It is well established that the identification process relies on a comparison and reconciliation of postmortem and antemortem information [1, 2]. Such reconciliation may occur in a timely manner, typically in cases where the circumstances of the recovery provide contextual information for a proposed identification hypothesis and the preservation of the remains allows suitable postmortem data to be collected. However, in cases where skeletonized human remains are found and there is little, or no, associated contextual information, the identification process is significantly hindered; often resulting in the individual remaining unidentified for extended periods of time [3, 4]. This paper presents the case of skeletonized human remains found in a context that provided no information to generate an identification hypothesis and details how five specialist disciplines worked for over 5 years to resolve the identification.

On 25 December 2017, a recreational snorkeler observed a human skull lying on the seafloor at Shallow Inlet, ∼25 m from the Sandy Point coastline, near Wilsons Promontory, southeast of Melbourne, Australia (Figure 1).

The State's Police Search and Rescue Team attended the scene and undertook the recovery of a near-complete skeleton lying supine on the seafloor buried under approximately 20 cm of sand. There were no clothing or personal effects associated with the remains. Upon recovery of the remains, the death was reported to the coroner, and the individual was transferred to the Victorian Institute of Forensic Medicine (VIFM) for examination.

The individual was completely skeletonized and very well preserved with an estimated 91.14% of the skeleton present [5]. Only the mandible, sternum, patellae, and select hand and foot bones had been lost postmortem. The skeletal elements showed some evidence of bioerosion. The upper and lower extremities and some right ribs were stained black, characteristic of contact with a sandy matrix [6], while the frontal bone and facial skeleton exhibited green staining, characteristic of contact with organic material such as algae [7].

An anthropological assessment using standard morphological and metric techniques [8] estimated the individual to be a male of European ancestry, aged between 21 and 37 years, with a stature of 169.7±3.27 cm. The individual had no apparent evidence of skeletal trauma or pathology and had, at some stage in his life, attended a dentist.

The maxillary dentition showed several interesting features. The left third molar was impacted, and the right first molar had been extracted. There was also diastema (i.e. a gap between the central maxillary incisors). The individual had a gold restoration on the distal aspect of the left second incisor and amalgam restorations on the right first premolar and first molar and the left first and second molars. The style of these restorations was not typical of contemporary Australian dental practice [9]. This indicated the individual may have had dental work undertaken overseas, or that the dental work had been done in Australia but was of some antiquity.

A sample of bone was collected from the shaft of the right proximal femur for DNA testing. Following the recovery of a DNA extract, a complete nuclear DNA (nDNA) profile (21 loci) as well as a complete mitochondrial DNA (mtDNA) profile (HVI and HVII) were obtained. At this time, it was noted that the mtDNA profile was indicative of the individual having a European maternal ancestry (including Scandinavian countries). These profiles were uploaded and searched on the Victorian Missing Persons DNA Database (VMPDD) [10]. When no match was obtained, coronial permission was subsequently provided for the DNA profiles to be uploaded to the National Criminal Investigation DNA Database-Integrated Forensic Analysis (NCIDD-IFA) [11]. Again, no match was obtained. Both databases have the capability to conduct direct and familial (also known as kinship) matching, with the view of reconciling the unknown deceased with long-term missing persons through DNA profiles that have been uploaded for the missing or their immediate family member (s) [12].

As no DNA matches were obtained following comparisons with the state and national missing persons DNA databases, additional DNA analysis in the form of forensic DNA intelligence (FDI) was undertaken. Single nucleotide polymorphisms (SNPs) were analysed to infer the biogeographical ancestry (BGA) and externally visible characteristics (EVCs) (such as eye and hair colour) of the individual. The BGA results indicated that the individual's ancestry was 95% European and 5% South-West Asian. The EVCs results showed that the individual most likely had blue eyes and hair that was a dark shade of brown. Further, the BGA (from a maternal point of view) was corroborated by the mtDNA haplotype. This intelligence, together with information obtained from the anthropology and odontology examinations, was compared with known information for missing persons cases. There were no reported missing persons in Victoria that appeared to be a match for this individual.

As the individual still had not been identified 1 year after being recovered and, in light of the dental findings, the possibility that the remains were of some antiquity was considered. Given that it is not possible to reliably estimate the time since death from a visual examination of skeletonized remains, radiocarbon dating was undertaken [13]. A maxillary left third molar and a sample of cortical bone from the left femur were collected for radiocarbon testing. The tooth sample was tested at Australia's Nuclear Science and Technology Organisation (ANSTO), and the bone sample was tested at the Australian National University. The radiocarbon ages were calibrated against the 2013 Southern Hemisphere atmospheric bomb radiocarbon dataset for zones 1–2 [14]. Calibration was performed using the OxCal v4.2.3 programme [15].

The results indicated that collagen in the tooth was formed prior to 1955. Further, both the tooth and cortical bone sample contained a fraction modern (F¹⁴C) concentration (i.e. a measurement of the deviation of the ¹⁴C/¹²C ratio of a sample from “modern”—i.e. 1950) that correlated with pre-bomb curve radiocarbon levels (Figure 2). Multiple formation ranges (Figure 2) were given for each sample that corresponded to instances of environmental radiocarbon matching the level found within the samples. These levels essentially indicated the bone and tooth were formed somewhere between 1666 and 1955, but it was only possible to say with accuracy that the remains were pre-bomb pulse (see [17] for an explanation of bomb pulse dating). The Victorian Coroners Act 2008 provides that the coroner may investigate a reportable death that has occurred between 50 and 100 years before the death was reported, and that the coroner must discontinue the investigation into the death if the coroner determines that the death probably occurred more than 100 years before it was reported to a coroner [18]. While the radiocarbon dating indicated that the remains were of some antiquity, it was not possible to provide a definitive answer as to whether the remains were of medico-legal significance.

Given the forensic anthropology, odontology, and radiocarbon findings, it was anticipated that the case was historic and that a possible cause of death was drowning at either Shallow Inlet or the adjoining Waratah Bay (Figure 1). While today a long sandy spit narrows the entrance into Shallow Inlet, in 1901 this spit was broken through, and the entrance to Shallow Inlet, where there is a strong tidal flow, was significantly widened [19]. Consequently, it was possible that a person who drowned in Waratah Bay could have been washed into Shallow Inlet with the incoming tide and subsequently buried in the sand. This might explain the rarity of finding an almost complete skeleton in the ocean after a significant period of time. This hypothesis prompted research into drownings in or near Waratah Bay in the 19th and 20th centuries through a search of newspaper articles in Trove [20]. The search yielded three shipwrecks, one plane crash, and three individual drownings.

In the 19th century Waratah Bay provided shelter to sailing vessels travelling between Melbourne, Hobart, and Sydney. This was a major shipping route where stormy weather would force boats into the Bay to wait for calmer waters. Unfortunately, many ships ran aground, or took on water, losing their crew, passengers, and cargo in the bay, including The Domain (1846) [21], The Spencer (1854) [22], and The Bertha (1870) [23]. A total of 18 adults drowned in these incidents, with only one body being recovered. However, the historical research identified that although dental amalgam was introduced into the USA in 1833 under the name of “Royal Mineral Succedaneum”, its use was highly contested and not adopted as the predominant restorative material until the late 1880s [24]. This suggested that the unidentified individual had likely attended a dentist sometime after 1880 and therefore could not have been one of the crew or passengers who drowned in Waratah Bay following these shipwrecks.

A search of historical newspaper reports identified at least four other young adult men who drowned in the area in the 19th and early 20th centuries. The first was an escaped criminal Martin Weiberg, who drowned in 1883 while sailing in rough weather from Walkerville to the Glennie Islands off Wilsons Promontory. The wreckage of his dingy was found on Sandy Point Beach, and no body was recovered [25]. Weiberg provided a plausible fit with the forensic profile of the unidentified skeletal remains, his prison records giving his height as 5 feet 6 and a half inches (168 cm), Norwegian ancestry, and grey eyes [26]. He was employed as a ship carpenter in the 1870s [27], and it may have been possible for him to have dental restorations in overseas ports. However, his prison record described Weiberg as having light brown hair, whereas phenotyping (see above) indicated that the individual had dark brown hair.

The second individual was a fisherman from Tarraville, John Jacobson, who drowned in 1911; his boat was reported to have been found anchored in Shallow Inlet [28]. The ancestry of Mr. Jacobson, born in Sweden, was a match to the unidentified individual; however, Tarraville is east of Wilsons Promontory, and it is much more likely that he drowned near Port Albert or in Corner Inlet (a significant fishery established in the late 1800s) rather than in Shallow Inlet, which is to the west of Wilsons Promontory.

The third man, Christopher Luke Moore, was a 29-year-old farmer and World War I veteran from Buffolo, who was reported to have drowned in Waratah Bay on 30 December 1928 during a family picnic at a popular swimming spot called “The Gap” [29]. Although no documentation was found to suggest that his body had been recovered, a gravestone at the nearby Meeniyan Cemetery included Christopher Moore's name, together with his wife and parents, which possibly eliminated him as being the unidentified individual.

Finally, the air-liner Miss Hobart disappeared off Wilsons Promontory when flying from Launceston to Melbourne in 1934, with a seat from its wreckage being found at Sandy Point Beach [30]. Among the nine passengers and two crew members who drowned, which included nine men, was a dentist from Launceston, Mr. Colin Jones [31]. The historical research provided some possible leads as to the identity of the unidentified remains; however, without mtDNA analysis from a maternal descendant for each contender, it was not possible to confirm an identification.

Although several possible identification hypotheses were generated based on the historical research (strongly supporting the forensic medical findings), advancements in forensic DNA analysis tools—namely forensic/investigative genetic genealogy (F/IGG)—were needed to progress the identification work. The F/IGG work was undertaken in 2022, led by Dr Runa Daniel (see Acknowledgments), as part of an ethically approved research project conducted by the VIFM and its collaborators to evaluate F/IGG for Australian casework (EC 11-2019 and EC 1151).

As the latest addition to FDI techniques, F/IGG combines DNA testing with traditional genealogical methods to attempt to generate an investigative lead as to the identity of the donor of a DNA sample [32, 33]. F/IGG relies on the use of SNP data, which is then uploaded to commercially available genealogy databases for familial searching; with the data set in the hundreds of thousands or millions of SNPs, depending on the methodology used to generate the SNP profile [32, 33]. In contrast to conventional DNA profiling, where only direct or kinship matching to close (first degree) relatives can occur, F/IGG enables kinship matching to occur to more distant relatives (e.g. second, third, fourth, and fifth degree) [34]. In 2018, F/IGG was thrust in the spotlight with the identification of the “Golden State Killer” Joseph James DeAngelo, where F/IGG was instrumental in the development of the investigative lead [32, 35, 36]. Since this time, F/IGG has been used in a range of contexts to assist with identifying so called “cold cases” [37–41].

In April 2022, a sample of DNA was sent to Othram (Texas, USA) for whole genome sequencing, with the VIFM's DNA laboratory receiving a SNP profile ready for upload to the two genealogy databases accessible to law enforcement (GEDmatch PRO and FTDNA). Comparisons returned matches to potential relatives of the unknown deceased—close enough relatives to make family tree building possible. Using the various tools available to query the match lists, both the maternal and paternal sides of the family tree were built back (Figure 3). Many of the potential relatives identified in the family tree-building lived in Southeast Gippsland, indicating that the skeletal remains belonged to a local man. Once the most recent common ancestors (MRCAs) and union couple (i.e. the deceased's likely parents) were identified, an identification hypothesis that the deceased may be Mr. Christopher Luke Moore was proposed.

It is worth noting that the BGA prediction (conducted prior to F/IGG) was instrumental in enabling case selection for F/IGG, as the (publicly available) genealogy databases used for this work have skewed ethnic representation. As such, only human remains thought to be of European origin, were selected for the F/IGG research project. Furthermore, the fact that the results of the bomb-pulse (radiocarbon) dating (conducted while F/IGG was in progress) indicated the remains were not considered “modern” also assisted the genealogy research.

The identification hypothesis that the remains may be that of Mr Moore was strongly supported by further historical data. On 15 January 1929, 2 weeks after Mr. Moore drowned, a mandible was found at Shallow Inlet. While it is not clear whether a formal dental examination was undertaken on the mandible that was recovered, the inquest documentation describes how the father of Mr. Moore, Mr. Cornelius Moore, “identified” the mandible as belonging to his son based on the dental work, namely one missing molar and four vulcanite fillings, two on each side [42]. Subsequently, on 24 January 1929, a Magisterial Inquiry determined that Mr. Christopher Luke Moore died from accidental drowning at Waratah Bay on 30 December 1928. It is important to note that the mandible was one of the few skeletal elements missing postmortem from the deceased, and that Waratah Bay is geographically close to where the deceased was recovered (Figure 1). The mandible was later buried with Christopher Moore's wife and parents at Meeniyan Cemetery.

Christopher Luke Moore was of Irish Catholic descent, and his military records confirmed that he was born in February 1899, was 5 feet 9 ¼ inches tall (1.76 m), with blue eyes and black hair (Figure 4). This was consistent with the anthropological findings, radiocarbon dating, biogeographical data, and DNA phenotyping of the skeletal remains.

While the F/IGG and circumstantial data strongly indicated the deceased was Mr. Christopher Luke Moore, a direct DNA comparison was required for formal identification by the coroner. Further ancestry research indicated that Mr. Moore had three living descendants (a grandniece and two grandnephews) on his maternal side. A mitochondrial profile obtained from a DNA sample provided by the grandniece of Mr. Moore was compared to the mitochondrial DNA of the human remains. The results of this DNA comparison supported the proposition that the grandniece was a maternal relative of the deceased. Consequently, the remains were formally identified by the State Coroner on 27 July 2023 as Mr. Christopher Luke Moore. Remarkably, this identification was made 95 years after Mr. Moore tragically drowned while swimming in 1928. His remains were finally returned to his family for burial in 2023.

It is widely acknowledged that the process of identifying human remains is a complex undertaking that necessitates the consideration of all available lines of evidence [43]. This case study serves as an exemplification of the fact that, when dealing with skeletonized remains, the identification process demands, more than ever, a multidisciplinary and collaborative approach, often entailing significant time and resource investments.

The authors are grateful to Emeritus Professor Richard Wright (University of Sydney) for comments on the cranial metrics and Dr. Eden Johnston-Belford (Monash University) for discussion about the radiocarbon results. For the F/IGG work, the authors greatly acknowledge the contributions of Dr. Nathan Scudder (Australian Federal Police), Dr. Jennifer Raymond and Alison Sears (New South Wales Police Force), Dr. Runa Daniel (University of New Haven), and Dr. Colleen Fitzpatrick (Identifinders International). The authors also acknowledge Mr. Moore's family who provided the photograph of Mr. Moore and their consent to publish this article.

Soren Blau, Dadna Hartman, Fiona Leahy, Jodie Leditschke, Samantha Rowbotham, Lyndall Smythe, and April Stock contributed to the identification process; Soren Blau, Dadna Hartman, Fiona Leahy, Jodie Leditschke, and Samantha Rowbotham were responsible for drafting the manuscript. All authors reviewed the manuscript.

This article does not contain any studies with human participants or animals performed by the authors. The publication of this article has been granted with the consent of Mr. Moore's family.

Soren Blau initial holds the position of Editorial Board member for Forensic Sciences Research and is blinded from reviewing or making decisions for the manuscript.

None.