THE DARK SIDE OF THE ALBANO CRATER LAKE

The Albano Lake is the deepest volcanic lake among the volcanoes located in the Italian peninsula. It belongs to the Colli Albani volcanic complex whose last largest eruptions are dated back to about ~30 Kyr, although minor events likely occurred during historical times at 7000 yr B.P. or earlier. After the end of the volcanic activity the Crater of Albano became a lake whose level changes are known since historical times. On November 2005, was performed the first very high resolution bathymetric survey of the Albano lake by means of a multibeam echo sounder, integrated with the GPS/RTK positioning technique A particular effort was devoted to produce a high resolution morphobathimetric map, which aims to provide a Digital Terrain Model of the lake floor for wide applications. The surveys did not revealed significant gas exhalative centres, which should indicate a current active gas release from the lake floor. Here we show the technical details of the bathymetric surveys, the very high resolution bathymetric map and the main morphological features of the Albano Lake bottom.


Introduction
The Colli Albani volcanic complex occupies a wide area about 25 km SE of Rome.Its general structure is a caldera with a central cone, although this complex displays two nested calderas and several more or less eccentric post-caldera vents, most of which produced by explosive activity.The highest point is Monte Cavo at 949 m, which consists in a scoria cone located in an eccentric position on the SW rim of the younger Faete caldera.The two crater lakes of Albano and Nemi fill the most recent craters of the volcano (Fig. 1).
The former geological studies performed by Fornaseri et al. (1963) andDe Rita et al. (1988), dated all deposits of the Colli Albani at an age older than Holocene (>10,000 years) and until recent time, they have been considered an extinct volcano, although some historical documents reported some eruptive activity around 114 B.C. (Funiciello et al., 2002) and 7000 yr B.P. (Andretta and Voltaggio, 1988).In recent times, evidences of an ongoing volcanic unrest based on instrumental seismological and geodetic data, have been detected (Amato and Chiarabba 1995;Chiarabba et al. 1997;Anzidei et al. 1998), and new researches indicates that likely occurred an eruptive activity during the Holocene (Funiciello et al. 2003;Porreca et al. 2003), as also previously suggested by Andretta e Voltaggio (1998) and Villa et al. (1999).
Measurements of ground deformation available from high precision levelling lines established by IGM in 1951, as well as other benchmarks measured at the end of the last century, indicated un uplift at 30 cm in 43 years at a rate at ~0.7 mm/yr (Amato and Chiarabba, 1995) and the broad deformation zone evidenced through DinSAR observations by Salvi et al. (2004), mainly across the two lakes of Albano and Nemi, were interpreted as related to a superficial source (3-6 km), producing a signal similar to many active volcanoes (Amato and Chiarabba, 1995;Chiarabba et al., 1997;Salvi et al., 2004).Dangerous gas release have been observed from the ground in an area densely populated (Chiodini and Frondini 2001;Annunziatellis et al. 2003;Carapezza et al. 2003), is likely the result of fracturing produced during the seismicity of 1989-1990and 1995, 1999(Beaubien et al. 2003).
From regional tectonic evidences and long-term behaviour of the volcanic complex, Karner et al. (2001a, b) and Marra et al. (2003) estimated that the volcano could be at the beginning of a new eruptive phase.The existence of a potentially active volcano so close to the centre of Rome and other minor but not less important villages, all densely inhabited, is now leading scientists to revise the volcanic hazards of this area not only for the scenario of new eruptions but also for the recurrent seismicity that periodically strikes the volcano (Amato et al. 1994) and of the possibility of dangerous sudden gas release, rich in CO 2 , from the bottom of the lakes of Albano and Nemi, as in case of the catastrophic events occurred in the African crater lakes of Monoun (Sigurdsson et al., 1987) and Nyos (Barberi et al.1989;Rice, 2000).
The Albano Lake, also known as the Castelgandolfo Lake, is located at 293 m above sea level and it is the deepest among the volcanic crater lakes of Italy, being nowadays 167 m deep.Presently, it is 3.5 km long and 2.3 km wide with an extension of about 6 km 2 .It was settled since pre-historical times (Meli, 1911;Ryves et al., 1996;Manca et al., 1996;Lowe et al., 1996) and during the roman epoch became a place of greater importance (Carandini, 1997).In recent times, this nice and quiet place frequented by tourists, is the summer residence of the Pope at Castel Gandolfo village, located just on the top of the crate rim which contains the Albano Lake.Due to its frequent level changes (Marra and Karner, 2005), produced by deep water circulation and a likely catastrophic overspill, occurred in 398 B.C., induced the Romans to excavate an artificial outlet, to control lake level (Funiciello et al., 2002(Funiciello et al., , 2003)).
Based on these data, under the umbrella of the Italian Dipartimento della Protezione Civile, was started a multi parametric study of the Colli Albani volcano, including a high resolution bathymetry of the groundfloor of the Lake of Albano, still not yet investigated by such surveys.This technique is capable to produce 2-D and 3-D images of the morphology of submerged volcanic areas, useful for wide applications, including hazard estimation (Anzidei, 2000;Anzidei et al., 2005;Esposito et al., 2006).

Bathymetric surveys
A multibeam high resolution survey that covered the whole area of the lake was performed for the first time in November, 2005, using the Alsea boat of Coastal Consulting and Exploration Company (Fig. 2), equipped with ultra high resolution multibeams and survey instrumentation (Fig. 3).
Particularly, due to the relevant depth of the lake, were used an ultra high resolution Reson Seabat 8125 multibeam (250 beams, 0.5°x1.0sector coverage, 455 Khz) up to 80 m depth and a Reson Seabat 8101 multibeam (101 beams, 0.5° x 0.5° sector coverage, 455 Khz), in the depth range 80-167 m, up to the deepest point of the lake (Tab.1).
Before starting the surveys, a check of the health of the GPS/RTK data link was performed trough the planned survey.The bathymetric datum was established by measuring the water level through some GPS measurements along the shore of the lake.The instrumental height of the zero level was referred to a WGS84 geodetic benchmark (named ALBA) previously set up close to the lake (Fig. 1; Tab.2).The latter was measured by geodetic space techniques using a couple of dual frequency GPS receivers, with reference to the GPS geodetic monument of INGR, located at INGV in Rome (Fig. 1), which belongs to the National GPS network of the INGV, whose 3-D coordinates are known at a few mm level (Anzidei et al., 1998;Serpelloni et al., 2005).Elevation data of lake floor were thus given into the WGS84 reference system (ellipsoidal heights) because the reference benchmark ALBA was not linked to any levelling line and the hortometric elevation were not available for this benchmark.
The centimetric positioning of the vessel was computed by GPS technique in RTK mode during surveys.Real time coordinates were obtained by installing an Ashtech Aquarius reference station located on the GPS station ALBA and transmitting the differential corrections by a High Frequency link at 1 Hz rate to the mobile Aquarius GPS receiver, placed on board of the vessel.In addition to this, a Sg-Brown Meridian Surveyor gyrocompass, a Tss DM505 MRU and a Fugro Omnistar Differential GPS, were interfaced to the Reson PDS2000 Navigation software for data acquisition (multibeam and positioning), control, calibration and pre-processing.An SBE 37-SI Microcat CTD probe was located at the sonar head and interfaced to the sonar processor, providing the real time speed of sound data for the beam forming, whereas the Navitronics SVP15 and a SeaBird CTD probes were also used for profiling the temperature, conductibility and the speed of sound along the water column (Fig. 6).Additional details on the employed instrumentation are reported in Table 1.
Navigation routes (Fig. 4) were performed in order to achieve the full coverage of the lake bottom, with at least 20-30% overlap of the nearby swaths.The Reson PDS2000 software was able to show in real time to the operator and the pilot the ongoing multibeam and Digital Terrain Model (DTM) and the positioning information that were used for guidance.
At the beginning of each survey, a full set of multibeam calibration lines were acquired, on flat bottoms and steep targets at about 30 m water depth.The roll, pitch and yaw correction angles were then used to correct the installation geometries.Calibration parameters were then taken into account during data analysis to correct the observations.

Data analysis and bathymetric map
Data analysis was performed by the Computer Aided Resource Information System -Hydrographic Information Processing System (CARIS -HIPS) PRO V5.2 software, specifically designed to process multibeam data under Windows NT ® and capable to manage images in a mosaic of the lake floor and produce raster and analytical maps.The processing sequence was as follows: 1. system calibration and multibeam data reprocessing; 2. data quality check: low quality data were discarded due to a not optimal signal to noise ratio; 3. lake level correction, using the RTK data; 4. High and medium frequency spike removal, but keeping intact eventual signatures produced by uprising gas bubbles from the lake floor.;

Production of high accuracy Digital Elevation Models (DEM).
To produce the MDEM, were used a total of 1.466.914 million of 3-D punctual data (Latitude, Longitude and depth), that were converted in the UTM33-WGS84 coordinate system.
The survey data set was analysed to reduce any positioning offset or error in the MDEM together with the analysis of the standard deviations of the mutibeam data,.The latter show values ranging between 10-15 cm up to depths of -20m; 15-30 cm at depths between -20 and -50 m and 30-50 cm at depths greater than -50 m.
Once the offsets and errors were analysed and corrected, the final MDEM was produced and made available for the morphostructural analysis, production of contour maps at 1:2.500 scale (Fig. 7) and shaded reliefs (Fig. 9,10), which show the roughness and complexity of the crater of the Albano lake.

Discussion and conclusions
High resolution multibeam technique provided the first 3-D detailed morphobathymetric map of the Albano Lake at < 1 m average pixel resolution (Fig. 7).These data are useful for a wide range of applications and to improve and support the geological, geomorphological, volcanological, geochemical, geophysical research and monitoring of this volcanic area.
As far as the a volcanological interpretations is not the specific goal of this paper, which aims to describe the technique used and to shows the first images of the submerged part of the Albano Lake crater, these new data can provide information on the still unknown morphological features of the submerged part of the crater.The groundfloor show the past episodes of the lake's history, strictly connected with the geological and volcanological evolution of the area.Surface features due to volcanic activity, recent surface and lake level changes and sliding or rock fall events, are evident from the data.
The first main results obtained from the bathymetric surveys can be summarized as follows: • The total water volume of the lake at the time of the surveys is 447.495.490m 3 .
• The lake is characterized in its northern side by a flat area at depth between 0 and 25 m below lake surface and by two concentric circular basins bordered by steep flanks, which can be addressed to crater rims.The first one is between about -50 and -120 m; the second between -120 and -167 m. • The deepest point of the lake is at -166.86 m, lower than the previous measurements ( 173m).This value could be explained with the lowering of the water table around the basin which can reduce the hydrologic balance of the area, as reported by Capelli et al., (2000).
• The deepest point coincides with a circular crater 1000 m wide, with steep inner flanks about 45 m height.The steep flanks shows erosion phenomena, likely sub aerial, that could be occurred before the crater was filled with water.The flat floor of this area suggest a continuous sedimentation at this depth in agreement with Oldfield et al. (1996).
• Other sub circular depressions size could be addressed to craters.But without further evidences from seismic soundings or drillings, with the exception of those reported in Oldfield et al. (1996), we cannot confirm or exclude this hypothesis.
• Slides or rock fall of different size occurred in the lake since its formation.They mainly occurred along the steepest inner flanks of the crater and partially in the central crater.
• Three levels of submerged shores, at the moment of unknown age, are clearly located at depths between -31 and -41 in the north-eastern side of the lake.They witness the past lake level standings at these heights likely due to change of the environment as also reported by Chondrogianni et al., (1996), Ryves et al. (1996), Lowe et al. (1996) and Marra and Karner (2005).
• Data did not revealed any relevant gas exhalative centres in the whole basin.This is in contrast with previous observations (Oldfield at al., 1996) that evidenced some exhalative points mainly in the eastern side of the basin.This can be explained with a temporary change of the gas exhalation from the ground or with the sealing of fractures which prevent the gas release up to the surface.
The issues related with the occurrence of slides and gas exhalative points are relevant for scientific discussion and hazard assessment of the Albano Crater Lake volcano and its bathymetry open new questions on its recent evolution, thus suggesting further investigations through the integration of different geological and geophysical studies.Mainly, if the central deep crater can be a suitable trap for CO 2 accumulation that can be suddenly released, as occurred in lake Nyos (Rice 2000), taking into account that the Albano lake has the highest CO2 concentration among the Italian crater lakes (up to 200 mg/l at -175 m) (Martini et al., 1994).In such case, it should be subjected to a water rollover with dangerous consequences of gas or hot fluids release from the deepest part of the lake.
Concerning slides, a slide hazard must be taken into account because their occurrence would induce tsunami lake waves, dangerous for humans and the environment.Some cases of tsunamis triggered by slides in closed basin, are reported in the literature such as that occurred in the artificial lake of Vajont, Italy, in 1963 which produced large desctruction and about two thousand of victims.So far, the occurrence of such events in the Albano lake should not be excluded due to the seismicity of the area, the features of the lake floor and to the steep slopes of the inner side of the crater facing the lake.The morphological features of the lake floor, suggest the existence of at least two larger craters and three more coalescent smaller eruptive centres (Anzidei et al., 2006).Further surveys, such as seismic soundings and sub bottom profiles, should be performed to provide data on the still unknown structural features of the lake floor.

Figure caption
Fig. 1 Regional setting of the Colli Albani volcanic complex (from De Rita al., 1992, modified).Key: 1=travertine; 2=Plio-Pleistocene sedimentary units; 3="final" hydromagmatic units; 4=air fall deposits; 5=lava flows; 6=pyroclastic flow units of the Colli Albani; 7=pyroclastic flow units of the Sabatini volcanic field (in northwestern part of the map); 8=Tortonian flysch; 9=caldera rims;    were collected above the central crater (deepest point of the lake) and were used to calibrate the multibeam system for sound speed velocity in the water to determine depths at 1 cm average formal accuracy.

Table caption
Tab.1 Instrumentation used during the bathymetric surveys.
Fig.2The ALSEA vessel used during bathymetric surveys of the Albano lake.

Fig. 3
Fig.3 Sketch of the instrumentation used during surveys Fig.4 Course over the ground performed by ALSEA vessel during bathymetric surveys.The 30% overlapping between nearby lines guarantee the full coverage of the area.

Fig. 5
Fig.5 sketch of the swath with the multibeam head in a) vertical and b) rotated positions to collect data up to the lake surface.

Fig. 6
Fig.6 a) Temperature and b) conducibility vertical profiles from lake surface to the bottom.Data

Fig. 7
Fig.7 Digital Elevation Model (DEM) showing the morphobathymetry of the area.Scale colour shows depths ranging between 0 and -167 m a) shaded relief b) at 5.0 m countour level.

Fig. 8
Fig.8 Multibeam image of the central crater Fig. 9 a,b,c,d Multibeam images of slides located along a,b) the north eastern side and c,d) the southern side of the crater, respectively.In 9d are also reported the isobaths to evidence the morphological changes in the slide area.