CLIMATE RESEARCH Clim Res Vol. 31: 151–165, 2006 Published July 27 Anomalous Po River flood event effects on sediments and the water column of the northwestern Adriatic Sea F. Frascari1, F. Spagnoli2,*, M. Marcaccio3, P. Giordano1 1 CNR-ISMAR, Sezione di Geologia Marina, Via Gobetti 101, 40129 Bologna, Italy 2 CNR-ISMAR, Sede di Ancona, Largo Fiera della Pesca 2, 60125 Ancona, Italy 3 ARPA Emilia Romagna, Via Triachini 17, 40138 Bologna, Italy ABSTRACT: The aim of this study was to investigate the spreading pattern and trophic impact of the anomalous Po River flood in the fall of 1994 in the marine area offshore of the Po River delta. The study was carried out by surveying the areal distribution of physical and biogeochemical parameters (pH, redox potential [Eh], water content, grain-size, mineralogy, organic and inorganic C, total N, total organic and inorganic P) in surficial and sub-surficial sediments, and physical-chemical properties (salinity, temperature and dissolved O2) of the water column. Water column investigations highlighted a fresh, turbid and relatively cold Po River water wedge on the basin surface that extended more than 40 km offshore, well above the usual limits of the Po River plume influence, which was a result of slow geostrophic currents and no wave motion in the basin. The distribution of minerals in sediments (such as serpentine and dolomite), derived from river catchments, provided cues on the maximum extension of Po and northern river influences. Biogeochemical tracers such as organic C and organic P, together with C:N ratios, provided indirect evidence on the development of algal blooms owing to the Po River flood run-off. Indeed, they indicated areas where there was the deposition of autochthonous reactive organic matter, derived from fresh plankton biomass. Their location suggests that the maximum development of plankton blooms occurred along the external edge of the plume, where warm basin waters mixed with colder nutrient-rich waters from the river. The data also suggest that significant amounts of newly formed and degrading organic matter reached bottom waters at the centre of the basin, thus increasing the degree of hypoxia. KEY WORDS: Adriatic Sea · Po River · Organic matter · Biogeochemical tracer · Algal bloom · Flood Resale or republication not permitted without written consent of the publisher To better understand how the spreading mechanisms of Po River waters and their particulate load affect the primary production dynamics in the northern Adriatic Sea, it is important to improve our knowledge of dispersion mechanisms of Po River inputs into the early impact zones of the North Adriatic basin in various discharge regimes and under various seasonal conditions. One of the main obstacles encountered when investigating the northwestern Adriatic Sea is the difficulty of identifying, with in situ observations, the impact zones and maximum expansion processes of Po River inputs towards the centre of the basin during exceptional floods in different seasons. Usually such extreme events cannot be foreseen, and thus marine surveys cannot be planned a priori. For the same reasons it is difficult to investigate how and where algal blooms develop when such exceptional events occur, and with what mechanisms and at what depths the settling autochthonous organic matter degrades and consumes oxygen in the water column, which eventually affects early diagenesis on the bottom sediments (Berner 1980, Barbanti et al. 1992). Actually, the autochthonous organic matter is particularly reactive and sensitive to a partial mineralization already present in the water *Corresponding author. Email: [email protected] © Inter-Research 2006 · www.int-res.com 1. INTRODUCTION 152 Clim Res 31: 151–165, 2006 2. STUDY AREA column, where it consumes dissolved oxygen. Furthermore, the organic matter, which is deposited together The study area extends about 40 km offshore from with fine inorganic particles on the bottom, feeds the the Po River delta; it is delimited to the north by the bacterial degradation processes that are first oxic Adige River mouth, and to the south by the Reno River and later anoxic (i.e. when degradation becomes more mouth (Fig. 1). intense or takes place at greater depths in the sediFrom a hydrodynamic point of view, the marine area ments) (Froelich et al. 1979, Berner 1980, Orel et al. in front of the Po River delta is characterised by 1993). In this manner, degradation processes consume summer water column stratification, with weak geodissolved oxygen in bottom waters and cause the slow strophic currents that form both counter-clockwise and release of nutrients. Indirect indications of deposition clockwise gyres, and an autumn-winter structure in and degradation of organic matter, relative to a single which the water column stratification disappears and a event or averaged over seasonal, annual or multiannual periods, may be obtained from bottom sediment studies. Bottom sediments record both the offshore dispersion of suspended river organic and inorganic particulate and the deposition of autochthonous organic materials. However, in order to be able to discriminate single sedimentation or reworking processes, seabottom sediment investigations are not enough; we need observations that are less averaged out over time, namely those that take place in the water column. From 8 to 21 November 1994, a survey cruise (PR94) was conducted within the framework of the PRISMA 1 National Programme (Frascari et al. 1998). The initial purpose of the cruise was the deployment of 3 automatic oceanographic measurement stations, but it happened to coincide with an anomalous Po River flood. As a consequence, we tried to obtain functional data to understand the processes related to the Po River flood in the marine area in front of the Po River delta by using all available instruments on board. To this aim, we performed both water column chemicalphysical measurements and bottom sediment samplings and analyses. The scope of this work is to show the sediment and water column processes and their extension that took place in the marine offshore area in front of the Po River delta during the November 1994 flood event, as deduced from the chemical-physical parameters of the water column and from the bioFig. 1. Study area. Numbered dots: multiparametric (temperature, salinity and disgeochemical, mineralogical and sedi- solved oxygen) water column profiles and box-corer stations. Solid lines: tracks mentological composition of bottom of multiparametric transects discussed in text (Transects A, B and C). Bathymetric contour interval: 10 m sediments. Frascari et al.: Po River anomalous flood event north-south geostrophic current, the stronger north Adriatic current (N-Ad), occurs (Franco et al. 1982, Malanotte-Rizzoli & Bergamasco 1983, Artegiani et al. 1989, Orlic et al. 1992). In this way, on a seasonal basis, Po River inputs extend further offshore in summer, whilst during winter they tend to be more confined along the western coast of the basin. As a general result the Po River inputs are conveyed almost entirely southwards. The bottom sediment grain-sizes in the investigated area (Brambati et al. 1983) are the result both of the general hydrodynamics of the northern Adriatic Sea and of the geological evolution of the last 20 000 yr (Colantoni et al. 1979b, Trincardi et al. 1994). Along the coast there is a thin coastal belt of sandy sediments, up to a depth of about 7 to 8 m, as a result of inputs of coarse material from rivers and coastal wave reworking; this is followed offshore by pelitic sediments in a belt along a north-south axis, deriving from suspended material inputs both from the Po River (between 18 × 106 t yr–1 [Dal Cin 1983] and 26 × 106 t yr–1 [IRSA 1977]), and to a lesser extent from other minor rivers. Towards the centre of the basin there is a gradual change to mainly sandy relict sediments (Colantoni et al. 1979a). Po particulate inputs are composed of rock fragments and minerals of heterogeneous origin with the presence of mainly silico-clastic elements, whilst the solid load transported by rivers north of the Po has a largely carbonate composition (Guerzoni et. al. 1984). The phytoplankton biomass and primary production dynamics in the northern Adriatic Sea are characterised by a permanent high production, especially in the coastal belt along the western coast (Fonda Umani et al. 1992). The area with the higher production is quantified off the Po River delta and related to the spreading of its plume. In this area, a marked west-toeast gradient of the standing crop and production is observed (Smodlaka & Revelante 1983). The phytoplankton standing crop in the northern Adriatic is related to thermohaline distribution: during periods of reduced vertical mixing and conspicuous river inputs, the surface community presents a cell density that is 25 times the bottom layers, dominated by microflagellates and diatoms; in contrast, when the Po River outflow is particularly reduced, phytoplankton density is relatively homogeneous throughout the water column and is always dominated by microflagellates and diatoms. The South Po River delta coast is characterised by a 6 to 10 km belt of spring diatom blooms and dinoflagellate summer monospecific blooms (Marchetti 1983) Generally, these blooms are linked to a highly stratified system that often reoccurs at the beginning and end of the summer period, but in almost every summer since 1975 they have given rise to red tide phenomena (Fonda Umani et al. 1990). 153 3. MATERIALS AND METHODS During the PR94 cruise (8 to 21 November 1994), in order to study the unforeseen flood event that took place at the Po River mouths between approximately 6 and 24 November, temperature, dissolved oxygen and salinity profiles of the water column were measured using a multiparametric probe (SBE 911 plus), and sediment box-corers were collected to sample bottom sediments. Stations were located in the main marine impact zones of the Po River flood inputs, along transects perpendicular to the coast line, with a frequency distance between stations of each transect of about 5 nautical miles near the Po River mouth and 12 nautical miles at the farthest extent (Fig. 1). Unfortunately, it was not possible to measure turbidity and chlorophyll a profiles owing to the lack of necessary equipment on board. The multiparametric profiles were conducted within a few days in order to have a synoptic view of the physical-chemical properties of the water column, with the main aim of evaluating the persistence of water column stratification in a season that is usually marked by high hydrodynamics, with constant values of temperature and salinity throughout the whole water column (Hendershott & Rizzoli 1976, Franco 1983, Franco et al. 1983, Fonda Umani et al. 1990, Artegiani et al. 1997). As far as bottom investigations are concerned, it should be pointed out that we decided not to sample coastal sands; our main purpose was to evaluate the extension mechanisms (in the direction of the open sea) of suspended material coming from the Po River during the flood, and to assess evidence on the bottom of the autochthonous organic matter being deposited. Sampling extended to zones further out to sea, where relict sands outcrop (Brambati et al. 1983), to evaluate the maximum extension of fine material deposition during exceptional floods. Each sediment box-corer was first described macroscopically to identify sedimentary structures, grain-size, bioturbation and macrofauna presence. In particular, the thickness of the recently deposited material, recognisable by the high degree of hydration and ochre colour, were recorded. Next, sub-samples were collected at 2 levels (surficial and sub-surficial) in each box-corer. As far as possible, the thickness of the surficial layer (0.5 to 1 cm thick) was chosen inside the thickness of the recently deposited material. The sub-surficial sample (about 2 cm thick) was collected immediately below the flood layer. This sampling strategy was based on the different information that surficial and sub-surficial samples can give. In fact, while surficial sediments provide information on the last sedimentological event (on the condition that sampling is carried out very soon after the event and that other processes, such as currents or 154 Clim Res 31: 151–165, 2006 wave motion, did not rework the sediment), sub-surficial samples provide information on more sedimentological events, averaged over time, because of the different reworking processes that act on sediments. Physical and chemical parameters (redox potential [Eh] and pH) were measured in each sub-sample immediately after the collection, and then each sample was subdivided into aliquots for sedimentological, mineralogical and geochemical analyses. Water content, grain-size and mineralogical composition of sediment samples were determined on an aliquot stored at 4°C. Organic C, total N, total, inorganic and organic P content analyses were performed on an aliquot stored at –20°C and then freeze dried. The water content was calculated as the difference between the weight of the wet and dried sediment samples; samples were dried at 70°C until they reach a constant weight (Gallignani & Magagnoli 1972). Grain-size analyses were conducted by wet sieving, in order to separate the finest fractions (< 63 µm) from the coarser fractions (> 63 µm), and then by X-ray sedigraphy on the finer fractions. Mineralogical composition was determined through X-ray-diffrattometry, and percentages of the most abundant minerals were evaluated (Cook et al. 1975). Total and organic C and total N (mainly organic N; Giordani & Angiolini 1983) were determined by means of a CHNS elemental analyser. Organic C was determined by decarbonisation with HCl 1 M and drying at 60°C for 2 h (Froelich 1980, Hedges & Stern 1984). Inorganic P was determined on the extract of sediment samples that were stirred at room temperature for 16 h with HCl 1 M (Aspila et al. 1976). Total P was determined on the extract after burning to ashes at 550°C for 2 h and stirring for 16 h with HCl 1 M. Reactive phosphates in all extracts were determined using colorimetric technique (Strikland & Parson 1972, Griffiths 1973). The organic P content was calculated as the difference between total and inorganic P. finished about 24 November (2270 m3 s–1) (Fig. 2); the maximum flood discharges, reached on 10 November, was 8630 m3 s–1; and average flood flow was 5479 m3 s–1, which was considerably above the average of the Po River’s discharge between 1917 and 1994 (1500 m3 s–1, Ufficio Idrologico e Mareografico pers. comm.) and that of the average discharge for 1994 (1870 m3 s–1). The flood brought into the sea 5.1 × 106 t of suspended solids (equal to 48% of the annual amount), 3297 t of total P (26%) and 27 600 t of N (25%) (Marchetti et al. 1996). In the survey period (8 to 21 November 1994), which coincided with the arrival at the sea of the flood waters, the weather-sea conditions of the northern Adriatic Sea were characterised by generally low hydrodynamic with regard to both geostrophic current intensity and wave motion, as observed by other field investigations in the same period in the northwestern Adriatic Sea (ELNA Project cruise; Hopkins et al. 1999, Hopkins 2001). The salinity, temperature and oxygen profiles measured in the water column allowed us to reconstruct the state of the basin at the moment of the flood discharge. Salinity, temperature and oxygen data were elaborated to obtain longitudinal transects located perpendicularly to the coastline, immediately to the south of the Po River delta (Transect A, Fig. 3), in front of the delta (Transect B, Fig. 4) and to the north (Transect C, Fig. 5) and also surface and bottom distributions of these parameters in front of the Po River delta (Figs. 6 & 7). The salinity and temperature transect just south of the Po River delta (Transect A, Figs. 1 & 3a,b) highlights a strong stratification of the water column, with a surface layer (thickness ~10 m) of colder and less saline water, relative to underlying water. The temperature and salinity values inside the upper layer increase gradually both from the surface down towards the lower layer and in a land-to-sea direction (Figs. 3a,b & 6a,b). This layer extends for over 40 km offshore in the zone to the southeast of the Po River delta (Fig. 6a,b) 4. RESULTS 10 m3 s–1 (103) The November 1994 Po River flood was caused by intense rainfall in a zone to the southwest of the Po River’s drainage basin, in Piedmont (northwest Italy). In this region, precipitation reached values of between 50 and 175 mm in 24 h (Autorità di Bacino del Fiume Po 1994) between 5 and 6 November 1994. Such intense rainfall caused a huge flood in the upper reaches of the Po River that was slightly reduced when it arrived at the sea mouths for normal discharges of tributaries into the lower part of the Po River. The flood reached the Po mouths 1 d after the rainfall began. The runoff into the sea began around 6 November (2290 m3 s–1) and Po River discharge 8 6 4 2 0 2 Nov 11 Nov 18 Nov 26 Nov 4 Dec 12 Dec Days (1994) Fig. 2. Po River flood runoff (5 to 25 November 1994) with annual (dashed line) and 1917–94 means (solid line) 12 16 20 24 25B 25 24 23 22B 32 28 36 5 . 36 37.0 –10 37.5 0 A1 a 22A Frascari et al.: Po River anomalous flood event –20 Transect A –30 Salinity PSU 12 16 20 24 28 32 36 40 –40 b 0 12 14 1615 15 16 Depth (m) –10 17 17 17 –20 17 Transect A –30 Temperature °C 12 13 14 15 16 17 18 19 20 –40 0 c 5 6 6 6 5 –10 4 5 –20 4 –30 Transect A Oxygen 2 2 mg l–1 0 1 2 3 4 5 6 7 8 9 10 3 3 3 4 –40 Fig. 3. (a) Salinity, (b) temperature and (c) dissolved oxygen distribution along Transect A 155 while it is confined to the coast, with a less accentuated stratification in zones in front of and immediately north of the delta (Figs. 4a,b, 5a,b & 6a,b). This situation is similar to that shown by Artegiani et al. (1997) for the autumn season, and by Franco (1983) with regard to the buoyancy of river waters over basin water, but with a major offshore extension of the less salty water front owing to the exceptional character of the flood relative to the averaged data of Artegiani et al. (1997). Moreover, the temperature transects show that in the deeper and eastern zones, an additional layer near the bottom is characterised by lower temperatures (Figs. 3a, 4a & 5a). The deep colder water layer is also visible in the central part of the basin in the bottom temperature areal distribution (Fig. 7a). Over the whole area, the salinity between the intermediate and deeper waters increases with a more gradual and regular trend than does variation in temperature (Figs. 3a, 4a, 5a & 7a), reaching a maximum in the eastern part of the studied area and forming a deep layer with both high salinity and low temperature. The dissolved oxygen distributions along the transects (Figs. 3c, 4c & 5c) allow us to note an oxygen-rich surface layer that extends as far as the external limits of the flood plumes (Fig. 6c) and a bottom hypoxic layer that corresponds to a north-south elongated area extending to the external limits of the flood plume (Fig. 7c). Both the deep hypoxic layer and the welloxygenated surface layer tend to become thicker and more evident in the frontal zones where the Po River fresh waters mix with the saltier waters of the basin (Figs. 3c, 4c & 5c). The distribution of water content in surficial sediments (Fig. 8a) shows 2 areas characterised by higher values. The 2 areas correspond to the 2 mouths of Po di Goro and Po della Pila. The first of these areas extends south-southeast, whilst the second— after initially moving towards the 16B 17 –20 5 37. –30 Transect B Salinity –40 0 b 13 –10 16 –20 Depth (m) Depth (m) 17 18 17 18 Transect B Temperature –20 17 °C 12 13 14 15 16 17 18 19 20 –40 0 12 13 14 15 16 17 18 19 20 c Transect C Temperature 18 °C –40 0 16 17 –30 –30 Transect C Salinity PSU PSU –10 37.0 37.5 12 16 20 24 28 32 36 40 12 16 20 24 28 32 36 40 –40 0 7 36 0 37. –20 b 6 –10 36.5 35.5 36 36. 5 5 35. 12 34.5 –10 –30 5 0 4 a 3 16BIS 15C 0 144 a Clim Res 31: 151–165, 2006 15TRIS 156 6 c 5 –10 –10 5 –20 5 4 –20 5 4 43 3 –30 2 Transect B Oxygen –30 Transect C Oxygen mg l–1 mg l–1 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 –40 Fig. 4. (a) Salinity, (b) temperature and (c) dissolved oxygen distribution along Transect B –40 Fig. 5. (a) Salinity, (b) temperature and (c) dissolved oxygen distribution along Transect C 157 Frascari et al.: Po River anomalous flood event a 45° 45’ N b Salinity surface data Temperature surface data PSU 45° 30’ -20 Venezia Chioggia Adige River 28 °C Chioggia Adige River Po di Goro 20 28 16 Po Pila Po di Goro 24 44° 45’ 13 14 -40 Reno River 20 32 44° 30’ 14 15 28 Ravenna Ravenna Cervia Cervia 44° 15’ 12°00’E c 45° 45’ N 15 -40 24 Reno River 0 0 32 15 -2 Po River Po Pila 20 19 18 17 16 15 14 13 12 -20 Venezia 36 45° 00’ Po River -2 Latitude 45° 15’ 40 36 32 28 24 20 16 12 12°30’E 13°00’E Longitude Oxygen surface data Fig. 6. Surface (a) salinity, (b) temperature and (c) dissolved oxygen distribution around the Po River delta mg l–1 45° 30’ Chioggia Adige River 45° 00’ Po River 54 5 4 0 -2 Latitude 45° 15’ 8 7 6 5 4 3 2 1 -20 Venezia Po Pila 44° 45’ 5 5 Po di Goro 4 6 -40 Reno River 5 44° 30’ Ravenna 6 7 Cervia 44° 15’ 12° 00’ E 12° 30’ E Longitude 13° 00’E open sea—heads southeast. They mark the main dispersion path into the sea of the Po River fine sediment inputs distributed from these 2 mouths; in fact, highly hydrated sediments indicate fine particles that had settled shortly before and were thus not greatly compacted. The thickness of this layer ranged from about 20 cm in the area near the Po River mouths to 1 or 2 cm near the farther border of the study area. In the subsurficial sediments (Fig. 8b), the water content was lower because of their older deposition and hence higher compactness. The grain-size fine-fraction distribution in the surficial sediments (Fig. 9) shows a prevalence of clay fractions (< 8 φ in the same areas in which highly hydrated sediments deriving from the Po River were deposited. The mineralogical analyses performed on surficial and sub-surficial sediment samples confirm the presence of (1) a silico-clastic facies (Dinelli et al. 1997) deriving from the Po River, in front of and south of the Po River delta, and (2) a carbonate facies, deriving from rivers further north. Serpentine and dolomite are among the minerals that can be used as tracers of these 158 Clim Res 31: 151–165, 2006 a 45° 45’ N b Salinity bottom data Temperature bottom data PSU 45° 30’ 42 40 36 32 28 24 20 16 12 10 -20 Chioggia Adige River Po Pila 37. 2 8 36. 16 17 13 16 16 -40 -40 17 Reno River .6 37 44° 30’ Reno River 17 Po di Goro 38 44° 45’ 20 19 18 17 16 15 14 13 12 0 Po di Goro Chioggia Adige River Po River .6 37 0 Po Pila °C -20 Venezia -2 45° 00’ Po River -2 Latitude 45° 15’ Venezia 18 .2 37 Ravenna Ravenna Cervia 44° 15’ Cervia 12°00’E c 45° 45’ N 12°30’E 13°00’E Longitude Oxygen bottom data Fig. 7. Bottom (a) salinity, (b) temperature and (c) dissolved oxygen distribution in the study area mg l–1 45° 30’ 45° 15’ Chioggia 2 3 0 -2 Latitude Adige River 45° 00’ Po River 8 7 6 5 4 3 2 1 -20 Venezia Po Pila Po di Goro 2 4 2 3 44° 45’ 3 Reno River 2 44° 30’ 3 Ravenna Cervia 44° 15’ 12° 00’ E 12° 30’ E Longitude 13° 00’ E -40 2 facies. The former prevails in the fine Po River inputs and the latter in the particulate of rivers north of the Po River (Nelson 1971, Dinelli et al. 1997, Tomadin & Varani 1998). The serpentine distribution, in both surficial (Fig. 10a) and sub-surficial (Fig. 10b) sediments, has higher concentrations in an area in front of and south of the Po delta and in another area further offshore, to the southeast; in a northeasterly direction, concentrations lower than those in the first 2 areas are found in an intermediate north-south belt. In this belt the dolomite concentration increases, marking the southward transition zone of northern river sediments. In the sub-surficial sediments, the serpentine distribution pattern shows that the belt with greater concentrations becomes more limited towards the coast, in front of the delta mouths, whilst to the southeast it is almost absent (Fig. 10b). The distribution of organic C (which includes both terrigenous and autochthonous organic matter) in surficial sediments (Fig. 11a) shows a high concentration zone in the area in front of the delta, which extends both northwards and southwards in the central areas 159 Frascari et al.: Po River anomalous flood event a b ISMAR ñ MARINE GEOLOGY 0 10 45° 30’ N 20 30 40 km ISMAR ñ MARINE GEOLOGY 0 10 40 km Chioggia Chioggia Latitude 30 Venezia Venezia 45° 00’ 20 H2O (%) -40 -40 44° 30’ 60 60 50 50 40 40 Ravenna 30 30 Cervia 20 12° 00’ E -10 Ravenna 20 -20 12° 30’ E 12°00’E 13° 00’E Cervia -10 12°30’E -20 13°00’E Longitude Longitude Fig. 8. Water content distribution in (a) surficial and (b) sub-surficial sediments around the Po River delta ISMAR ñ MARINE GEOLOGY 0 10 45° 30’ N 20 30 40 km Venezia Latitude Chioggia 45° 00’ CLAY (%) -40 70 60 50 40 44° 30’ 30 Ravenna 20 10 12° 00’ E Cervia -10 12° 30’ E -20 13° 00’ E Longitude Fig. 9. Clay content distribution in surficial sediments around the Po River delta of the northern Adriatic basin. The highest concentrations are found in the most easterly and southerly zone of the study area. In sub-surficial sediments, organic C has the same distribution pattern but with less pronounced southward and northward lobes (Fig. 11b). Furthermore, to distinguish the nature and origin of the organic matter, the organic C:total N ratio was considered: it indicates, for high values (> 9 to 10), a continental origin or an advanced degradation of organic matter, and, for low values (< 7 to 8), an autochthonous marine organic matter with limited degradation (Müller 1977, Müller & Mathesius 1999). The C:N ratio presents a zone in surficial sediments (Fig. 12a) with high values — representative of prevalently terrigenous organic matter — that is located in front of the river mouths and is extended further offshore than in sub-surficial sediments (Fig. 12b). Low surficial sediment C:N ratio values are present in areas far from the Po River mouths. Organic P shows a different distribution between surficial and sub-surficial sediment. In surficial sediment (Fig. 13a), the organic P presents 2 enrichment areas: one in front of the Po River delta with a southeast protuberance, and another offshore, characterised by a south-southwest extension of the limits of the muddy deposits distributed from the Po River. The subsurficial sediments have their maxima in the areas to 160 Clim Res 31: 151–165, 2006 a 0 10 45°30’ N Latitude b ISMAR ñ MARINE GEOLOGY 20 30 40 km ISMAR ñ MARINE GEOLOGY 0 10 Venezia 45°00’ Chioggia Chioggia SERPENTINE SERPENTINE -40 -40 -4 3.0% 2.5% 2.5% 2.0% 2.0% 1.5% 1.5% Ravenna 1.0% Ravenna 1.0% 0.5% Cervia 12°00’ E 30 40 km Venezia 3.0% 44°30’ 20 -10 0.5% -20 12° 30’ E 13° 00’ E -10 Cervia 12° 00’ E -20 12° 30’ E 13° 00’ E Longitude Longitude Fig. 10. Serpentine content distribution in (a) surficial and (b) sub-surficial sediments around the Po River delta a b ISMAR ñ MARINE GEOLOGY 0 10 45°30’ N 20 0 30 40 km Latitude 10 20 30 40 km Venezia Venezia Chioggia 45°00’ ISMAR ñ MARINE GEOLOGY Chioggia ORGANIC C ORGANIC C 1.5% 1.5% -40 44°30’ -40 1.3% 1.3% 1.1% 1.1% 0.9% 0.9% 0.7% 0.7% 0.5% Ravenna 0.3% 0.5% 0.1% 12°00’ E Cervia 0.3% -10 0.1% -20 12° 30’ E Longitude 13° 00’ E 12° 00’ E Ravenna -10 Cervia 12° 30’ E -20 13° 00’ E Longitude Fig. 11. Organic C content distribution in (a) surficial and (b) sub-surficial sediments around the Po River delta 161 Frascari et al.: Po River anomalous flood event a b ISMAR ñ MARINE GEOLOGY 0 10 45°30’ N 20 ISMAR ñ MARINE GEOLOGY 0 30 40 km 30 40 km Chioggia 45°00’ C/N C/N -40 11 44°30’ 20 Venezia Venezia Chioggia Latitude 10 -40 11 10 10 9 9 8 8 7 7 6 Ravenna 6 Ravenna 5 5 Cervia 4 12°00’ E -10 12° 30’ E Cervia 4 -20 13° 00’ E 12° 00’ E -10 -20 12° 30’ E Longitude 13° 00’ E Longitude Fig. 12. C:N molar ratio distribution in (a) surficial and (b) sub-surficial sediments around the Po River delta a 0 10 45°30’ N Latitude b ISMAR ñ MARINE GEOLOGY 20 ISMAR ñ MARINE GEOLOGY 0 30 40 km 10 20 Venezia Venezia Chioggia Chioggia ORGANIC P –1 45°00’ (µg g ) ORGANIC P (µg g–1) 495 495 -40 445 44°30’ -40 445 395 395 345 345 295 295 245 245 195 195 145 Ravenna 145 Ravenna 95 95 45 12°00’ E 30 40 km Cervia -10 45 -20 12°30’ E Longitude 13° 00’ E 12° 00’ E Cervia -10 -20 12° 30’ E 13° 00’ E Longitude Fig. 13. Organic P content distribution in (a) surficial and (b) sub-surficial sediments around the Po River delta 162 Clim Res 31: 151–165, 2006 Table 1. Mean values (range) of parameters analysed in surficial and sub-surficial sediments collected during the PR94 cruise Parameter H2O (%) Silt (%) Clay (%) Serpentine (%) Organic C (%) C:N Organic P (µg g–1) Surficial sediments Sub-surficial sediments 47.2 (23.3–64.7) 42.8 (25.9–60.3) 36.5 (5.9–87.1) 37.6 (4.6–80.9)0 32.4 (0–71.5) 0 32.3 (0–69)0000 1.02 (0–4)0000 1.07 (0–3)00000 0.85 (0.2–1.5)0 0.86 (0.2–1.7)00 8.02 (4.5–12.2) 8.34 (5.1–12.9)0 245.3 (67.5–757.0) 239.2 (83.3–666.4) the south and offshore of the Po River delta (Fig. 13b). These areas are located in the same areas where organic C is also high. The maximum values of organic P in surficial and sub-surficial sediment are mainly distributed along the edges of the various plumes visible from the areal distribution of surface salinity and temperature (Fig. 6a,b). Table 1 shows mean values and ranges, in the surficial and sub-surficial sediments, of parameters discussed in the text. 5. DISCUSSION The surface water salinity is minimal near the Po River mouths and becomes gradually higher offshore for the continuous mixing of river and basin waters (Fig. 6a). The temperature transects (Figs. 3b, 4b & 5b) and bottom temperature areal distribution (Fig. 7b) highlight a vertical stratification in the water column. The stratification shows a division into 3 layers that are very clear in Transect A (Fig. 3b): (1) the surface layer presents cold waters coming from the Po River; (2) the intermediate layer is characterized by the warmest waters, which are still affected by the marine summer warming; and (3) the deepest and eastern layer again represents cold waters, corresponding to the dense water of the central northern Adriatic Sea characterised by low temperature and high salinity (Franco et al. 1982, Malanotte Rizzoli & Bergamasco 1983, Orel et al. 1993), which had remained confined to the bottom from the previous winter. The salinity and temperature surface stratifications are consequences of the fresh water from the Po River that floats offshore and fans out over the basin waters, mainly eastwards and southwards (Figs. 3, 4 & 6). The strong floating and the spreading eastward and southward of the fresher and colder river waters are consequences of the high speed and quantity of the Po River runoff in a basin characterised by low hydrodynamics, with no geostrophic currents and wave motion. In fact, the latter cause a disruption of the water column stratification when they become established in fall (Hendershott & Rizzoli 1976, Franco 1983, Franco et al. 1983, Fonda Umani et al. 1990, Artegiani et al. 1997). The dissolved oxygen concentration pattern could be used as an indicator of depositional processes of decomposing organic particulate, mainly from primary production. In fact, the low oxygen concentrations in the bottom waters are ascribed to oxygen consumption by the microbic degradation of reactive organic matter. These hypoxic conditions are more intense and extend upwards in the zones where the Po River and basin waters mix together (Figs. 3c & 4c). This is an indication that oxygen consumption is due mainly to the degradation of newly formed organic material settling on the bottom, the latter being generated by the surface algal blooms in the mixing zones of the Po River nutrient-rich waters with the warmer basin waters. This increase in surface primary production in the frontal zones of the river plume is confirmed by an increase in dissolved oxygen in the surface layer near to the front zones (Figs. 3c & 4c). The parameters measured in surficial sediments allow us to obtain direct information on depositional processes, as well as indirect observation and confirmation of the processes that took place in the water column when the survey cruise was carried out and the flood was underway. In contrast, the same parameters measured in the sub-surficial sediments allow us to deduce the processes that occurred in the study area, averaged over time, before the flood occurred. The muddy material of early deposition from the river plume is very rich in water, because it is the result of flocculation that occurs when fresh water mixes with salt water (Olsen 1982). Generally, the flocculated material is deposited in the proximal part of the prodelta area (Barbanti 1989). In our case, owing to the fact that fresh river water was floating on salty basin water, the fine and flocculated material contained in the floating water was spread out to sea in directions of river water distribution currents, and deposited all over the area covered by the flood plume as more hydrated material. Indeed, the water content distribution map of the surficial sediments (Fig. 8a) can be seen as traces of the flood plumes of the Po della Pila and Po di Goro mouths, which develop southward and south-southeastward. The origin of these zones with high water content is confirmed by the sub-surficial sediments, which are less hydrated over the whole area, having not been influenced by recent floods. Furthermore, the elevated thickness of the flood layer (between 20 cm near the Po River mouths and 1 to 2 cm near the farther borders of the study area) indicates a very high particulate load coming out from the Po River during the flood event. Frascari et al.: Po River anomalous flood event These observations are also confirmed by the clay and serpentine surficial distributions (Figs. 9 & 10a), which show—while deposition is underway— a higher offshore extension, which is unlike that under normal conditions when higher geostrophic currents occur and the fine Po River load is deposited near the coast. The spatial irregularities of the serpentine surficial distribution and, in particular, the lobed northward extension, in coincidence with the northward protuberance of clay and organic C (as traces of plume extension), indicate the low strength of general circulation currents during the flood that allowed the formation of an anti-cyclonic secondary gyre. In contrast, the organic C distribution in the surficial sediments (Fig. 11a) and its relationship with surficial N concentrations (Fig. 12a) highlights algal bloom process formation that is not related to the flood plume distribution pattern. The maximum organic C values, concomitant with low C:N ratios encountered in the zones most easterly and southerly of the Po River prodelta, may be explained by the deposition of organic matter of primary production. The production of organic matter is fostered by the mixing of nutrientrich river waters with the warmer basin waters along the plume front, as can be seen by the high oxygen value distribution pattern; this mixing also favours the depositional processes in the front zones due to the disappearance of the pycnocline, which maintains the particulate in suspension (Franco et al. 1989). Depositional processes are also favoured in the frontal area owing to the retarded intruding river currents, which reach minimum velocities in this region. The distribution of organic P in surficial sediments (Fig. 13a) confirms the presence of newly formed organic matter at the plume front. Furthermore, the high organic P concentrations, coinciding with high organic C contents and concomitant with low C:N ratios in surficial sediments, are found in areas of low bottom oxygen concentrations — namely where newly produced degrading organic matter is settling. Organic P and organic C are thus excellent tracers, not only for identifying zones affected by the flood plume but also for defining the areal limits of the plume itself and of the main processes of primary production in fall. These processes produce, in the presence of a stratified water column, a general hypoxia or anoxia in the dense bottom waters. The distribution of the C:N ratio in surficial sediments (Fig. 12a) shows high values in a zone near the river mouths, which extend southwards; these high values indicate the zones subject to the deposition of organic matter of continental origin, as a result of the early deposition of coarser material. In light of our findings deduced from the surface sediment parameter distributions and the temperature 163 and salinity distributions in the water column, it is possible to reconstruct the depositional processes of the Po River suspended particulate during the 1994 flood. The difference in salinity between the surface layer and the deeper basin waters, the high velocity of the river water inputs (rich in suspended solids) and the low hydrodynamics of the basin have favoured the buoyancy of the fresh river waters over the salty basin waters that are spread out towards the central basin areas. This has increased the surface transport of particulate coming from the river, which has deposited along the plume dispersion axes and particularly at the fronts where fresh water and salt water mix and where the reducing current speed, the disappearance of the pycnocline and the higher weight of the flocculated material favour the settling of the particulate. In this manner, only a small portion of the finer suspended particulate of river origin has undergone a process of early deposition that usually occurs just outside the Po River mouths as a result of flocculation processes initiated by the mixing of fresh and salt water. 6. CONCLUSION It can be seen from the PR94 cruise data and also from the literature (Hopkins et al. 1999, Hopkins 2001) that the geostrophic north-south current, generally present in this area, was very slow during the 1994 Po River flood and, therefore, that the typical cyclonic autumn-winter Adriatic circulation had not yet set in. This is deduced in particular from the strong temperature and salinity stratification shown in the landoffshore transects — in fact, the north Adriatic Sea water column stratification is disrupted when the N-Ad current is present (Hendershott & Rizzoli 1976, Franco 1983, Franco et al. 1983, Fonda Umani et al. 1990, Artegiani et al. 1997). The fresh and cold waters coming from the Po River spread out offshore, floating above the basin waters and undergoing subjacent and frontal mixing processes at the pycnocline limit. The main fresh water branch, coming out from the Po della Pila (as deduced from sediment and water column data), tends first to be distributed directly offshore, as far as ~40 km from the coast, and then to split into 2 branches: the main branch proceeds southwards while the less pronounced branch heads northwards. Waters coming from mouths south of the Po della Pila are distributed southeastwards and south-southeastwards. The low hydrodynamic of the basin together with the high speed and high amount of Po River runoff favour the large extension of the Po River plume into the basin, The Po River water distribution pattern in the basin is confirmed by a comparison of the parameters mea- 164 Clim Res 31: 151–165, 2006 sured in the surficial and sub-surficial sediments. The distribution of high values of water content, serpentine, organic C and organic P in the surficial sediments (compared to sub-surficial sediments) in areas extended southeastwards and eastwards mark the fresh deposition of terrigenous particulate. In particular, the area occupied by the main plumes and the frontal masses of flood waters can be superimposed on the fine highly hydrated surficial sediments. The elevated thickness and extension of the freshly deposited hydrated sediment shows that water coming from the Po River during the flood must have been particularly rich in suspended particulate. Furthermore, because of the considerable concentration of organic matter (high organic C) with a low C:N ratio and high organic P content, whose maxima were found in the bottom sediments in areas at the frontal zones, it is hypothesised that water coming from the Po River flood was rich in nutrients. In these areas, the fresh and cold nutrient-rich river waters mix on the surface with the saltier and warmer basin waters, and this favours the growth of algal blooms. The formation of algal blooms in the surface waters along the plume fronts, and their subsequent remineralisation within the water column through the pycnocline down to the bottom waters and the bottom sediments, is confirmed by the (1) gradually decreasing dissolved oxygen concentration from bottom water to the surface plume fronts, and (2) accumulation of organic matter with low C:N ratio and high organic P contents in the surface sediment of the area located at the plume fronts. The reduced oxygenation of the confined bottom waters is increased by the fresh organic mater mineralization processes. The persistence of the stratification of the water column after this event (Regione Emilia Romagna 1994) has led to a generalised anoxia of the bottom waters of the central Adriatic basin. The anoxia of the bottom water was resolved only with the early fall storms and with the onset of the general winter Adriatic Sea circulation, which disrupted the water column stratification. In contrast, the lack of zones containing high concentrations of organic C, a low C:N ratio and high concentrations of organic P in the sub-surficial sediments indicate that, in the medium term, the particulate that accumulates in front of the Po River delta is reworked and redistributed by the currents and wave motion in more extensive areas, especially to the south. In this way, the Po River prodelta bottom sediments act as temporary reservoirs of autochthonous organic matter and inorganic terrigenous particulate. These solid materials are then redistributed to other parts of the basin located further south, after degradation processes of the more reactive organic matter had already partially taken place. Acknowledgements. This work was conducted under the PRISMA I Project program. We thank Dr. Cristina Bergamini and Dr. Barbara Garrone for their field and laboratory support and for advice on interpretation, Vladimiro Landuzzi, Gabriella Rovatti, Angelo Magagnoli and Maurizio Mengoli, for their analyses, and Dr. Fabio Zaffagnini for help in constructing figures. This paper is Contribution No. 1380 of ISMAR, Sezione di Geologia Marina, CNR, Bologna, Italy. LITERATURE CITED Artegiani A, Azzolini R, Salusti E (1989) On the dense water in the Adriatic Sea. Oceanologica Acta 12:151–160 Artegiani A, Bregant D, Paschini E, Pinardi N, Raicich F, Russo A (1997) The Adriatic Sea general circulation. Part II: baroclinic circulation structure. J Phys Ocean 27:97–114 Aspila KI, Agemian H, Chau ASY (1976) A semiautomated method for the determination of inorganic, organic and total phosphate in sediments. Analyst 101:187–197 Autorità di Bacino del Fiume Po (1994) Indicazioni per lo sviluppo del piano di bacino conseguenti all’evento alluvionale del 4–6 Novembre 1994. Autorità di Bacino del Fiume Po, Parma Barbanti A (1989) Comportamento geochimico del fosforo e di alcuni metalli pesanti in ambiente fluviale e marino: indagini alla foce del fiume Po. PhD thesis, University of Bologna and Modena Barbanti A, Bergamini MC, Frascari F, Miserocchi S, Ratta M, Rosso G (1992) Early diagenesis and nutrient benthic fluxes in an area South of Po River Delta, Northern Adriatic Sea, Italy. Rapp Comm Int Mer Méditerr 33:321 Berner RA (1980) Early diagenesis: a theoretical approach. Princeton University Press, Princeton, NJ Brambati A, Ciabatti M, Fanzutti GP, Marabini F, Marocco R (1983) A new sedimentological textural map of the Northern and Central Adriatic Sea. Boll Ocean Teor Appl 1: 267–271 Colantoni P, Gallignani P, Lenaz R (1979a) The sediments of the Po river delta and the adjacent continental shelf. Rapp Comm Int Mer Méditerr 25/26:127–128 Colantoni P, Gallignani P, Lenaz R (1979b) Late Pleistocene and Holocene evolution of the north Adriatic continental shelf Italy. Mar Geol 33:41–50 Cook HE, Johnson PD, Matti C, Zemmels I (1975) Methods of sample preparation and X-ray diffraction data analysis. X-ray mineralogy laboratory, University of California, Riverside, Initial Report DSDP 28:999–1007 Dal Cin R (1983) I litorali dal delta del Po alle foci dell’Adige e del Brenta: caratteri tessiturali e dispersione dei sedimenti, cause dell’arretramento e previsioni sull’evoluzione futura. Boll Soc Geol 102:9–56 Dinelli E, Frascari F, Marcaccio M, Matteucci G, Roffi C, Spagnoli F (1997) Caratteristiche composizionali dei sedimenti del Mare Adriatico: alcuni risultati del progetto di ricerca PRISMA I PLINIUS. Suppl Ital all’Eur J Mineral 18:99–101 Fonda Umani S, Franco P, Ghirardelli E, Malej A (1990) Outline of oceanography and the plankton of the Adriatic Sea. In: Colombo G, Ferrari I, Ceccherelli VU, Rossi R (eds) Marine eutrophication and population dynamic. Proc 25th Eur Mar Biol Symp. Olsen & Olsen, Fredensborg, p 347–366 Franco P (1983) L’Adriatico settentrionale: caratteri oceanografici e problemi. Atti V Congresso AIOL, Stresa, p 1–27 Franco P, Jeftic L, Malanotte Rizzoli P, Michelato A, Orlic M (1982) Descriptive model of the Northern Adriatic. Oceanol Acta 5:379–389 Frascari et al.: Po River anomalous flood event 165 Franco P, Rabitti S, Cioce F (1989) Adriatico settentrionale in condizioni di stratificazione. 2. Distribuzione del materiale sospeso e delle proprietà ottiche (1983–1984). Boll Oceanol Teor Appl NS:77–87 Frascari F, Bergamini MC, Marcaccio M, Matteucci G (1998) Relazione conclusiva. Sottoprogetto ‘Flussi da e verso i fondali’ (PRISMA 1). ISMAR Tech Rep, Sezione di Geologia Marina, Bologna Froelich PN (1980) Analysis of organic carbon in marine sediments. Limnol Oceanogr 25:242–248 Froelich PN, Klinkhammer GP, Bender M, Luedtke NA, Hartmann B, Mainard V (1979) Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxis diagenesis. Geochim Cosmochim Acta 43:1075–1090 Gallignani P, Magagnoli A (1972) Metodologie e tecniche di sedimentologia fisica. Rapp Tec No. 1, CNR Laboratorio per la Geologia Marina, Bologna Giordani P, Angiolini L (1983) Chemical parameters characterizing the sedimentary environment in a NW Adriatic coastal area (Italy). Estuar Coast Shelf Sci 17:159–167 Griffiths EJ (1973) Environmental phosphorus handbook. John Wiley & Sons, New York Guerzoni S, Frignani M, Giordani P, Frascari F (1984) Heavy metals in sediments from different environments of a northern Adriatic Sea area, Italy. Environ Geol Water Sci 6:111–119 Hedges JI, Stern JK (1984) Carbon and nitrogen determinations of carbonate-containing solids. Limnol Oceanogr 29: 657–663 Hendershott MC, Rizzoli P (1976) The winter circulation of the Adriatic Sea. Deep-Sea Res 23:353–370 Hopkins T (2001) Abiotic variabily and biocomplexity in the northern Adriatic Sea, some research perspectives. Biol Mar Med 8:39 Hopkins T, Artegiani A, Kinder A, Pariente R (1999) A discussion of the northern Adriatic circulation and flushing as determined from the ELNA hydrography. In: Hopkins TS, Artegiani A, Cauwet G, Degobbis D, Malej A (eds) The Adriatic Sea, ecosystem research report No. 32. European Commission, Brussels, p 85–107 IRSA (1977) Indagine sulla qualità delle acque del Po. Quaderno 32. Istituto di Ricerca Sulle Acque, Milano Malanotte Rizzoli P, Bergamasco A (1983) The dynamics of the coastal region of the northern Adriatic Sea. J Phys Oceanogr 13:1105–1130 Marchetti R (1983) Quadro di sintesi delle indagini svolte dal 1978 sul problema dell’eutrofizzazione nelle acque costiere dell’Emilia Romagna. In: Eutrofizzazione dell’Adriatico. Ricerche e Linee d’Intervento, Regione Emilia-Romagna, Bologna, p 39–75 Marchetti R, Baraldi O, Crosa G, Garibaldi L, Varallo A (1996) I carichi di fosforo e azoto del Po durante la piena del novembre 1994. Acqua Aria 5:499–503 Müller PJ (1977) C/N ratios in Pacific deep-sea sediments: effect of inorganic ammonium and organic nitrogen compounds sorbed by clays. Geochim Cosmochim Acta 41: 765–776 Müller A, Mathesius U (1999) The paleoenvironments of coastal lagoons in the southern Baltic Sea, I. The application of sedimentary C org/N ratios as a source indicators of organic matter. Palaeogeogr Palaeoclim 145:1–16 Nelson BW (1971) Mineralogical differentiation of sediments dispersed from the Po Delta. In: Stanley DJ (ed) The Mediterranean Sea: a natural sedimentation laboratory. Dowden, Hutchinson & Ross, Stroudsbury, PA, p 441–453 Olsen CR, Cutshall NH, Larsen IL (1982) Pollutant-particle association and dynamics in coastal marine environments: a review. Mar Chem 11:501–533 Orel G, Aleffi F, Umani Fonda S (1993) Ipossie e anossie di fondali marini. L’Alto Adriatico e il Golfo di Trieste. Regione autonoma del Friuli Venezia Giulia, Trieste, p 46–77 Orlic M, Gacic M, La Violette PE (1992) The currents and circulation of the Adriatic Sea. Oceanol Acta 15:109–124 Regione Emilia Romagna (1994) Rapporto sullo stato di eutrofizzazione delle acque costiere dell’Emilia Romagna nel 1994. Assessorato Territorio, Programmazione e Ambiente. Servizio Promozione, Indirizzo e Controllo Ambientale U.O.O. DAPHNE II, Cesenatico Smodlaka N, Revelante N (1983) The trends of phytoplankton production in the Northern Adriatic Sea; A twelve year survey. Rapp Comm Int Mer Mediterr 28:89–90 Strickland JDH, Parsons TR (1972) A practical handbook of sea-water analysis. Fisheries Research Board Canada, Ottawa Tomadin L, Varani L (1998) Provenance and downstream mineralogical evolution of the muds by the Po River (Northern Italy). Miner Petrogr Acta 61:205–224 Trincardi F, Corregiari A, Roveri M (1994) Late Quaternary trasgressive erosion and deposit in a modern epicontinental shelf: the Adriatic semienclosed basin. Geo-Mar Lett 14:41–51 Submitted: November 21, 2004; Accepted: March 20, 2006 Proofs received from author(s): July 11, 2006
© Copyright 2024 Paperzz