Forensic Science International 229 (2013) e6–e12

Case report

An investigation on body displacement after two drowning accidents

Marcos Mateus a,*, Hilda de Pablo a, Nuno Vaz b

a MARETEC, Instituto Superior Tecnico, Universidade Tecnica de Lisboa, Av. Rovisco Pais, 1049-001 Lisboa, Portugalb CESAM, Departamento de Fısica, Universidade de Aveiro, 3810-193 Aveiro, Portugal

A R T I C L E I N F O

Article history:

Received 2 August 2012

Received in revised form 28 February 2013

Accepted 6 March 2013

Available online 8 April 2013

Keywords:

Drowning

Drifting bodies

Accumulated degree days

Post-mortem submersion interval

Oceanography

Forensic oceanography

A B S T R A C T

The finding of human remains in aquatic environments is usually attributed to causes such as work-

related or recreational accidents, suicides, discarded homicide victims, and natural disasters. When the

point and date of entry in the water is unknown, these findings pose serious challenges to forensic

analysis given the difficulty to estimate the drift of the body. In this context, the information retrieved

from cases where the point of entry and body recovery sites are known, as well as the timing, is

significant. Two drowning accidents in marine coastal waters were analyzed. In both cases the post-

mortem submersion interval (PMSI) is known, as well as the accident (point of entry) and body recovery

sites. Accumulated degree days (ADD) was estimated in both cases using satellite sea surface

temperature data. In both cases the bodies were recovered in the vicinity of the accident site (�2 km in

case 1 and less than 1 km in case 2). Results were interpreted in terms of oceanographic conditions,

physical settings and ADD. The results provide some relevant clues on the fate of human cadavers in

coastal marine environments that can be used by officials and agencies involved in the recovery of

bodies, as well as by forensic investigators when dealing with these findings.

� 2013 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at SciVerse ScienceDirect

Forensic Science International

jou r nal h o mep age: w ww.els evier . co m/lo c ate / fo r sc i in t

1. Introduction

The finding of human remains in aquatic environments is notuncommon. The most significant causes for such findings can beattributed to work-related accidents such as boat disasters andman overboard occurrences [1–3], recreational accidents [2,4–6],suicides [7–9], discarded homicide victims [5,10–12], and naturaldisasters such as floods and tsunamis [13].

A significant number of water-related deaths are usuallycharacterized by uncertainties related to the cause of death, pointof entry in the water and post-mortem submersion interval (PMSI).The challenge to forensic analyses starts with the determination ofthe circumstances surrounding death [4]. Usually an autopsy willsuffice to establish the manner of death, and artifactual alterationof tissue structure and microscopic features may provide clues onthe PMSI [14–16]. Determining the place where the body enteredthe water or, in some cases, predict the drift trajectory afterdrowning accidents, might prove difficult given the complexdynamics of some aquatic systems, and its interaction with thedecomposition process of the body.

Understanding the horizontal displacement of a body afterdrowning requires knowledge on the decomposition process, sincethis has a direct influence in the vertical displacement. Vertical

* Corresponding author. Tel.: +351 91 9865029; fax: +351 21 8419423.

E-mail addresses: mmateus.maretec@ist.utl.pt (M. Mateus),

hildadepablo@ist.utl.pt (H. de Pablo), nuno.vaz@ua.pt (N. Vaz).

0379-0738/$ – see front matter � 2013 Elsevier Ireland Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.forsciint.2013.03.010

movement of a body depends on its specific gravity, the relationbetween the density of the body and the density of the surroundingwater [17]. The human body specific gravity is very similar to thatof water, which means that small variations in the specific gravityhave considerable effect on buoyancy [18]. These variations are thekey to understand horizontal displacement because they deter-mine if the body sinks to the bottom where current intensity isusually lower and friction forces higher, or if the body floats and istransported by surface currents.

Increasing buoyancy due to the formation of putrefaction gasesplays a crucial role in the drift behavior of human bodies in waterenvironments [19], because in time they enable the body to gainsufficient flotation force. Decomposition rate can be estimated bythe accumulation of thermal energy needed for the chemicalreactions and biological processes of decomposition to take place,frequently expressed in accumulated degree days (ADD) [20,21].Accurate information on the PMSI and ADD is only possible toobtain in cases where the places and times of drowning and bodyrecovery are known. Such occurrences also provide significantinformation on the mechanisms involved in the displacement ofthe body. Ultimately, the knowledge and experience gained bystudying these cases can be applied afterwards to similarsituations where information is scarce.

The present study aims to investigate two drowning cases incoastal marine waters where the PMSI is known, as well as theaccident sites (point of entry) and body recovery. The ADD iscalculated for both cases in an attempt to relate this parameterwith the timing of body recovery. Also, oceanographic features

M. Mateus et al. / Forensic Science International 229 (2013) e6–e12 e7

such as tidal propagation are taken into account in the displace-ment of the bodies.

2. The cases and area of study

Water-related accidents are usually conveyed in mediacoverage. Frequently newspaper clips can provide more timelydata on drowning accidents, such as approximate time andlocation, but they are also incomplete sources of information [22].Local Maritime and Port Authorities were contacted then forprecise information on the location and time of the accidents andbody recoveries.

2.1. Case 1

On 25 April 2012 around 1745 h (Western European SummerTime) an 8-year old girl was dragged by the sea when playing onthe sand at Memoria beach (41813.850 N; 8843.300 W), nearOporto, Portugal (Fig. 1A). The body was missing for approxi-mately 9 days, and was recovered after being spotted in the sandof Corgo beach (41815.140 N; 8843.470 W, Fig. 1B), a fewkilometer north from the accident site, at around 0815 h on 4May 2012.

Memoria and Corgo beaches are located on the NW Portuguesecoast, and they are on the same sand stretch with an almost N–Sorientation (Fig. 1B). The bottom is mainly sand but some rockyprotuberances can be found. Intense wave regime is frequent inboth beaches and surface currents are usually southward in late-spring and summer and northward during the rest of the year.Tides are semidiurnal and the tidal amplitude is approximately3 m, modulating the circulation in this region.

Fig. 1. Map of the areas of the Portuguese coast where the study sites are located (left pan

the accident (point of entry) and body recovery are marked with numbers 1 and 2, res

2.2. Case 2

On 9 June 2012 around 1620 h, a 17-year old boy disappearedwhile swimming at St Amaro de Oeiras beach (38841.050 N; 9818.70

W, Fig. 1D), near the mouth of the Tagus Estuary in Lisbon, Portugal(Fig. 1C). The body was found floating on 15 June 2012 around0650 h by fisherman less than 1 km away (38841.100 N; 9818.250

W) from the site of disappearance.The St Amaro beach is located in the mouth of the Tagus Estuary

(Fig. 1C), the largest estuary in the Iberian Peninsula. The beach hasa W–E orientation and is on the right margin of the estuary nearcoast. Tidal action, wave regime and river discharge control thecirculation and water temperature in this area. As a consequencethere is a significant change in the direction and intensity of localcurrents and water temperature. Tide is also semidiurnal,presenting tidal amplitude of about 3 m.

3. Methods

PMSI was calculated for both cases, since the initial time of submersion and body

retrieval are known. Based on the coordinates of the sites, total body drift distance

was estimated in the aerial imagery provided by Google Earth1 (Fig. 1B and D). Tidal

conditions for each site [23] were analyzed for the time of body recovery and

information on sea surface temperature (SST) was retrieved from satellite imagery

(daily values), and then used to calculate the ADD. The SST product used is based on

the night-time images collected by the infrared sensors mounted on different

satellite platforms, and correspond to the Mediterranean Sea Ultra High Resolution

Sea Surface Temperature Analysis (product ID: SST-MED-SST-L4-NRT-OBSERVA-

TIONS-010-004-a) available in the project MyOcean catalog (available at

www.myocean.eu).

3.1. Accumulated degree days (ADD)

ADD is typically calculated considering the duration of immersion and average

daily temperature, by summing the average temperature over each 24 h period (Ti)

els), and the respective close-up aerial image of each site (right panels). Locations of

pectively.

Fig. 2. Sea surface temperature (SST) from remote sensing data for 25 April–3 May 2012 (top row), and 9 June–15 June 2012 (bottom row). White triangle marks the accident

sites. Different temperature scales are used to highlight the temperature gradients. SST data retrieved from the MyOcean catalog product ID: SST-MED-SST-L4-NRT-

OBSERVATIONS-010-004-a.

M. Mateus et al. / Forensic Science International 229 (2013) e6–e12e8

for the number of days (n) that made up the PMSI [24]:

ADD ¼Xn

i¼1

Ti (1)

Considering that a single water temperature daily field is usually obtained from

satellite imagery for SST, and to account for the exceeding hours outside a complete

24 h cycle, we propose a slightly modified expression to calculate ADD:

ADD ¼Xn

i¼1

Ti �hi

24

� �(2)

where hi is the number of hours considered for each single day (i) during the entire

period (n), and the temperature value (Ti) corresponds to the daily satellite

temperature value for the site.

4. Results

PMSI was �8.6 days for case 1 and �5.6 days for case 2.Estimated body displacement assuming a linear path is around2.8 km and 0.7 km for case 1 and case 2, respectively. SST fromsatellite images in Fig. 2 for both cases shows little variation intime. The top panel depicts the SST for the north coast of Portugal,near Oporto, and the lower panel depicts the SST results for theTagus Estuary region, near Lisbon. In case 1 (top panel) the SST is

Table 1Information on the time of accident, body recovery, postmortem submersion interval

calculate according to Eq. (2) for each case. Values for total ADD have been round up.

Case 1

Accident: 25-04-2012 1745 h

Body recovery: 04-05-2012 0800 h

PMSI: 8.6 days

Day Accumulated time (h) Temp. (8C) ADD (8C)

25-04 6.1 13 3.3

26-04 24.0 13 13.0

27-04 24.0 13 13.0

28-04 24.0 13 13.0

29-04 24.0 14 14.0

30-04 24.0 14 14.0

01-05 24.0 14 14.0

02-05 24.0 14 14.0

03-05 24.0 14 14.0

04-05 8.0 14 4.7

Total 117

fairly constant, with values around 13–14 8C during the PMSI.During the PMSI in case 2 (lower panel) the temperature is alsoconstant and around 16–17 8C. For cases 1 and 2 the ADD was117 8C and 95 8C, respectively. Daily water temperature valuesused to calculate ADD are presented in Table 1, along with thesummary of the most relevant information. In both cases the bodywas found during low tide when tide elevation was at theminimum values, as seen in Fig. 3.

5. Discussion

Reports on body drift in the ocean usually focus on exampleswhere bodies have been recovered at significant distances from thesite of drowning [7,25,26]. Some extreme situations are reportedwith bodies being recovered hundreds of kilometers away from thepoint of entry [1,6]. The cases reported in this paper show a strikingdifferent pattern, in which the bodies have undergone littledisplacement, even after a significant PMSI (�8.6 days in case 1)under high tidal amplitude and wave regime. To understand thesemarked differences in the horizontal displacement, it is imperativeto understand the influence of vertical motion on the horizontaldisplacement.

(PMSI), surface water temperature (Temp.) and accumulated degree days (ADD)

Case 2

Accident: 09-06-2012 1620 h

Body recovery: 15-06-2012 0650 h

PMSI: 5.6 days

Day Accumulated time (h) Temp. (8C) ADD (8C)

09-06 7.7 17 5.5

10-06 24.0 17 17.0

11-06 24.0 17 17.0

12-06 24.0 17 17.0

13-06 24.0 17 17.0

14-06 24.0 17 17.0

15-06 6.8 17 4.8

Total 95

Fig. 3. Tidal elevation prediction (data from [23]) for the port of Leixoes during 4 May 2012 (left panel), and port of Lisboa during 15 June 2012 (right panel). Time of body

recovery marked with a black dot.

Fig. 4. Schematics for the main processes involved in the vertical and horizontal displacement of human cadavers in aquatic environments: (A) relation of specific gravity (SG)

with floatability; SG increases with water aspiration or ingestion, and decreases with insufflation caused by gas formation during the decomposition process; (B) forces acting

in a body in water: buoyancy, gravity (both affected by the SG), drag and friction (both affected by local currents); (C) phases of movement [17]: settling to the bottom, motion

along the bottom, ascent to the surface, drift along the surface. In warm and shallow water the body can rise to the surface within a few days, while deep and cold water it may

take as long as several weeks. Note that the dimension of the arrows does not represent proportional intervals of time. (D) Possible relation between water level (tide) and

time of resurface. In an advanced stage of the bloating phase when the SG starts to decrease, the gradual decrease in the hydrostatic pressure over the tidal cycle, followed by

gas expansion, may trigger the resurface stage. The black dot in the lower panel shows the water level in the tidal cycle for each diagram.

M. Mateus et al. / Forensic Science International 229 (2013) e6–e12 e9

M. Mateus et al. / Forensic Science International 229 (2013) e6–e12e10

5.1. Sink, float and body displacement

In both cases it is assumed that the victims died from drowning,which means that an increase in lung weight by water aspiration[18,25,27] or stomach weight increase by water ingestion mighthave occurred [28,29]. The result is an increase in the body specificgravity leading to decreases in floatability (Fig. 4A). Once the bodystarts to sink, the increase in the hydrostatic pressure willcompress gas containing cavities leading to the increase in specificgravity, making the body sink to the bottom [18,30]. After sinkingtwo major factors can explain why the horizontal displacementwas so reduced in these cases: increased bottom roughness causedby bed effects (rocks, debris, snags, etc.), and insufficient velocity tomove the body, which means that the drag forces cannot overcomegravitational and frictional forces (Fig. 4B). In case 1, the bed rocksmay have also conditioned the displacement by retaining the body(trapped near the bottom), and only once freed it floatedimmediately to the surface. The distance that the body istransported is a function of the water velocity. Even significantstrong currents of about 0.6 m s�1 are known to be insufficient togenerate sufficient drag force to overcome bottom friction [17].Apparently the strong alongshore currents observed in case 1 (insome cases >1 m s�1) for the studied period, and the tidal currentsin case 2 were not sufficient to move the bodies along the bottomfor greater distances.

The PMSI in these cases are somehow below the range presentedin other studies that mention bodies resurfacing after 12 days [24].However, these values are from water systems in the U.K., withrelatively lower temperature ranges (closer to temperatures in case1). The displacement in the two cases here discussed seems ratherinsignificant (2.8 km and 0.7 km for case 1 and case 2, respectively)when compared to other reported occurrences [1,6,26,31]. One ofthe reasons could be that the bodies were stranded (case 1),preventing further drift, or recovered moments after resurfacing(case 2). Apparently, significant body displacement only occurs afterresurfacing and if the body is not immediately recovered. In suchcases the body will drift away dragged by surface currents. Theprobability of this occurring increases in places less frequented bypeople, during night or under low visibility conditions.

It can be argued that in case 1 the body might have resurfacedafter what it has been subject to surface transport along the coastto finally beach. Coastal currents may induce a surface displace-ment of water to the north. Moreover, the strong wave actionobserved during the 9 day period of the PMSI poses seriousdifficulties to this hypothesis. Under strong wave regime a floatingbody is eventually pushed to the beach by wave motion, unless itresurfaces several kilometers away from the shore. If this was thecase, the body would have been transported by the strongalongshore currents (northward). In a similar way, for case 2 itcan be argued that the alternating flow of water imposed by thecyclic movement of the tide makes impossible to know how far thebody has drifted from the point of entry. However, if the body wastransported along the same stretch of water recurrently beforediscovered, it could only have happened during the previous night,since the area is heavily populated and in a waterway crossed dailyby numerous boats and ships. As such the body would have beeneasily spotted in daylight.

Other causes that justify long distance drift may be found in thefact that the victim could have died by other reason than drowning,entered the water already death, or could have been wearingflotation devices such as life jackets. In these cases the body neversunk and drifted with local currents upon entering the water.Finally there is also the possibility of the victims be immersed in amass of water traveling at a significant velocity, such as thefreshwater of swelled rivers discharging their excess volume intocoastal areas after heavy rainfall and under the influence of strong

wind. Case 2 happened near the Tagus Estuary mouth, which is avery dynamic region, mainly controlled by the tidal motion.However, in such a locations, physical features like the residualcirculation may be responsible to the surface displacement ofobjects (or bodies) in the water. Under these circumstances thebodies of drowned victims have been reported to travel as much as380 km in only 60 h [6].

5.2. Accumulated degree days estimation and body resurfacing

Some studies show that body size and weight, against what iscommonly assumed, do not have significant relevancy in the decayrate of a human body; this process is mostly controlled by ambienttemperature [32], so that the putrefaction rates decreases withlower water temperature. It is known that a body will resurfaceafter submersion in the bloating stage (Fig. 4C), unless it isstrapped to an object, closed in a compartment or lodged withindebris or structures at the bottom [3,33], or if it has sunk to depthsgreater than 100 m [10,33]. The body can resurface even whenstrapped to heavy objects such as anchors and stones [12,34]. Thereason is the accumulation of putrefactive gases inflating the body,decreasing the specific gravity of the body and giving it sufficientfloating force.

ADD measures the energy input as the accumulation oftemperature over time. Studies have shown that ADD is anadequate indicator of the rate of decomposition as a function oftemperature [20,24], especially to compare cases with differenttemperature ranges [21] such as the cases addressed here.However, this method requires knowing the mean daily tempera-ture, which can prove to be difficult, if not impossible, in mostaquatic environments. In the cases described here this value is anapproximation (Table 1) since environmental temperature datawas only available as a single daily value (Fig. 2, see Section 3).

Since water temperature changes are not drastic in the ocean,this can be considered an adequate approximation to the meandaily temperature. Nevertheless, water temperature changesconsiderably in depth and spatially in systems with significanttidal shifts or under the influence of estuarine ecosystems like theTagus and the Douro estuaries. Thus, the surface temperatureprovided by remote sensing can be considered a rough estimation.The results for ADD in both cases (117 8C and 95 8C) show somevariation, but given the limitation imposed by the lack of detaileddata, one can assume that the difference can be lower. However, avariation around the temperature values used to calculate ADD canbe ascribed to account for potential deviation from the usedsatellite data (low resolution, near coast masks, etc.) or in situ

variation caused by depth, local temperature increase due tosunlight, or any other physical processes that may have aninfluence on water temperature.

Differences of up to 0.5 8C are expected between differentsatellite sensors and between satellite and in situ measurements.Frequently the satellite measurements are made during nighttime(as the product used in this study), and it is known that solarheating of the ocean surface can cause warming of up to 3 8C duringday time [35–37]. The temperature differential is more significantin spring and summer months when solar radiation is higher, so ahigher deviation in the temperature could be expected in case 2. Toaccount for this variation, an estimate can be made by adding atemperature variation range (Dt) in Eq. (2), thus determining thecumulative lower (ADDmin) and upper (ADDmax) limits for ADDduring the PMSI:

ADD range

ADDmin ¼Xn

i¼1

ðTi � DtÞ � hi

24

� �

ADDmax ¼Xn

i¼1

ðTi þ DtÞ � hi

24

� �

8>>>><>>>>:

(3)

Fig. 5. Calculated ADD (black squares and dots) and estimated ADD range (gray

areas) during the PMSI in both cases. Values calculated according to Eqs. (2) and (3)

(see discussion for details).

M. Mateus et al. / Forensic Science International 229 (2013) e6–e12 e11

By assuming a �1.5 8C variation range on the temperature valuesused to calculate ADD in Table 1, the ADD range in case 1 is 104–130 8C, while in case 2 it becomes 87–104 8C at the end of the PMSI(Fig. 5). While both cases converge to an approximated narrowinterval when considering this ADD range, it is impossible to establishany definitive link between ADD and the time of resurfacing. Such alink can only be inferred from a larger data set.

It may be misleading, however, to assume ADD as the onlyreason to explain the timing of resurfacing in both bodies.Considering the significant differences in the accidents location,the bodies could have sunk to different depths, which mean thathydrostatic pressure could have play a different role in both cases.It is known that in aquatic environments, putrefaction andautolysis advance more rapidly at low hydrostatic pressure [19].Also, human specific gravity is affected by its intrinsic character-istics such as height, weight and composition of body tissue, but itcan be considerably changed by clothes, or any objects attached tothe body. As such, differences in size between both subjects mustalso be taken into consideration, and the fact that in case 1 thevictim was clothed at the time of the accident, while in case 2 thevictim was only wearing bathing trunks. Moreover, gas accumu-lates in different parts of the body in male and female bodies thusaffecting floatability. Finally, the body in case 1 could have becomesnagged or lodged in rocks protruding from the bottom, thusdelaying the resurface process.

5.3. The effect of tide

In both cases the body was found and recovered during low tide(Fig. 3), and as discussed previously it is fairly plausible to assumethat the bodies had not resurfaced much earlier than the time theywere found. The fact that both bodies seem to have resurfacedduring low tide, can be explained by the gradual decrease inhydrostatic pressure due to the decrease of the water columnheight above the body (Fig. 4D). The change in depth leads to anexpansion of the putrefactive gases that have accumulated duringthe decomposition process. A slight increase in floatability(increase in body volume causing a decrease in the specificgravity) will cause the body to ascend, and the subsequentdecrease in pressure will have a positive feedback in the buoyancyforce [19,38]. For bodies reaching the equilibrium betweenbuoyancy and gravity forces (Fig. 4B), the gradual decrease inthe hydrostatic pressure in just a few hours could, in fact, be thetrigger for resurfacing. Moreover, tidal motion may inducehorizontal displacement of the bodies, In fact, both Douro (case1) and the Tagus Estuary lie in regions of energetic tides, whichmay induce surface current velocities higher than 1 m s�1. Also, the

estuarine discharge and the coupling with the alongshore coastalcurrents may play a key role in the body displacement at theselocations of the Portuguese coast. Near the Tagus mouth, underfavorable environmental conditions, such as significant riverdischarge and strong wind conditions, the estuarine water canbe transported several nautical miles along the coast over a fewdays [39].

6. Conclusions

Body displacement in aquatic environments is highly variableand its understanding requires knowing how ambient conditionsaffect the decomposition rate of the body and, at the same time,influence its vertical and therefore also horizontal drift. In this way,the interaction between forensics and operational oceanography isuseful in the study/investigation of these events.

The two cases analyzed here provide relevant information tounderstand the displacement of bodies in marine coastal areas,because the time and place of death are known with precision, aswell as the time and place where the bodies were recovered. Basedon this information we have calculated ADD with SST from satelliteimages and analyzed basic oceanographic features of both accidentsites.

Some general remarks can be drawn from this study:

� Submerged bodies resurface when the decomposition processreaches the bloating phase and production of putrefactive gasesis enough to expand body cavities and decrease the specificgravity of the body which in turn increases its buoyancy; in thecases reported here this phase was estimated to have occurred at95 and 117 ADD. These values can be expanded to account theuncertainty in temperature values.� Significant drift of the bodies only takes place when the body

resurfaces. Bottom friction, low intensity currents or obstaclesseem to impose serious restrictions on body displacement whilestill on the bottom.� The coincidence of resurfacing happening during low tide

suggests that the decrease in the water column height abovethe body (decrease in the hydrostatic pressure), decreases thespecific gravity of the body and subsequently increase itsbuoyancy, thus triggering the resurface process.

The results are still preliminary since further cases must beanalyzed before establishing any patterns. Further studies mayinclude the use of hydrodynamic/circulation models to estimatethe trajectory of the bodies, since these numerical tools canaccurately predict the displacement of passive tracers using anEulerian or Lagrangean approach, and to understand the influenceof oceanographic conditions on body drift. For the Portuguesecoast, information on currents and water temperature is alreadyavailable in the form of hindcast, nowcast and 3-day forecast [40](data available through the portal http://forecast.maretec.org/).

Finally, the results presented in this work provide somerelevant clues on the fate of human cadavers in aquaticenvironments that can be used by officials and agencies involvedin the recovery of bodies, as well as by forensic investigators whendealing with these findings.

Acknowledgements

The authors are thankful to the Maritime Police in the port ofLeixoes, and to Teixeira Pereira, Assistant Captain of the Port ofLisbon, for providing the details on the place and time of theaccidents and body recovering. Also, we would like to express ourgratitude to Sibila Pando and Constanca Belchior for valuableassistance in the access to information.

M. Mateus et al. / Forensic Science International 229 (2013) e6–e12e12

References

[1] J.C. Giertsen, I. Morild, Seafaring bodies, Am. J. Forensic Med. Pathol. 10 (1989)25–27.

[2] J.A. Stoop, Maritime accident investigation methodologies, Inj. Control Saf. Pro-mot. 10 (2003) 237–242.

[3] T. Kahana, J. Almog, J. Levy, E. Shmeltzer, Y. Spier, J. Hiss, Marine taphonomy:adipocere formation in a series of bodies recovered from a single shipwreck, J.Forensic Sci. 44 (1999) 897–901.

[4] S.A. Papadodima, S.A. Athanaselis, E. Skliros, C.A. Spiliopoulou, Forensic investi-gation of submersion deaths, Int. J. Clin. Pract. 64 (2010) 75–83.

[5] P. Lunetta, J.H. Modell, in: M. Tsokos (Ed.), Macroscopical, Microscopical, andLaboratory Findings in Drowning Victims, Forensic Pathology Reviews, vol. 3,Springer/Humana Press Inc., New York/Dordrecht/Heidelberg/London/Hamburg,Germany, 2005, pp. 3–77.

[6] J.B. Pampın, B.A. Rodrıguez, Surprising drifting of bodies along the coast ofPortugal and Spain, Leg. Med. 3 (2001) 177–182.

[7] B. Kringsholm, J. Jakobsen, B. Sejrsen, M. Gregersen, Unidentified bodies/skullsfound in Danish waters in the period 1992–1996, Forensic Sci. Int. 123 (2001)150–158.

[8] S.P. Avis, Suicidal drowning, J. Forensic Sci. 38 (1993) 1422–1426.[9] G. Cetin, Y. Gunay, S.K. Fincanci, R. Ozdemir Kolusayin, Suicides by jumping from

Bosphorus Bridge in Istanbul, Forensic Sci. Int. 116 (2001) 157–162.[10] J.C. Giertsen, I. Morild, The fjord murders, Am. J. Forensic Med. Pathol. 10 (1989)

124–129.[11] J. Lucas, L.B. Goldfeder, J.R. Gill, Bodies found in the waterways of New York City, J.

Forensic Sci. 47 (2002) 137–141.[12] G. Haupt, Drowning investigations, FBI Law Enforcement Bull. 75 (2006)

14–22.[13] J.-P. Beauthier, E. De Valck, P. Lefevre, J. De Winne, Mass disaster victim identifi-

cation: the Tsunami experience, Open Forensic Sci. J. 2 (2009) 54–62.[14] M.H.A. Piette, E.A. De Letter, Drowning: still a difficult autopsy diagnosis, Forensic

Sci. Int. 163 (2006) 1–9.[15] V.N. Ambade, A.N. Keoliya, R.B. Deokar, P.G. Dixit, Decomposed bodies – still an

unrewarding autopsy? J. Forensic Leg. Med. 18 (2011) 101–106.[16] W.D. Haglund, Disappearance of soft tissue and the disarticulation of human

remains from aqueous environments, J. Forensic Sci. 38 (1993) 806–815.[17] D.R. Dilen, The motion of floating and submerged objects in the Chattahoochee

River, Atlanta, GA, J. Forensic Sci. 29 (1984) 1027–1037.[18] E.R. Donoghue, S.C. Minnigerode, Human body buoyancy: a study of 98 men, J.

Forensic Sci. 22 (1977) 573–579.[19] A. Reisdorf, R. Bux, D. Wyler, M. Benecke, C. Klug, M. Maisch, P. Fornaro, A. Wetzel,

Float, explode or sink: postmortem fate of lung-breathing marine vertebrates,Palaeobiodivers. Palaeoenviron. 92 (2012) 67–81.

[20] M.S. Megyesi, S.P. Nawrocki, N.H. Haskell, Using accumulated degree-days toestimate the postmortem interval from decomposed human remains, J. ForensicSci. 50 (2005) 618–626.

[21] T. Simmons, R.E. Adlam, C. Moffatt, Debugging decomposition data—comparativetaphonomic studies and the influence of insects and carcass size on decomposi-tion rate, J. Forensic Sci. 55 (2010) 8–13.

[22] P. Lunetta, K. Tiirikainen, G.S. Smith, A. Penttila, A. Sajantila, How well does anational newspaper reporting system profile drowning? Int. J. Inj. Contr. Saf.Promot. 13 (2006) 35–41.

[23] C. Antunes, Previsao de Mares dos Portos Principais de Portugal, 2007 FCULWebpage, http://webpages.fc.ul.pt/�cmantunes/hidrografia/hidro_mares.html.

[24] V. Heaton, A. Lagden, C. Moffatt, T. Simmons, Predicting the postmortem sub-mersion interval for human remains recovered from U.K. waterways, J. ForensicSci. 55 (2010) 302–307.

[25] C. Yorulmaz, N. Arican, I. Afacan, H. Dokgoz, M. Asirdizer, Pleural effusion inbodies recovered from water, Forensic Sci. Int. 136 (2003) 16–21.

[26] C.C. Ebbesmeyer, W.D. Haglund, Drift trajectories of a floating human bodysimulated in a hydraulic model of Puget Sound, J. Forensic Sci. 39 (1994) 231–240.

[27] J.E. Azparren, C. Cubero, E. Perucha, P. Martınez, G. Vallejo, Comparison betweenlung weight and blood strontium in bodies found in seawater, Forensic Sci. Int.168 (2007) 128–132.

[28] R.H. Fuller, Drowning and the post-immersion syndrome: a clinicopathologicalstudy, Mil. Med. 128 (1964) 22–36.

[29] F.S. Golden, M.J. Tipton, R.C. Scott, Immersion, near-drowning and drowning, Br. J.Anaesth. 79 (1997) 214–225.

[30] L. Snyder, Homicide Investigation: Practical Information for Coroners, PoliceOfficers, and Other Investigators, 3rd ed., Charles C Thomas Pub Ltd., Springfield,Illinois, USA, 1977.

[31] S. Carniel, G. Umgiesser, M. Sclavo, L.H. Kantha, S. Monti, Tracking the drift of ahuman body in the coastal ocean using numerical prediction models of theoceanic, atmospheric and wave conditions, Sci. Justice 42 (2002) 143–151.

[32] R.W. Mann, W.M. Bass, L. Meadows, Time since death and decomposition of thehuman body: variables and observations in case and experimental field studies, J.Forensic Sci. 35 (1990) 103–111.

[33] T.K. Dumser, M. Turkay, Postmortem changes of human bodies on the bathyal seafloor—two cases of aircraft accidents above the open sea, J. Forensic Sci. 53 (2008)1049–1052.

[34] P. Lunetta, J.H. Modell, in: M. Tsokos (Ed.), Macroscopical, Microscopical, andLaboratory Findings in Drowning Victims: A Comprehensive Review, ForensicPathology Reviews, Humana Press Inc., Totowa, NJ, 2005, pp. 3–77.

[35] J.F. Price, R.A. Weller, R. Pinkel, Diurnal cycling: observations and models of theupper ocean response to diurnal heating, cooling, and wind mixing, J. Geophys.Res. 91 (1986) 8411–8427.

[36] R. Yokoyama, M. Konda, Sea surface effects on the sea surface temperatureestimation by remote sensing—Part 2, Int. J. Remote Sens. 17 (1996) 1293–1302.

[37] R. Yokoyama, S. Tanba, T. Souma, Sea surface effects on the sea surface tempera-ture estimation by remote sensing, Int. J. Remote Sens. 16 (1995) 227–238.

[38] H. Peltier, W. Dabin, P. Daniel, O. Van Canneyt, G. Doremus, M. Huon, V. Ridoux,The significance of stranding data as indicators of cetacean populations at sea:modelling the drift of cetacean carcasses, Ecol. Indic. 18 (2012) 278–290.

[39] N. Vaz, L. Fernandes, P.C. Leitao, J.M. Dias, R. Neves, The Tagus estuarine plume as aresponse to wind and river runoff: winter 2007 case study, J. Coastal Res. SI56(2009) 1090–1094.

[40] M. Mateus, G. Riflet, P. Chambel, L. Fernandes, R. Fernandes, M. Juliano, F.Campuzano, H. de Pablo, R. Neves, An operational model for the West Iberiancoast: products and services, Ocean Sci. Discuss. 9 (2012) 1651–1689.