«Document Title: Assessing Error in PMI Prediction Using a Forensic Entomological Computer Model Author(s): Daniel Slone, Susan Gruner, Jon Allen ...»
The author(s) shown below used Federal funds provided by the U.S.
Department of Justice and prepared the following final report:
Document Title: Assessing Error in PMI Prediction Using a
Forensic Entomological Computer Model
Author(s): Daniel Slone, Susan Gruner, Jon Allen
Document No.: 211760
Date Received: October 2005
Award Number: 2000-RB-CX-0002
This report has not been published by the U.S. Department of Justice.
To provide better customer service, NCJRS has made this Federally- funded grant final report available electronically in addition to traditional paper copies.
Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S.
Department of Justice.
ASSESSING ERROR IN PMI PREDICTION USING A FORENSIC
ENTOMOLOGICAL COMPUTER MODEL
Byrd. It was also enhanced with a maggot mass temperature submodel using data gathered from the linear probe, and individual temperatures of maggot masses. Using this expanded model, we ran simulations using weather data from the Florida test plots, and compared those PMI predictions to actual results found in the field trials.
With the few useable field trials, the time of death estimates have been found to be quite accurate in our limited testing, with preliminary error rates in the 1-2% range. The majority of the field trials were not useable at this time because we have no phenology data for the most common fly found, Phaenicia coeruleiviridis. When phenology is obtained for that species, we can run those trials and obtain a more robust set of error measurements. We also will perform validaiton using cold-case criminal data.
Sensitivity analysis showed that the most sensitive parameter in the model is the growth rate of the maggots. This is followed closely by air temperature, which becomes less important as the mass grows larger and has better internal temperature regulation. Other parameters were less sensitive, such as the rate at which heat contained in the air, soil, or body transfers to other locations. We also analyzed error rates brought about by sampling imprecision and found, unsurprisingly, that larger sample sizes lead to greater precision in the model.
Forensic entomology is the broad field where arthropod science and the judicial system
interact (Hall 2001). The field of forensic entomology has been divided into three areas:
medicocriminal entomology (also referred to as medicolegal entomology), urban entomology and stored product entomology. Information gained from forensic entomology typically is used to determine time of death, place of death and other issues of medical or legal importance (Gordh and Headrick 2001). Urban entomology concentrates mainly on controversies involving termites, cockroaches, and other insect problems accruing to the human environment, whereas stored product entomology involves disputes over arthropods and arthropod parts in food and other products (Hall 2001).
When human remains are found, the most important questions are usually how, when, where and why the person died. Historically, determination of the postmortem interval has been estimated through observation and measurement of body conditions such as core body temperature (Nelson 1999), muscular flaccidity, rigor mortis, lividity, pallor of the skin and others (Smith 1986; Bass 2001; Byrd and Castner 2001). Entomological specimens in medicolegal death investigations can be reliable indicators for estimating the postmortem interval (PMI) in both early and advanced stages of cadaver decomposition (Nuorteva 1977;
Smith 1986; Goff et al. 1988; Kashyap and Pillay 1989; Greenberg 1991; Byrd 1998).
Insects and other invertebrates feeding on carrion form a distinct faunal succession associated with the various stages of decay (Smith 1986). Recognition of the different immature stages of each species involved, together with the knowledge of their rates of development, can give an indication of the PMI (Smith 1986). A forensic entomologist can also determine the age of immature insects, based upon knowledge of the variables regarding insect invasion of human remains. Evaluation and interpretation of entomological evidence at a crime scene can address other complicated issues in addition to time of death, including: season of death, geographic location of death, movement or storage of the remains following death, location of specific sites of trauma on the body, sexual molestation and use of drugs (Haskell et al. 1997).
In case studies conducted in varying temperate and tropical climates, where human remains were exposed to the environment for 2.5 months or less, entomology-based PMI estimates differed by ± 48 hours when compared with the intervals determined by independent corroboration such as confessions and eyewitness testimony (Greenberg 1985; Goff, Omori et al.
1988; Lord 1990; Byrd 1998). Clearly, entomological evidence is the most reliable scientific means of estimating a PMI of 72 hours or more (Kashyap and Pillay 1989; Catts and Haskell 1990; Anderson 2001).
The study of insects important to forensic entomology has been conducted mainly through the use of non-human animal models. Decomposition studies worldwide have used a variety of different carcass types and sizes including: dogs (Reed 1958; Jiron and Cartin 1981;
Early and Goff 1986), cats (Early and Goff 1986), voles (Lane 1975), rats (Greenberg 1990;
Tomberlin and Adler 1998; Faucherre et al. 1999; Kocarek 2001), squirrels (Johnson 1975), foxes (Easton and Smith 1970; Smith 1975), pigs (Payne 1965; Tullis and Goff 1987; Haskell 1989; Anderson and VanLaerhoven 1996; Tessmer and Meek 1996; Richards and Goff 1997;
deCarvalho et al. 1999; Shahid et al. 1999; Davis and Goff 2000; deCarvalho and Linhares 2001;
Wolff et al. 2001; Tenorio et al. 2003), seals (Lord and Burger 1984), guinea pigs (Bornemissza 1956), mice (Putnam 1978; Blackith and Blackith 1989), lizards and toads (Cornaby 1974), raccoons (Joy et al. 2002), turtles (Abell et al. 1982), poultry (Hall and Doisy 1993; Tessmer, This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice.
Meek et al. 1995), sheep (Deonier 1940), rabbits (Denno and Cothran 1975; Tantawi et al. 1996;
Bourel et al. 1999), elephants (Coe 1978), opossums (Goddard and Lago 1985), black bears (Anderson 1998), and impala (Braack 1981). The only faunal succession research on human remains was conducted in Tennessee (Rodriguez and Bass 1983; Catts and Haskell 1990).
Human cadavers are not easily obtainable for detailed decomposition studies. Pigs, Sus scrofa, are omnivorous, have similar gut fauna, are relatively hairless and have skin that is very similar to that of humans (Anderson and VanLaerhoven 1996). The putrefaction of pigs proceeds approximately at the same rate as for human bodies that are of the same torso weight (Campobasso et al. 2001). Haskell’s 1989 study in Tennessee (unpublished) compared the insect community structure and decomposition rates between adult and infant human remains and that of pigs and found no significant difference in the composition of the insect communities in human and pig carcasses (Campobasso et al. 2001). Therefore, twenty-two kg pigs have been recommended as suitable human models for adult decomposition (Catts and Goff 1992).
Biology of Calliphorid Flies Two major groups of insects are predictably attracted to cadavers and provide the majority of information in forensic investigation; the flies and the beetles (Castner 2001). This study focuses on the Family Calliphoridae, commonly called the blow flies, which are the first to find and colonize human corpses. Experimental studies indicate that these flies arrive at carcasses within minutes of their exposure ( Shean et al. 1993; Byrd and Castner 2001).
There are more than 1000 species of blow flies throughout the world. This family includes the green bottle flies (genus Phaenicia), blue bottle flies (genus Calliphora), the screwworm flies (genus Cochliomyia) and the black blow flies (tribe Phormiini). The common name is derived from the manner in which these flies deposit their eggs (Hall 1948). The family name means ‘beauty bearer’ in Greek (Greenberg and Kunich 2002).
Calliphorid flies have highly specialized sense organs on their antennae that are stimulated by putrefaction odors and gases that are released during post-mortem decomposition of organic matter. Studies indicate that some species of Phaenicia are attracted to various organic sulphur compounds, either alone or in combination with hydrogen sulphide, and also by ammonia (Cragg 1956; Cragg and Cole 1956; Ashworth and Wall 1994; Wall and Warnes 1994).
Nilssen used insect flight traps baited with dimethyl trisulphide and found that the chemical was a strong attractant for some calliphorids (Nilssen et al. 1996). Odors from Proteus mirabilis, a bacterial infection that occurs in the fleece in sheep, are attractants to some calliphorid flies (Morris et al. 1998).
Landing behavior of calliphorids is also dependent on visual cues such as color (Wall et al. 1992, Hall et al.1995). Oviposition is elicited primarily by the presence of ammonia-rich compounds, moisture, pheromones, and tactile stimuli (Ashworth and Wall 1994) yet was rarely stimulated by chemicals used alone (Cragg 1956). Unfortunately, the complex interaction of semiochemical and visual cues used for resource location remains little studied in calliphorids (Wall and Fisher 2001).
Blow flies are heliotropic and usually rest at night. Eggs are not usually laid at night although clearly there are exceptions. Green (1951) observed that Calliphora deposited eggs at night under artificial light in slaughter houses. He wrote that “under laboratory conditions it has been found that Calliphora erythrocephala (now called C. vicina), Lucilia sericata and Phormia terrae-novae will all oviposit in total darkness, although Wardle (1930) asserts that blowflies do not oviposit in the complete absence of light”. Greenberg (1990) observed Phaenicia sericata, This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice.
Phormia regina (Meigen) and Calliphora vicina (Robineau-Desvoidy) ovipositing a very small number of eggs on rat carrion. Singh (2001) pointed out that the flies in Greenberg’s experiment probably were resting on a nearby bush and literally crawled over to oviposit on the rat carrion, thus indicating that blow flies were not actively searching for an oviposition site. Nocturnal oviposition has not been observed in large-scale studies in other areas (Greenberg 1990; Byrd 1997; Haskell et al.1997.
Other factors that affect blow fly activity are temperature, size of the carcass, geographical location, humidity, light and shade, seasonal and daily periodicity, availability of food and competition, maggot mass temperature and manner of death.
Description of original “maggot model” Our model of forensic fly phenology was developed using the Matlab/Simulink software simulation package (Figure 1). Mathematically speaking, we use distributed delays (Manetch 1976, van Sickle 1977) to model the developmental delay process. MacDonald (1978) has called this method the ‘linear chain trick’. For each biological stage (egg, larva1, larva2,…) a chain of differential equations is written as,
This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice.
MATERIALS AND METHODSThe use of human corpses for field studies in maggot development is generally not legal or practical, so a substitute decomposition subject was needed. As determined in other decomposition studies (Haskell 1989, Anderson and VanLaerhoven 1996, Campobasso et al.
2001) the rate decomposition and fly colonization in pigs is very similar to that of humans:
therefore, dead pigs, Sus scrofa L, were used as animal models for this study.