A Longitudinal Study on the Outdoor Human Decomposition Sequence in Central Texas
MetadataShow full metadata
Estimating the postmortem interval (PMI), or how much time has passed since an individual died, is an important aspect of investigating a death. Traditionally, forensic anthropologists have relied on non-standardized decomposition stages, anecdotal evidence, and personal experience to make an estimation of the PMI (Love and Marks 2003). Decomposition sequences have been proposed for specific geographic regions (Mann et al. 1990; Galloway 1997; Komar 1998; Rhine and Dawson 1998; Love and Marks 2003), but these stages may not be applicable to different climates and most were developed from cross-sectional data (Galloway et al. 1989; Komar 1998; Rhine and Dawson 1998). Recently, Megyesi et al. (2005) developed a quantitative method of estimating the PMI using accumulated degree-days (ADD), temperature data, and total body score (TBS), a system of numerically ranked qualitative observations of decomposition. This method was developed from cross-sectional data and has never been tested in a longitudinal experimental study using human cadavers. In addition, scavenging and its effect on using ADD to estimate the PMI has not been addressed (Simmons et al. 2010a). The present study tested Megyesi et al.'s (2005) model of scoring decomposition and its relationship to ADD using human cadavers. The goals of this study were to test the system outlined by Megyesi et al. (2005) using longitudinal data and examining the decomposition process directly. This study examined the assumption that all of the stages and decomposition characteristics used by Megyesi et al. (2005) and based on Galloway et al.'s (1989) decomposition stages follow a sequential order. The degree in which scavenging animals in this environment affect the decomposition rate and the estimation of ADD from TBS was incorporated. From November 2009 to July 2010, 10 donated human cadavers were placed outdoors at the Forensic Anthropology Research Facility (FARF) at Texas State University-San Marcos. Decomposition was ranked using the TBS system for each day of observation over time. Observations support the general decomposition stages found in high temperature and high humidity environments (Galloway et al. 1989; Galloway 1997) with accelerated autolysis, high rates of maggot activity when scavengers are controlled for, and rapid skeletonization. TBS, however, is not linear, with changes in certain decomposition characteristics able to influence the observer's recorded TBS. Statistically significant differences were found between the estimated ADD and the actual mean ADD for each major decomposition stage. The differences were still significant after cadavers that were scavenged were removed from analysis, meaning that these differences were not caused by scavengers alone. In this study, longitudinal data collection allowed for a comparison between scavenged and non-scavenged human bodies. Scavenged bodies had significantly lower ADD (i.e. faster rates) to reach major decomposition stages than protected cadavers. This study shows in a quantitative manner that scavenging animals can have a significant impact on the estimation of the PMI from ADD. Exact binomial tests tested the rate of the equation produced by Megyesi et al. (2005) to successfully predict ADD against an expected success rate. The method had 100% accuracy rates for decomposition scores less than 22, but this was found to be indicative of a lack of precision stemming from a large standard error. Bodies skeletonized much faster than what was estimated with the equation, and the low success rates for scores 22 and above make the equation not recommended for severely decomposed remains. Only score 23 effectively predicted ADD from TBS (~90%), but all successes were recorded from one donation. The results of this study demonstrate that different environments may contain significant variables that the Megyesi et al. (2005) decomposition scoring system does not specifically address. In addition, low success rates for the Megyesi et al. (2005) equation to predict ADD from TBS above 22 and the wide standard error ranges provided demonstrate the need to reevaluate the equation for PMI estimation from TBS.