to the effluent and measuring in-situ. Simultaneous measurements of oceanographic conditions were made by Acoustic Doppler Current Profilers, thermistor strings, and profiling instruments. Four experiments were performed, two during unstratified conditions when the plume was surfacing, and two during conditions of strong stratification when the plume was submerged. The minimum dilution varied from 30 to 130. The measurements reflect the worst case conditions as the campaigns were all made for weak currents.