Anne E. Perez

Date of Graduation

2014

Document Type

Dissertation

Degree Type

PhD

College

Eberly College of Arts and Sciences

Department

Sociology and Anthropology

Committee Chair

Jeffrey D Wells

Committee Co-Chair

Kevin C Daly

Committee Member

Lynn R LaMotte

Committee Member

Keith B Morris

Committee Member

Rita VM Rio

Abstract

For academic carrion insect succession studies to be applicable to forensic casework, the study must provide statistically defensible postmortem interval (PMI) estimates. Multiple statistical approaches have been used to describe insect succession, but no study to date has been able to determine a confidence interval about a succession based PMI estimate. The ability to reject PMI values, and thereby create a confidence interval is determined by statistical power, of which sample size is a positive correlate. A proposed model established prospective sample sizes for desired levels of statistical power, indicating a target sample size of ~50 carcasses for estimating PMI based on two carrion insect species (LaMotte and Wells 2000). A surrogate for human decomposition has been identified, the domestic pig, yet studies to date have failed to include more than 5 carcasses in a single treatment group. This lack of replication has lead to strictly observational findings that are unsuitable for forensic use. As replication of carcasses increases, in an attempt to generate a suitable reference insect succession dataset, additional obstacles are encountered. Constraints of time and space, as well as development of an appropriate species list and sampling procedure are of specific concern. Throughout a century of research concerning insect succession on carrion, issues central to development of large datatsets, including the effect of year, fine-scale spatial discrepancies, repeated sampling and intercarcass distance, remain largely unknown.;The use of temperature to describe succession is an additional area of succession research critical to casework application. Because temperature directly affects decomposition rates, insect development rates and insect activity, succession data collected across time and space (potentially experiencing differing temperature histories) may most accurately be described using a physiological time measure as compared to an absolute time measure. Further, succession described in physiological time may be more accurate for application to a case that occurred at a different time and location than the reference dataset. The proposed research aims to create a succession dataset that is suitable for statistical analysis and estimation of time since death by: (1) investigating current sampling methods as well as the assumption that succession is consistent among year and small-scale variation in location, (2) determination of minimum intercarcass distance to ensure independence of carcasses, (3) identifying candidate insect species appropriate for analytical methods prescribed by LaMotte and Wells (2000), and (4) assessing accuracy of PMI estimates using an absolute and physiological time measure.

Download

Share

COinS