Microbial community succession on submerged vertebrate carcasses in a tidal river habitat: Implications for aquatic forensic investigations

Abstract

Death investigations in aquatic ecosystems are challenging due to abiotic and biotic factors that may influence the estimation of a postmortem submersion interval (PMSI).In this study, we examined bacterial changes throughout the decomposition process on porcine carcasses submerged in a tidal-influenced river and identified predictors of epinecrotic community succession. Fetal porcine (Sus scrofa) carcasses (N = 6) were submerged with epinecrotic samples collected every 3 days (6 collections) over a period of 19 days (~7415 accumulated degree hours (ADH)). Amplicon sequencing was performed using the Illumina MiSeq platform (16S V4 region, 2 × 250 bp format) to identify changes in bacterial relative abundance and diversity. To match bacterial succession with rough taphonomy, carcasses were visually assessed at each sampling time point to determine the decomposition stage. Notably, the three most abundant families were Moraxellaceae, Burkholderiaceae (Proteobacteria), and Clostridiaceae (Firmicutes), though communities composition varied significantly across decomposition stages. Greater bacterial phylogenetic diversity was observed in in latter decomposition stages (advanced floating decay, sunken remains). Random Forest Models were built to predict ADH and explained 77%-80.8% of variation in ADH with an error rate of +/−1943.2 ADH (Root Mean Square Error) or approx. ±2.7 days at the mean water temperature of this study. This study provided a useful model that could be used to estimate a PMSI in this river system utilizing bacterial community succession, and thus, potentially improve the accuracy of such estimations to be used in the court of law.

Microbial community succession of submerged bones in an aquatic habitat

“…Despite differences among study location and design, the predominate phyla identified here were similar among other decomposition studies. Firmicutes, Bacteroidetes, and Proteobacteria were predominate taxa in this study, but were also prevalent in terrestrial decomposition [7,50,53], other aquatic decomposition [9,10,20,21,61], and bone postmortem microbiomes [16,17]. Basidiomycota and Mucoromycota were predominate taxa in terrestrial [8,18,19,55,57] and aquatic decomposition and identified as important taxa in our study as well [11].…”

“…For aquatic decomposition, many previous studies used nonhuman surrogates. These include tiles for biofilm formation [60], swine remains [9,11,20,21,61],…”

“…For aquatic decomposition, many previous studies used nonhuman surrogates. These include tiles for biofilm formation [60], swine remains [9,11,20,21,61], salmon [62], and swine bones without a full carcass [10,12]. The aquatic systems also differed among studies, ranging from tidal and freshwater systems [10–12], to small streams [21,60], rivers [61], and marine environments [20].…”

“…3 weeks) aquatic decomposition using skin and clothing communities and were able to explain about 88% of the microbial community variation over PMSI [9]. Other short‐term PMSI models explained 77%–80.8% of variation in ADH with an error rate of 1943 ADH or ~2.7 days [61]. Importantly, in this study with a much longer submergence study timeframe (i.e., 353 days) and different source substrate (i.e., bone), the random forest model was 94% accurate with an error of 472.3 ADD using family‐level taxonomic classification for bacteria [10] and 87% accurate using family‐level taxonomic classification for eukaryotes with an error of 536.8 ADD [12].…”

“…salmon [62], and swine bones without a full carcass [10,12]. The aquatic systems also differed among studies, ranging from tidal and freshwater systems [10][11][12], to small streams [21,60], rivers [61],…”