G Model

FSI-8010; No. of Pages 9

Forensic Science International xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Forensic Science International

journal homepage: www.elsevier.com/locate/forsciint

Forensic anthropology population data

Human decomposition and the reliability of a ‘Universal’ model

for post mortem interval estimations
Diane L. Cockle a,*, Lynne S. Bell b,1
a
Department of Archaeology, Centre for Forensic Research, Simon Fraser University, 8888 University Drive, Burnaby, BC V5A 1S6, Canada
b
School of Criminology, Centre for Forensic Research, Simon Fraser University, 8888 University Drive, Burnaby, BC V5A 1S6, Canada

A R T I C L E I N F O A B S T R A C T

Article history: Human decomposition is a complex biological process driven by an array of variables which are not
Received 20 January 2015 clearly understood. The medico-legal community have long been searching for a reliable method to
Received in revised form 18 April 2015 establish the post-mortem interval (PMI) for those whose deaths have either been hidden, or gone un-
Accepted 17 May 2015
noticed. To date, attempts to develop a PMI estimation method based on the state of the body either at
Available online xxx
the scene or at autopsy have been unsuccessful. One recent study has proposed that two simple
formulae, based on the level of decomposition humidity and temperature, could be used to accurately
Keywords:
calculate the PMI for bodies outside, on or under the surface worldwide. This study attempted to validate
Regional decomposition
Post mortem interval estimations
‘Formula I’ [1] (for bodies on the surface) using 42 Canadian cases with known PMIs. The results indicated
Forensic anthropology and population data that bodies exposed to warm temperatures consistently overestimated the known PMI by a large and
inconsistent margin for Formula I estimations. And for bodies exposed to cold and freezing temperatures
(less than 4 8C), then the PMI was dramatically under estimated. The ability of ‘Formulae II’ to estimate
the PMI for buried bodies was also examined using a set of 22 known Canadian burial cases. As these
cases used in this study are retrospective, some of the data needed for Formula II was not available. The
4.6 value used in Formula II to represent the standard ratio of time that burial decelerates the rate of
decomposition was examined. The average time taken to achieve each stage of decomposition both on,
and under the surface was compared for the 118 known cases. It was found that the rate of
decomposition was not consistent throughout all stages of decomposition. The rates of autolysis above
and below the ground were equivalent with the buried cases staying in a state of putrefaction for a
prolonged period of time. It is suggested that differences in temperature extremes and humidity levels
between geographic regions may make it impractical to apply formulas developed in one region to any
other region. These results also suggest that there are other variables, apart from temperature and
humidity that may impact the rate of human decomposition. These variables, or complex of variables, are
considered regionally specific. Neither of the Universal Formulae performed well, and our results do not
support the proposition of Universality for PMI estimation.
ß 2015 Elsevier Ireland Ltd. All rights reserved.

1. Introduction decomposition in exceptional circumstances when the death of a

Death investigators face a daunting task to estimate the post
mortem interval (PMI) based on the visible state of decomposition
as the rate is dependent on an array of poorly understudy variables.
The majority of individuals in our society are embalmed or
cremated within a few days of death. A body is found in a state of
* Corresponding author. Tel.: +1 604 317 8803; fax: +1 604 666 2242.
E-mail addresses: dianecockle@hotmail.com, diane.cockle@rcmp-grc.gc.ca
(D.L. Cockle).
1
Tel.: +1 778 782 7240.
marginalized individual goes unnoticed, or the victim of a crime is
intentionally hidden. Without eyewitness accounts, the onus is on
the death investigators to establish the circumstances and timing
of death. The timing of death may not be as critical for those who
have died accidentally or due to natural causes; however, for those
whose lives have been wrongly taken from them, the PMI
estimation can be important. A PMI can limit the extensive list
of missing persons to facilitate a positive identification by
fingerprints, DNA or dental records. At the start of a homicide
investigation, a PMI can provide time ranges for cell phone records
or CCTV footage and in some instances confirm or refute alibis
offered by suspects. The quest to develop a reliable method of PMI

http://dx.doi.org/10.1016/j.forsciint.2015.05.018
0379-0738/ß 2015 Elsevier Ireland Ltd. All rights reserved.

Please cite this article in press as: D.L. Cockle, L.S. Bell, Human decomposition and the reliability of a ‘Universal’ model for post mortem
interval estimations, Forensic Sci. Int. (2015), http://dx.doi.org/10.1016/j.forsciint.2015.05.018
G Model
FSI-8010; No. of Pages 9
e2 D.L. Cockle, L.S. Bell / Forensic Science International xxx (2015) xxx–xxx
estimation has been pursued by forensic pathologists and
taphonomists over the past half century with limited success
[2–11].
Human decomposition is an incredibly complex biological and
chemical process which involves the interaction of a diverse range
of enzymes, bacteria, fungi and protozoa, which each operate
within an optimal environmental range, and set of conditions
[12]. After death, the deprivation of oxygen triggers a chain of
events which results in cell necrosis and the onset of autolysis and
then putrefaction. Enzymes, which are proteins can be destroyed,
crystallized or coagulated in temperature extremes [12]. The
optimal temperature for enzymes is 37 8C with a range between
40 and 50 8C. The principle of the ‘rule of ten’ states that with each
10 8C increase in temperature, the velocity of the chemical reaction
doubles or even triples [12]. Water is critical to chemical reactions
as it moderates the impact of acidity and temperature. Water is
also critical for the action of hydrolyzing enzymes which break
down biological polymers into carbohydrates, lipids and protein
hydrolysis. Once the process of autolysis has created sufficient
quantities of nutrient-rich fluid from cellular disruption, putrefac-
tion can take over to self-digest and liquefy the soft tissue [12,13].
It is generally agreed that temperature, water, acidity, and
availability of oxygen can moderate the internal chemical and
biological progression of decomposition [1,12–14]. What has not
been established is the effect that the external environmental
conditions; manner and cause of death; scene context (burial,
indoors, outdoors, water; clothed, wrapped etc.) or variability
between individuals (age, sex, body size) has on those internal
biochemical variables during the progression of decomposition
[12,15–18]. Most researchers would attribute temperature and
moisture as the most significant dependant variables to the rate of
decomposition [4,11,17–21], however there are others who have
suggested a number of other variables. It has been proposed that
insect and scavenger activity [16,18–20], extent of clothing [4],
burial, soil pH, burial depth [12,18], trauma [18,19], body size
[16,18], sun exposure and rainfall [4,12,17] have a significant
impact on the rate of human decomposition. Due to the
complications of conducting actualistic experimentation with
donated human bodies [13,22,23] and the inability to find a
sufficient number of retrospective cases with similar intrinsic and
contextual conditions [11,18,24–26], much of the research
conducted on human decomposition has used pigs as a proxy
for humans [15,27–33]. Without the corresponding research to
establish the similarities or differences between human and pig
decomposition, we have merely created a large compendium of
knowledge on pig decomposition. Therefore, the question of which
variables impact the rate of decomposition and whether or not
those variables vary from one region to another is still unanswered.
There have been a number of studies conducted in an attempt to
understand the rate and variability of decomposition within a
specific geographic region [4,18,25,26,34]. The regional study in
Arizona [18] was the largest with 189 cases and it still sets the
standard for the description of decomposition in an arid
environment. The authors of this study did caution others not to
apply their regionally specific classification system or general
schedule of progression to any other geographic region [18]. Many
of the other regional models suffer from small sample sizes
[25,26]; or a focus on bone decomposition as opposed to soft tissue
decomposition [4]. In relation to other studies with a regional
concentration of cases, the rate of decomposition appears to be a
by-product of their original focus [11,35].
There are some researchers who have suggested that it might be
possible to take some of these regional models and attempt to
provide a global perspective of human decomposition. Rhine and
Dawson [4] determined that based on their New Mexico study, in
combination with the studies from the Northwest [35] and
Please cite this article in press as: D.L. Cockle, L.S. Bell, Human decomposition and the reliability of a ‘Universal’ model for post mortem
interval estimations, Forensic Sci. Int. (2015), http://dx.doi.org/10.1016/j.forsciint.2015.05.018
Southwest [18], that the data provided a general template for
rates of decomposition which could be applied continent-wide as a
first order of approximation. Although some practitioners may
attempt to provide a generalized PMI estimation for found remains
based on their personal experience, there currently no scientifi-
cally recognized PMI estimation method for any specific or general
geographic region.
The search for a method to establish the PMI has been
undertaken not only by forensic taphonomists, but also forensic
pathologists and entomologists, all with disappointing results.
Historically, the most common method to establish the PMI either
at the scene or autopsy, was the temperature method [36,37]. A
comprehensive validation study of this method was undertaken in
the 1990s which found that there were flaws in the actual models
and algorithms used and a subjectivity in measurements taken
which contributed to errors in the final calculation [37]. Methods
once used by Forensic Pathologists to estimate the PMI such as;
potassium levels in the aqueous vitreous or digestion levels of
stomach content have also been demonstrated not to be as reliable
as once thought [7,9]. Until recently, PMI estimations using
forensic entomology had been regarded as relatively reliable,
however recent research in Canada [33] has suggested that there is
an unpredictable delay in the arrival of insects to decedents both
indoors and outdoors regardless of season, which introduces an
unknown error rate with any PMI estimate. Any general PMI
estimations in Canada are provided reluctantly and with trepida-
tion with the belief that witness statements are still the most
reliable means to determine a time of death [38].
Based on our current level of understanding of the variability of
human decomposition, many forensic anthropologists tend to
accept that the estimation of a PMI based on the physical state of
the body is inherently unreliable and constitutes an educated
guess based the individual’s previous experience
[10,18,19,39,40]. There are those who assert that it is possible
to predict the PMI with a relative degree of reliability and accuracy.
Vass [1] has proposed that two basic formulae can used to reliably
estimate the PMI for any body found under or on the surface
anywhere.
The proposed Formulae are based on the assumption that soft
tissue takes a standard amount of time or accumulated tempera-
ture (1285 ADD) to decompose, and that only atmospheric or
environmental temperature and humidity have any impact on the
rate of decomposition. A previous study by Vass [2] has established
that it takes a total of 1285 (ADD) accumulated temperature for all
soft tissue to decompose. Therefore, if the temperature averaged
20 8C every day, it would take a maximum of 64 days for all soft
tissues to decompose and disappear. Vass [1] has identified only
four variables, which in his opinion has any measurable impact on
the rate of decomposition: temperature, moisture, pH and partial
pressure of oxygen. A constant factor of 0.0103 was calculated by
Vass [1] for the effect of moisture on decomposition based on his
personal experience observing human decomposition at the
Archaeological Research Facility (ARF) in Knoxville, Tennessee
[1]. The proposed method scores the level of atmospheric or soil
humidity between 1 and 100 at the site of discovery on the day the
corpse is discovered. The ambient temperature at the scene is
taken on the day of discovery or averaged over several days and
corrected by a several day comparison to the nearest National
Weather Service station [1]. As Vass [1] has already calculated how
much temperature (ADD) it takes for all soft tissue to decompose,
the method uses certain values for humidity and lack of oxygen in
burials to compensate for the conditions at the site and calculate an
accurate PMI.
In an attempt to address the variability of external pH or oxygen
pressure in a burial, Formula II utilizes a ratio of 4.6 to represent
the overall deceleration in the rate of burial compared to surface
G Model
FSI-8010; No. of Pages 9
D.L. Cockle, L.S. Bell / Forensic Science International xxx (2015) xxx–xxx
decomposition. Vass developed this ratio based a study conducted
with four buried cadavers at ARF in 1985 [22] and the
decomposition of three pigs in clays soils in England [28]. This
4.6 value was developed by Vass as a result of numerous other
unpublished experiments conducted over many years [1] presum-
ably at ARF in Tennessee. The contribution of pH in the process of
human decomposition, although identified in the introduction as
one of the four widely recognized factors that influence the rate
and ultimate completeness of the decomposition process [1] was
unfortunately not defined or quantified in the methodology. For
the classification of decomposition, Vass assessed the extent of soft
tissue decomposition using a scale of 1–100%. This method was
developed based on years of observation and experience [1] at ARF.
Vass states that these formulas have been applied to many cases
worldwide with remarkable success [1]. He suggests, that with the
application of these two formulas, Medical Examiners or forensic
personnel can estimate the percentage of decomposition, and
therefore the PMI to very accurate levels [1].
The aim of this study was to validate these two formulae
proposed by Vass [1] using a set of 118 known Canada
retrospective cases. A group of 42 outdoor surface cases with
known PMIs were selected according to the criteria set out by Vass
[1] to test the accuracy of Formula I for surface decomposition. A
group of 96 surface and 22 buried cases were used to determine
whether decomposition in a burial was 4.6 slower than exposure
on the surface. By validating these two formulae, this study would
be able to comment on the appropriateness of universal rules or
concepts in our attempt to understand and predict human
decomposition.
2. Materials and methodology
A total of 118 outdoor and buried cases were taken from a larger
study on human decomposition in Canada to assess the ability of
the two Universal Formulas [1] to accurately estimate the PMI for
found bodies in Canada. These cases were randomly selected from
a group of 7328 Canadian death investigations using the Violent
Crimes Linkage System (ViCLAS) and the Ottawa Police Service
(OPS) sudden death databases. Selection of the cases was based on
the quality of data and photographs and availability of a known
PMI. For all cases in the dataset, the PMI was established by
independent means such as phone, computer or CCTV records or
confirmed eye-witness accounts. Ethics permission for this study
was granted by Simon Fraser University (SFU) and permissions
granted to access and mine the data from the police files (including
coroner/medical examiner, forensic identification, forensic an-
thropology and forensic lab reports) from the Royal Canadian
Mounted Police (RCMP) and Ottawa Police Service (OPS).
The cases used for the present study include solved homicides
or suspicious deaths with found bodies located outside on the
surface (96 cases) or buried in a clandestine grave (22 cases). An
outdoor case is defined as such when the body was located
outside of a temperature controlled structure, exposed to
ambient temperature and humidity ranges. Of the 22 buried
cases, 17 cases were buried in a shallow grave (some portion of
the body exposed and visible) and 5 cases deeply buried (not
visible). The cases were selected from British Columbia, Alberta,
Saskatchewan, Ontario and New Brunswick, however the
majority of the outdoor cases (81%) in this study are from
British Columbia and Ontario.
3. Methodology
In order to estimate the PMI using Formula 1 for above ground
cases; information relating to the level of decomposition, the
average temperature and humidity conditions were needed for the
Please cite this article in press as: D.L. Cockle, L.S. Bell, Human decomposition and the reliability of a ‘Universal’ model for post mortem
interval estimations, Forensic Sci. Int. (2015), http://dx.doi.org/10.1016/j.forsciint.2015.05.018
e3
duration of the PMI or the day of discovery. The level of
decomposition was classified using photographs of the body at
the scene on the day of discovery. The original police operational
files were accessed to view the scene photographs. With
permission from the RCMP and OPS, digital copies were made of
the relevant scene and autopsy photographs. The images were
captured using a Cannon colour CanoScan 4400F high resolution
flatbed scanner for printed photographs. A Nikon Coolscan V ED
negative scanner was used for film negatives and any digital
images were copied from the file CD or DVD to a secure external
hard drive. The case photographs were numerically catalogued and
separated from all case information. Any information which could
be used to identify the decedent or case was removed.
The level of decomposition was classified for each decedent
based using the crime scene and autopsy photographs by one
author (DC) for all cases. The classification system was developed
for the larger human decomposition study and designed to
describe decomposition found within the environmental ranges
of Canada [41]. This classification system described soft tissue
decomposition and assigned a numerical stage to the visible
progression of autolysis and putrefaction: stage ‘00 as no visible
signs of post mortem changes, stages 1 and 2 include autolysis;
stages 3–5 putrefaction and stages 6–8 the final stages of
liquefaction, soft tissue removal and skeletonization.
The average temperature and humidity conditions for each case
for the duration of their post mortem interval were calculated
using data available from the ‘‘Canadian National Climate Data and
Information Archive’’, which is available on the internet (http://
climate.weatheroffice.gc.ca/climateData/canada_e.html). The
temperature and humidity data was taken from the geographically
closest weather station. The Accumulated Degree-days (ADD)
score, was calculated for each case using the methodology
described by Megyesi et al., [11]. The temperature, humidity
and stage of decomposition for each case were exported from Excel
into Minitab 16 Statistical Software, version 16.2.2.0 (by Minitab
Inc.) for analysis.
3.1.1. Formula 1 (Vass [1]) – above ground – aerobic decomposition
A subset of 42 cases was selected from the 96 outdoor cases
from the larger study to test the ability of the ‘Universal’ Formula I
[1] to estimate the PMI for bodies exposed outside in Canada. The
criterion provided by Vass [1] was used select 36 cases. These cases
had a PMI of 1 day or more, no evidence of scavenging, not exposed
to freezing temperatures, and without adipocere formation
[1]. Due to the prevalence of colder environmental conditions in
Canada, a further six cases with longer PMIs which had been
exposed to freezing temperatures, were selected to determine the
impact that cold temperatures had, if any, on the accuracy of
Formula I [1]. These six cases were exposed to warm temperatures
during their initial deposition and then cold and freezing
temperatures (average temperatures of 4 8C or less) for the
remainder of their post mortem interval.
Vass [1] used a cumulative percentage to express the level of
decomposition for the calculation of the PMI in both ‘Universal’
Formulas. There is unfortunately, no explanation as to what visible
signals or states are used to assess the overall extent of
decomposition and express it in a numeral range from 0 to
100. In order to approximate the same methodology used by Vass
[1], the classification system used to describe the level of
decomposition for Canadian cases, was adapted so that the
numerical stage of decomposition could be expressed in a
percentage format. For each case, the numerical stage of
decomposition assigned from 1 to 8 was converted into a 12.5–
100% range by multiplying each stage of decomposition by a factor
of 12.5 (i.e., stage 2 became 25% decomposed).
G Model
FSI-8010; No. of Pages 9

e4 D.L. Cockle, L.S. Bell / Forensic Science International xxx (2015) xxx–xxx

The percentage of decomposition, the temperature (ADD score) ratio was then calculated as an expression of the extent of error; for
and average humidity were used within Formula I (PMI Aerobic) example an estimated PMI of 40 days, when the actual PMI was
[1] to calculate the estimated PMI: 20 days was express as a ration of ‘ 20 over estimation. The results
have been provided in Table 1 and scatter plot formats in Fig. 1.

1285  ðdecomposition=100Þ 3.1.2. Formula II – below ground – anaerobic decomposition

PMI Aerobic ¼
0:0103  temperature  humidity
The values of ‘12850 have been used by Vass to represent the
empirically determined ADD value at which soft tissue decomposi-
tion ceases [1]. The 0.0103 value represents the empirically
determined measure of the effect of moisture on decomposition
rates [1]. The percentage of decomposition for each of the cases in
this study was divided by 100 and multiplied by 1285. This number
was then divided by the multiplication of the average temperature,
humidity and 0.0103. The results of the calculation for Formula I
produced an estimation in PMI days (24 h period). The PMI
estimation was then compared to the actual known PMI for each
case. The error rate was calculated in actual PMI days: ‘number of
days’ difference between the actual PMI and estimated PMI. The error
Formula II was designed by Vass [1] to provide accurate PMI
estimations for bodies recovered from sub surface graves. The
formula requires the soil moisture at the site on the day the corpse
was discovered as well as the temperature of the soil on that day.
These values were unavailable for the cases in this study, and as
such the same method used to evaluate Formula I could not be
used for Formula II. To calculate the PMI estimate for buried bodies,
Vass uses a constant ratio of 4.6 which ‘‘represents a slowdown in
the rate of decomposition due to lack of oxygen’’ [1]. This study
examines the relative speed of decomposition to determine
whether the rate of decomposition is 4.6 slower for bodies under
the ground compared to on the surface in Canada.

Table 1
Comparison of known and predicted PMI calculations using the Universal Formula I (Vass [1]) for cases in Canada.

Case Location Ecozone Decomp Average Average Actual PMI PMI using Error in days
score %* temp humidity formula I** and ratio diff.

1 Fort Chipewyan, AB 9 25 13.3 66 3 36 +33 ( 12)

2 Ottawa, ON 8 25 6.5 82 2 59 +57 ( 29.5)
3 Salmo, BC 14 12.5 4.4 85 5 43 +38 ( 8.6)
4 St. Denis, SK 10 37.5 17.9 55 2 48 +46 ( 24)
5 Richmond, BC 13 50 16.7 85 17 44 +27 ( 2.6)
6 Penticton, BC 14 87.5 5.6 58 25 336 +311 ( 13.4)
7 Coquitlam, BC 13 87.5 15.6 84 68 83 +15 ( 1.2)
8 Langley, BC 13 62.5 16.3 82 9 58 +49 ( 6.4)
9 Squamish, BC 13 75 14.5 83 31 78 +47 ( 2.5)
10 St. Denis, SK 10 37.5 15.4 83 11 37 +26 ( 3.4)
11 Dawson City, YK 12 12.5 17.3 73 2 13 +11 ( 6.5)
12 Burnaby, BC 13 12.5 13.2 95.5 2 13 +11 ( 6.5)
13 Port Alberni, BC 13 12.5 17 67 2 14 +12 ( 7)
14 Kamloops, BC 14 12.5 21 44.6 42 17 S25 ( 2.5)
15 Penticton, BC 14 12.5 14 63 3 18 +15 ( 6)
16 Midway, BC 14 12.5 21.5 41 2 18 +15 ( 6)
17 Prince George, BC 14 12.5 17 40.1 3 24 +21 ( 8)
18 Salmo, BC 14 12.5 4 78.6 5 52 +47 ( 10.4)
19 Chatham Head, NB 7 12.5 10.9 71.7 3 21 +18 ( 7)
20 Neguac, NB 7 12.5 14 65 2 18 +16 ( 9)
21 Newcastle, NB 7 12.5 22.1 75 2 10 +8 ( 5)
22 Ottawa, ON 8 12.5 20.7 79.7 3 10 +7 ( 3.33)
23 Ottawa, ON 8 12.5 11.4 52.3 3 27 +24 ( 9)
24 Ottawa, ON 8 12.5 18 85 2 11 +9 ( 5.5)
25 Ottawa, ON 8 12.5 20.3 70 2 11 +9 ( 5.5)
26 Ottawa, ON 8 12.5 22 65.6 2 11 +9 ( 5.5)
27 Parksville, BC 13 25 17.9 53.5 5 33 +28 ( 6.6)
28 Cowichan, BC 13 25 16 66.5 4 29 +25 ( 7.3)
29 Fort Chipewyan, AB 9 25 14.6 48.3 3 44 +41 ( 14.7)
30 Ottawa, ON 8 25 17.9 53.8 4 32 +28 ( 8)
31 Surrey, BC 13 62.5 10 71.7 31 110 +79 ( 3.5)
32 Courtney, BC 13 62.5 9.7*** 72 340 113 227 ( 3)
33 Ottawa, ON 8 62.5 23.1 89 6 38 +32 ( 6.3)
34 Ottawa, ON 8 62.5 22.2 78 13 45 +32 ( 3.5)
35 Prince George, BC 14 75 11.3*** 62 229 134 95 ( 1.7)
36 Prince George, BC 14 75 9.1*** 60 395 172 223 ( 2.3)
37 Ottawa, ON 8 75 20.2 74 32 63 +31 ( 2)
38 Ottawa, ON 8 75 22.4 66.4 9 63 +54 ( 7)
39 Port Alberni, BC 13 87.5 9 70 67 174 +107 ( 2.6)
40 Chilliwack, BC 13 87.5 9*** 69 336 177 159 ( 1.9)
41 Surrey, BC 13 87.5 14*** 63 260 125 135 ( 2.1)
42 Burns Lake, BC 14 87.5 10.5*** 61 247 171 76 ( 1.4)

Results in red: represent an overestimation of the actual PMI.

Results in blue: represent an underestimation of the actual PMI.
NB: each of the longer PMI cases (with the exception of case 14) with exposure to colder and freezing temperatures resulted in an underestimation of the actual PMI.
*
The decomposition score from 0 to 8 has been multiplied by 12.5 to equate to the 0–100% scale used by Vass [1]. The average temperature and humidity have been
calculated from the day of death to discovery.
**
This represents the calculation of the post mortem interval using Formula I for bodies above the surface [1].
***
Cases with longer PMI in days were included if the first 30 days resulted in an average positive temperature. The average temperature provided for these cases is for the
first 30 days only.

Please cite this article in press as: D.L. Cockle, L.S. Bell, Human decomposition and the reliability of a ‘Universal’ model for post mortem
interval estimations, Forensic Sci. Int. (2015), http://dx.doi.org/10.1016/j.forsciint.2015.05.018
G Model
FSI-8010; No. of Pages 9

D.L. Cockle, L.S. Bell / Forensic Science International xxx (2015) xxx–xxx e5

Comparison of Actual PMI and Formula I (Vass 2011) estimated PMI

400 Variables
Actual PMI
PMI using Formula I

300

PMI in Days 200

100

0 10 20 30 40
Case Number

Fig. 1. Comparison between the actual PMI (in black) and the estimated PMI using Formula I (Vass [1]) (in red) for each case. Formula I [1] consistently underestimated the PMI
for cases exposed to cold or freezing temperatures. The cases exposed to colder temperatures are those which were deposited in warmer conditions, but have been exposed to
colder seasonal conditions with the passage of time in Canada. Formula 1 consistently overestimated the PMI for cases selected using the criteria provided by Vass [1].(For
interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

A set of 22 buried cases was compared to 96 surface cases. As
bodies were discovered at a range of decompositional stages, the
cases were grouped by stage of decomposition and the average
time (PMI in days) and temperature (ADD score) calculated for that
stage of decomposition for the surface and buried groups. Due to
the smaller sample sizes within the burial set, it was only possible
to compare the differences in the rates of decomposition for stages
0, 1, 3, 5, 6 and 7 [41] (There were no burial cases discovered at
decompositional stages 2, 4 or 8). The average ADD scores and PMI
for each group at each stage of decomposition was compared. If
there was a difference in the average time (PMI in days) or
temperature (ADD score) for that stage of decomposition between
the surface and buried group, the difference was expressed as a
ratio to present the number of times that the rate of decomposition
for one group was faster or slower (in time and temperature) than
the other.
4. Results
Fig. 1 demonstrated that Formula I underestimated the PMI for
cases exposure to cold temperatures and underestimated those
exposed to warm temperatures, but also that there is an increase in
the error rate with a corresponding increase in the PMI.
This study has found that the rate of decomposition for buried,
compared to surface cases is variable and dependant on the stage
of decomposition. For some stages of decomposition, the progres-
sion of decomposition was faster for buried compared to surface
cases (Table 2). The buried cases reached stage 1, 1.3 faster in days,
with half as much temperature (ADD score) compared to the
surface cases at the same stage. For the stages of putrefaction
however, the surface cases did demonstrate a faster rate of
decomposition compared to the buried cases. It took considerably
more temperature and more time for buried cases to commence
the onset of putrefaction in a burial compared to a surface
environment. The rate of decomposition for the later stages of
liquefaction was only slightly faster for buried cases compared to
surface cases, both with longer PMIs (Table 2).

In brief, this study determined that Vass’s Formula I [1] either

over-estimated the actual PMI of bodies found on the surface in
Table 2
Canada in warmer temperatures or underestimated those exposed Ratio differences for the rate of decomposition of surface compared to buried bodies
to colder temperatures outside (Table 1). It was also determined in the Canadian dataset.
that the ratio of 4.6 [1] used in Formula II to express the delay in the
Stage of Ratio Ratio Relation of surface
speed of decomposition under versus on top of the ground could decomposition difference difference to buried bodies
not be validated. for PMI for ADD
The degree of over or underestimation was calculated to in days score
determine if a constant error rate existed which could account for 0 3 6.7 Buried faster than surface
the environmental and ecological differences between the region 1 1.3 2 Buried faster than surface
for which the formula was designed (Knoxville Tennessee) and 3 0.08 9.25 Surface faster than buried
Canada. For the 36 cases in the Canadian dataset with shorter PMIs 5 9.6 89 Surface faster than buried
6 6.4 163 Surface faster than buried
and no exposure to cold or freezing temperatures, Formula I 7 1.9 1.5 Buried faster than surface
overestimated the PMI in days an average of seven times, with an
Note: The ratio difference is the number of times decomposition is faster or slower
unpredictable error rate ranging from 1.2 to 30 times.
from the surface compared to the buried bodies for that stage of decomposition. The
For the six cases with longer PMIs with exposure to colder or results are provided in time (PMI days) and temperature (ADD score). The results
freezing temperatures, Formula I provided an estimated PMI which provided in blue represent the stage of decomposition which was found to be
was on average, half of the actual length of time from death to achieved with a smaller amount of time and temperature for buried versus surface
discovery. The estimations provided by Formula I [1] resulted in an cases. The results provided in red represent those stages of decomposition which
were found to be achieved with a smaller amount of time and temperature for cases
inconsistent error rate and underestimated of the actual PMI from on the surface compared to those at the same stage of decomposition under the
113 to 223 days. The disparity between the PMI estimated by surface. N = 22 for buried cases and N = 96 for surface cases. Not all stages were
Formula I and the actual PMI for each case is provided in Fig. 1. represented by the 22 buried cases.

Please cite this article in press as: D.L. Cockle, L.S. Bell, Human decomposition and the reliability of a ‘Universal’ model for post mortem
interval estimations, Forensic Sci. Int. (2015), http://dx.doi.org/10.1016/j.forsciint.2015.05.018
G Model
FSI-8010; No. of Pages 9
e6 D.L. Cockle, L.S. Bell / Forensic Science International xxx (2015) xxx–xxx
As the rate of decomposition is variable depending on the stage
of decomposition for cases in Canada, a constant rate of 4.6 cannot
be used to express the ratio of deceleration in the rate of
decomposition of a buried body discovered at any stage of
decomposition. As the validity of the 4.6 ratio used for Formula II
has not been validated here, then PMI estimations for cases in
Canada using this formula would produce erroneous results, and
therefore the idea of ‘universality’ is too, unsupported.
5. Discussion
It has been demonstrated therefore, that the formulae proposed
by Vass [1] based on research conducted on decomposition in the
Knoxville Tennessee region of the United States of America, cannot
be successfully applied to cases within Canada. The reasons for this
may be two fold; that not all variables contributing to the rate of
decomposition have been identified, and that these variables may
differ from one geographic region to another.
5.1. Freezing temperatures
One of the main differences between the southern United States
and the majority of Canada, is that Canada experiences more
temperature extremes during winter and summer. These temper-
ature extremes especially during winter, may have a significant
impact on the rate of decomposition for an overwintered body
thawing in spring. There have been a number of studies which have
examined the impact of cold temperatures on tissue deterioration
or rates of decomposition [25,42–46]. A study conducted in
Australia and Hawaii [45,46] conducted controlled experiments on
hard and soft tissue in an attempt to identify changes caused by
freezing. The study found that freezing small disassociated soft
tissue samples did not impact the re-thawed rate of decomposition
or the soil chemistry compared to unfrozen samples. However, this
experiment was conduct to determine whether freezing would be
deleterious to forensic chemistry analysis as opposed to the overall
rate of decomposition for an intact organism.
There have been a few studies using whole animal proxies in the
northern United States of America [42,44] as well as one study which
examined human decomposition in cold environments [25] in the
Edmonton region of Alberta, Canada. Extreme scavenging for the
20 Alberta cases meant that there was very little remnant soft tissue
and it was not known if the freezing process had accelerated the rate
of decomposition upon thawing [25]. Micozzi’s [42] has asserted
that freezing temperatures will act as a biocide to eliminate or slow
down endogenous gut flora and thus result in a slower rate of
decomposition, which may result in the re-colonization of aerobic
bacteria on the body from external sources. In the larger study of
Canadian retrospective cases [41], there was one case which
exhibited a deceleration in the rate of decomposition and external
colonization of the body by algae during cold and freezing
temperatures. This particular body was exposed to freezing
temperatures (to a minimum of 118) with an average temperature
of 48 during the 37 days post mortem interval. There was no visible
sign of tissue change due to autolysis or putrefaction; there was
however an obvious colonization of the tissue by an unidentified
algae. This visible appearance of this case is consistent with
Micozzi’s [42] prediction that bodies exposed to extreme cold will
decompose from the ‘outside in’ versus the ‘inside out’.
The implication of these results is that death investigators in
Canada must consider the detrimental effect of freezing or cold
temperatures on putrefaction and the later stages of decomposi-
tion. A body which has overwintered and subject to freezing
temperatures may have a different rate of tissue liquefaction
compared to a body with an equivalent overall ADD score not
exposed to freezing temperatures. There may be instances where a
Please cite this article in press as: D.L. Cockle, L.S. Bell, Human decomposition and the reliability of a ‘Universal’ model for post mortem
interval estimations, Forensic Sci. Int. (2015), http://dx.doi.org/10.1016/j.forsciint.2015.05.018
frozen body will not putrefy. In the cooler moist and dark
environments of western Canada, there are also instances of bodies
colonized by fungi or algae in a visible fresh state after a long post
mortem interval.
5.2. Humidity
The variability in the rate and trajectory of decomposition may
be not only the result of temperature extremes, but also humidity.
As already discussed, moisture is critical to the internal processes
of both autolysis and putrefaction. The lack of external moisture
results in the preservation of tissue with the development of
mummification. Due to the high temperatures and low humidity
rates, mummification was an inevitable end state for the bodies
decomposing outdoors in Arizona [18]. In Canada, instances of
mummification are seen more commonly with indoor cases in a
climate controlled environment [41]. With higher levels of
humidity, it took on average three times more time to reach the
end stages of decomposition in Canada compared to Arizona.
However in Arizona the end state was generally considered to be a
mummified envelope over hard tissue, whereas in Canada soft
tissue did liquefy leaving only mummified connective tissue. It was
noted however, that there was a difference in the overall rate of
decomposition from one Ecozone to another within Canada. The
main environmental differences between those two Ecozones were
that one had higher levels of humidity in summer and more
temperature extremes compared to the other Ecozone with lower
average yearly humidity with a more temperate climate [41]. It
would seem therefore, that based on temperature and humidity
alone, there is variability in the rate and trajectory of decomposi-
tion between one geographic region and another.
5.3. The progression of decomposition in a burial
In an attempt to bring order to the chaos of human
decomposition, taphonomists have attempted to develop ‘rules
of thumb’ for the rate of decomposition in different contexts. Fisher
[47] has suggested that the progression of decomposition for a
body decomposing on the surface for 1 week is equivalent to
2 weeks in the water or 8 weeks in the soil. However Rhine and
Dawson [4] have stated that it is a generally accepted rule, that
1 week on the surface equates to 2 weeks in water, and 3 in a burial
to achieve the same level of decomposition. These two rules of
thumb are based to some extent, on the early research conducted
at ARF in Knoxville, Tennessee [22,40]. Rodriguez and Bass [22]
conducted a ground breaking experiment with six un-embalmed
cadavers at ARF. They evaluated the decomposition of these six
corpses buried at a variety of depths for a period of 2 years. Two of
the corpses had been autopsied and all were placed in a variety of
positions with and without clothing. The bodies were exhumed at
regular intervals for evaluation. The bodies were described as
progressing through stages of ‘‘tissue shrinkage’’ [22] and
adipocere development with skeletonization of the extremities
and head. They concluded by stating that ‘‘the decomposition of
buried human cadavers occurred at a much slower rate than that of
cadavers placed above the ground’’ (Ibid: 849). Rodriguez and Bass
also suggested that the deeper the burial, the less oxygen available
and the slower the rate of decomposition [22,40]. Shepherd also
stated that burial would slow the rate of decomposition
considerably [48]. It may be that the rate of decomposition is
depressed due to lower temperatures and lack of oxygen, however
the question must be asked whether any other variable has an
impact and whether the same rule of thumb can be applied for to
all environmental regions.
The impact of depth of burial has been proposed as a significant
contributing variable to decomposition in burials [22]. There are
G Model
FSI-8010; No. of Pages 9
D.L. Cockle, L.S. Bell / Forensic Science International xxx (2015) xxx–xxx
those who assert that a shallow burial may accelerate the rate of
decomposition compared to deeper internment based on the
absence of bacteria [49]. The bacterium that decomposes the body
from outside in, is limited at deeper stratas due to the reduction in
available oxygen and organic material. Knusel et al., [50] proposes
that the depth of burial is irrelevant and suggests that the presence
of porous and permeable the soil, which allows the free exchange
of water and oxygen, is more important to the rate of decomposi-
tion. This study found that depth of the burial was not a significant
contributing variable, and more importantly, the rate of decom-
position did not appear to be constant through the stages of
autoloysis and putrefaction. For the cases in this study, the rate of
decomposition for the early stages (stages 0 and 1) was slightly
faster for the buried cases, at a ratio of 1.3–3 times, compared to
the cases on the surface (Table 2). From stage 5 onward, the time
spent in each stage of decomposition was extremely variable. From
stages 6 to 8, the buried cases took 5.36 times more time (PMI days)
and 87 times more temperature compared to the rate of
decomposition on the surface (Table 2). The cases which were
deeply buried achieved stage 6 decomposition after an average of
191 days, compared to 203 days for shallowly buried cases
[41]. The rule of thumb that decomposition in a burial is a constant
8 times, 3 times or 4.6 times slower than surface decomposition
was not validated in this study. The assumption that the rate of
decomposition decreases with the increasing depth of burial was
also not validated.
6. PMI estimations
The notion that the rate of human decomposition is subject to
a limited number of external environmental variables common to
all geographic regions was examined during this study with
discouraging results. It was found that Formula I cannot be used
to accurately or reliably estimate the PMI for bodies in Canada.
The Formula will either overestimate the actual PMI for bodies
exposed to warm temperatures, or dramatically underestimate
those exposed to colder temperatures in Canada. The extent, or
ratio of over or underestimation for Formula I is unpredictable
and inconsistent (Table 1). In the larger study, the rate of
decomposition for bodies exposed to cold temperatures com-
pared to bodies not exposed to the cold with the same ADD score
showed a slower overall rate of decomposition [41]. The
differences in ‘moisture’ between one region and another could
affect the ratio of 0.0103 calculated by Vass for Tennessee [1]. The
average humidity levels decrease with each increase in the
geographic latitude. Warmer air can suspend water vapour to a
much larger extent that colder air. The lower average tempera-
tures found in more northern latitudes may therefore have a
negative impact on the relative humidity levels and the
contribution that humidity makes to the progression of decom-
position. Due to the higher average levels of both temperatures
and relative humidity in the Knoxville Tennessee, the ‘1285 ADD’
calculation for the end of soft tissue decomposition may be
dependent on higher levels of humidity and is therefore
regionally specific. This point is illustrated with 10 cases in this
study which had an ADD score well in excess of 1285 ADD with
ongoing soft tissue liquefaction. Five of these cases were
discovered in later stages of putrefaction (average ADD score of
2337) and the other five in a state of putrefaction in combination
with mummification or adipocere formation (average of
3673 ADD). The constant factor of ‘0.01030 calculated by Vass
[1] for the contribution that moisture makes to decomposition in
Knoxville Tennessee, may not be a constant for other geographical
regions in different latitudes.
This study also illustrated that the rate of decomposition in a
burial is not a constant 4.6 times slower compared to the rate on
Please cite this article in press as: D.L. Cockle, L.S. Bell, Human decomposition and the reliability of a ‘Universal’ model for post mortem
interval estimations, Forensic Sci. Int. (2015), http://dx.doi.org/10.1016/j.forsciint.2015.05.018
e7
the surface (Table 2). This original ratio of 4.6 for burial
decomposition was established by Vass [1] using a small sample
size (six bodies) [22] in one specific geographic region. A total of
118 cases from two main Canadian Ecozones (Ecozone 13; Pacific
Maritime and 8; Mixedwood Plain) were used to validate this
ratio with negative results. The rate of decomposition was found
to be variable depending on the stage of decomposition.
Autolysis on the surface appeared to progress at a similar speed
to that in a burial, however the onset and duration of
putrefaction took considerably longer for the buried cases.
Based on these results, it would seem that any attempt to use
Formula II [1] to predict the PMI for buried cases in Canada
would produce erroneous estimations. It would also appear that
the depth of burial is not as significant as perhaps soil type and
drainage. It is also likely that the variability in moisture levels in
the soil as well as temperature plays a substantial role in the
progression decomposition.
There is always the temptation for any researcher to attempt to
search for patterns or laws to understand complex processes. In
this instance Vass [1] identified only four variables: temperature,
moisture, pH and partial pressure of oxygen, to attempt to
calculate standard values to represent their contribution to the
process of decomposition. The study attempted to validate one of
these two formulae with limited success. The accuracy of Formula
II was tested using three buried cases with relatively long PMIs,
presumably from the Knoxville, Tennessee area. The results of
Formula I estimation were as follows: 23 months (actual
22 months), 18.2 years (actual 18 years), and 5.4–6.5 years (actual
8 years). For two of the cases the PMI estimates were within a few
months of the actual PMI, however the estimate for the last cases
was in error by 2.6–1.5 years. These results suggest that there may
still be an imperfect understanding of all of the variables which
affect the rate of decomposition even within the Knoxville
Tennessee region.
The results of this study seem to suggest that understanding
the variables which affect human decomposition may be more
complex than previously thought. There does appear to be an
impact created by regionally specific humidity and temperature
levels. It may also be that these are not the only contributing
variables to the progression of decomposition. The individual’s
age, manner of death and level of exsanguination, alcohol and
drug intake as well as immediate context such as clothing,
coverings, burial, in water, indoors or outdoors may also play a
significant role in the rate and trajectory of human decomposition
[41].
The failure of the universal formulae to accurately estimate the
PMI for cases in Canada suggests that the quest for a universal
method is still ongoing. There remains the hope that chemical
analysis of decompositional fluids can produce a reliable means of
PMI estimation [13,28,51–53], however the reliability and
practicality of these methods for a medico-legal context have
yet to be established. With our current level of understanding of
human decomposition, we may not be at the point of discerning
universal patterns. With more retrospective studies and experi-
mentation, we may however be able to create regional specific
models by identify variables particular to that area and context.
7. Conclusion
This study attempted to validate the two Universal Formulas
proposed by Vass [1] using 118 Canadian cases with discouraging
results. Formula I (for surface decomposition) was tested using
Canadian 42 cases and found to either overestimate the known PMI
by as much as 30 times for bodies decomposing in warmer climates
or underestimate the PMI as much as 3 times for bodies exposed to
colder temperatures. The ratio of 4.6 proposed by Vass [1] to
G Model
FSI-8010; No. of Pages 9
e8 D.L. Cockle, L.S. Bell / Forensic Science International xxx (2015) xxx–xxx
represent the standard deceleration in the rate of burial versus
surface decomposition was not validated by the Canadian
dataset. The rate of decomposition for a group of 22 burial cases
was compared to the rate of decomposition for 96 cases on the
surface in Canada. It was found that the rate of decomposition
was not constant in either environment, however the rate was
comparable during the early stages of decomposition. The
progression of putrefaction was much slower for the burial
compared to surface cases. This study suggests that the use of
these formulas for any cases in Canada would result in a
significant error in the PMI estimation for cases found on, or
under the surface. The Universal formulas therefore may not be
as universal as previously thought.
There is a complex of environmental, immediate context and
intrinsic variables which affect manner and speed by which we as
human beings decompose. Our level of understanding of human
decomposition given the plethora of variables which affect it is
not at the level where we can attach any level of predictability to
it. Therefore, there does not appear to be any ‘‘universal’’
constants which can currently be applied to human decomposi-
tion globally. It may however be possible to create regionally
specific models of human decomposition given a better under-
standing of all the variables which impact the rate and trajectory
of decomposition within that geographic region. It appears
therefore that based on our current level of understand in relation
to all of the variables which affect the rate and trajectory of human
decomposition, any concept of universality may be overly
optimistic.
Acknowledgements
We would like to thank the Royal Canadian Mounted Police
(RCMP) and The Ottawa Police Service (OPS) for allowing us access
to their investigational files. To Melissa Martineau of the
Behavioural Sciences Department of the RCMP for assistance with
the ViCLAS data and to Sgt. Pam Scharf of the Forensic
Identification Section of OPS for all her assistance with the death
investigation files lastly, to Dave Thompson and Nadine Schuur-
man for their never ending support. No funding was received for
this research.
References
[1] A. Vass, The elusive universal post-mortem interval formula, Forensic Sci. Int. 204
(2011) 34–40.
[2] A. Vass, W. Bass, Time since death determinations of human cadavers using soil
solution, J. Forensic Sci. 37 (5) (1992) 1236–1253.
[3] B. Swift, I. Lauder, S. Black, J. Norris, An estimation of the post-mortem interval in
human skeletal remains: a radionuclide and trace element approach, Forensic Sci.
Int. 117 (2001) 73–87.
[4] S. Rhine, J. Dawson, Estimation of time since death in the Southwestern United
States, in: Advances in the Identification of Human Remains, Charles C Thomas,
Springfield, Illinois, 1998, pp. 145–161.
[5] R. Pickering, D. Bachman, The Use of Forensic Anthropology, Second ed., CRC
Press, Boca Raton, FL, 2009.
[6] V. Heaton, A. Lagden, C. Moffatt, T. Simmons, Predicting the postmortem submer-
sion interval for human remains recovered from U.K. Waterways, J. Forensic Sci.
55 (2) (2010) 302–307.
[7] B. Knight, Methods of dating skeletal remains, Med. Sci. Law 9 (1969) 247–252.
[8] B. Madea, C. Henssge, Eye chages after death, in: The Estimation of the Time
Since Death in the Early Postmortem Period, Edward Arnold, London, 1995, pp.
106–137.
[9] C. Henssge, B. Knight, T. Krompecher, B. Madea, L. Nokes, The Estimation of the
Time Since Death in the Early Postmortem Period, Edward Arnold, London, 1995.
[10] W. Bass, Outdoor decomposition rates in Tennesse, in: Forensic Taphonomy: The
Post Mortem Fate of Human Remains, CRC Press, New York, 1997, pp. 181–185.
[11] M. Megyesi, S. Nawrocki, N. Haskell, Using accumulated degree-days to estimate
the postmortem interval from decomposed human remains, J. Forensic Sci. 50 (3)
(2005) 1–9.
[12] H. Gill-King, Chemical and ultrastructural aspects of decomposition, in: Forensic
Taphonomy: The Postmortem Fate of Human Remains, CRC Press, New York,
1997, pp. 93–104.
Please cite this article in press as: D.L. Cockle, L.S. Bell, Human decomposition and the reliability of a ‘Universal’ model for post mortem
interval estimations, Forensic Sci. Int. (2015), http://dx.doi.org/10.1016/j.forsciint.2015.05.018
[13] A. Vass, G. Barshick, J. Sega, J. Caton, J. Skeen, C. Love, J. Synstelien, Decomposition
chemistry of human remains: a new methodology for determining the postmor-
tem interval, J. Forensic Sci. 47 (3) (2002) 542–553.
[14] M. Micozzi, Taphonomy of Human and Animal Remains: Systematic Study of
Post-Mortem Changes, Charles C. Thomas, Springfield, 1991.
[15] B.B. Horenstein, X. Linhres, B. Rosso De Ferradas, D. Garcia, Decomposition and
dipteran succession in pig carrion in Central Argentina, Ecol. Aspects Import.
Forensic Sci. 24 (March (1)) (2010) 16–25.
[16] T. Simmons, R. Adlam, C. Moffat, Debugging decomposition data-comparative
taphonomic studies and the influence of insects and carcass size on decomposit-
ing rate, J. Fornesic Sci. 55 (1) (2010) 8–13.
[17] P. Sledzik, Forensic taphonomy: postmortem decomposition and decay, in:
Forensic Osteology: Advances in the Identification of Human Remains, Charles
C Thomas Publisher Ltd, Springfield, Illinois, 1998, pp. 109–119.
[18] A. Galloway, W. Birkby, A. Jones, T. Henry, B. Parks, Decay rates of human remains
in an arid environment, J. Forensic Sci. 34 (3) (1989) 607–616.
[19] R. Mann, W. Bass, L. Meadows, Time since death and decomposition of the human
body: variables and observations in case and experimental field studies, J.
Forensic Sci. 35 (1) (1990) 103–111.
[20] M. Weitzel, A report of decomposition rates of a special burial type in Edmonton,
Alberta from an experimental field study, J. Forensic Sci. 50 (3) (2005) 1–7.
[21] C.P. Compobasso, G. Di Vella, F. Introna, Factors affecting decomposition and
Diptera colonization, Forensic Sci. Int. 120 (2001) 18–27.
[22] W. Rodriguez, W. Bass, Decomposition of buried bodies and methods that may aid
in their location, J. Fornsic Sci. 30 (3) (1985) 836–852.
[23] T. O’Brien, A. Kuehner, Waxing grave about adipocere: soft tissue change in an
aquatic context, J. Forensic Sci. 52 (2) (2007) 294–301.
[24] M. Manhein, Decomposition rates of deliberate burials: a case study of preserva-
tion, in: Forensic Taphonomy: The Postmortem Fate of Human Remains, CRC
Press, New York, 1997, pp. 469–478.
[25] D. Komar, Decay rates in a cold climate region: a review of cases involving
advanced decomposition from the medical examiners office in Edmonton,
Alberta, J. Fornesic Sci. 43 (1) (1998) 57–61.
[26] J. Prieto, C. Magana, D. Ubelaker, Interpretation of post mortem change in
cadavers in Spain, J. Forensic Sci. 49 (5) (2004) 1–6.
[27] A. Wilson, R. Janaway, A. Holland, H. Dodson, E. Baran, A. Pollard, D. Tobin,
Modelling the buried human body environment in upland climes unsing three
contrasting field sites, Forensic Sci. Int. 169 (2007) 6–18.
[28] B. Turner, P. Wiltshire, Experimental validation of forensic evidence; a study of
the decomposition of buried pigs in a heay clay soil, Forensic Sci. Int. 101 (2)
(1999) 113–122.
[29] M. Archer, Rainfall and temerpature effects on the decomposition rate of exposed
neonatal remains, Sci. Justice 44 (1) (2004) 35–41.
[30] N. Reeves, Taphonomic effects of vulture scavenging, J. Forensic Sci. 54 (3) (2009)
523–528.
[31] C. Pakosh, T. Rogers, Soft tissue decomposition of submerged, dismembered pig
limbs enclosed in plastic bags, J. Forensic Sci. 54 (6) (2009) 1223–1228.
[32] P. Cross, T. Simmons, The influence of penetrative trauma on the rate of decom-
position, J. Forensic Sci. 55 (2) (2010) 295–301.
[33] G. Anderson, Comparisson of decomposition rates and faunal colonization of
Carrion in indoor and outdoor environments, J. Fornesic Sci. 65 (1) (2011)
136–142.
[34] M. Sorg, E. Dave, A. Rebmann, Cadaver dogs, taphonomy, and postmortem interval
in the Northwest, in: Forensic Osteology, Charles C Thomas, Springfield, 1998, pp.
120–145.
[35] W. Haglund, Dogs and coyotes: postmortem involvement with human remains,
in: Forensic Taphonomy: The Postmortem Fate of Human Remains, CRC Press,
New York, 1997, pp. 367–381.
[36] J. Davey, Researches, Physiology and Anatomical, Stewart and Murray, Old Bailey,
London, 1893.
[37] B. Knight, Errors in estimating time since death, in: The Estimation of the Time
Since Death in the Early Postmortem Period, Edward Arnold, London, 1995, pp.
32–33.
[38] M. Pollanen, May 24, 2005 report of Dr. Michael Pollanen, in: The Gouge Inquiry:
Inquiry into Pediatric Forensic Pathology in Ontario, Hon. Steven Gouge, Com-
missioner, Toronto, Ont., 2008.
[39] M. Nafte, Flesh and Bone An Introduction to Forensic Anthropology, Carolina
Academic Press, Durham, North Carolina, 2000.
[40] W. Rodriguez, W. Bass, Insect activity and its relationship to decay rates of human
cadavers in East Tennessee, J. Forensic Sci. 28 (2) (1983) 423–432.
[41] D. Cockle, Human Decomposition and the Factors that Affect it: A Retrospective
Study of Death Scenes in Canada, Simon Fraser University, Burnaby, BC, 2013.
[42] M. Micozzi, Experimental study of postmortem change under field conditions, Eff.
Freezing Thawing Mech. Inj. 31 (1986) 953–961.
[43] M. Micozzi, Frozen environments and soft tissue preservation, in: Forensic
Taphonomy: The Postmortem Fate of Human Remains, CPC Press, New York,
1997, pp. 171–180.
[44] A. Bunch, The impact of cold climate on the decomposition process, J. Forensic
Identific. 59 (1) (2009) 26–44.
[45] M. Tersigni, Frozen human bone, Microsc. Investig. 52 (1) (2007) 16–20.
[46] K. Stokes, S. Forbes, M. Tibbett, Freezing skeletal muscle tissue does not affect its
decomposition in soil: evidence from temporal changs in tissue mass, microbial
activity and soil chemistry based on excised samples, PLoS ONE 183 (2009) 6–13.
[47] R. Fisher, Time of death and chages after death, Part 1 anatomical consiserations,
in: Medicolegal Investigation of Death: Guidelines for the Application of Pathol-
ogy to Crime Investigation, Charles C Thomas, Springfield, USA, 1973, pp. 12–38.
G Model
FSI-8010; No. of Pages 9

D.L. Cockle, L.S. Bell / Forensic Science International xxx (2015) xxx–xxx e9

[48] R. Shepherd, Simpson’s Forensic Medicine, Arnold, London, 2003.
[49] P. Allison, D. Briggs, Taphonomy of nonminerlzed tissues, in: Taphonomy: Re-
leasing the Data Locked in the Fossil Record, Penium Press, New York, New York,
1991, pp. 25–70.
[50] C. Knusel, R. Janaway, S. King, Death, decay and ritual reconstruction, Archaeol.
Evid. Cadaveric Spasm 15 (July (2)) (1996) 121–128.
[51] S. Forbes, B. Stuart, B. Dent, S. Fenwick-Mulcahy, ‘‘Characterization of adipocere
formation in animal species’’, J. Forensic Sci. 50 (3) (2005) 633–640.
[52] M. Tibbett, D. Carter, T. Haslam, R. Major, R. Haslam, A laboratory incubation
method for determining the rate of microbiological degradation of skeletal
muscle tissue in soil, J. Forensic Sci. 49 (3) (2004) 560–565.
[53] B. Dent, S. Forbes, B. Stuart, Review of human decomposition processes in soil,
Environ. Geol. 45 (4) (2003) 576–585.

Please cite this article in press as: D.L. Cockle, L.S. Bell, Human decomposition and the reliability of a ‘Universal’ model for post mortem
interval estimations, Forensic Sci. Int. (2015), http://dx.doi.org/10.1016/j.forsciint.2015.05.018