Forensic Science International 257 (2015) 242–251

Contents lists available at ScienceDirect

Forensic Science International

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

Design and evaluation of a bioreactor with application to forensic

burial environments
Melissa A. Dunphy a, Katherine E. Weisensee b, Elena A. Mikhailova c, Melinda K. Harman a,*
a
Department of Bioengineering, 301 Rhodes Engineering Research Center, College of Engineering and Science, Clemson University, Clemson, SC 29634, USA
b
Department of Sociology and Anthropology, 123A Brackett Hall, College of Business and Behavioral Sciences, Clemson University, Clemson, SC 29634, USA
c
Department of Forestry and Environmental Conservation, 206 Lehotsky Hall, College of Agriculture, Forestry and Life Sciences, Clemson University, Clemson,
SC 29634, USA

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

Article history: Existing forensic taphonomic methods lack specificity in estimating the postmortem interval (PMI) in
Received 23 April 2014 the period following active decomposition. New methods, such as the use of citrate concentration in
Received in revised form 8 June 2015 bone, are currently being considered; however, determining the applicability of these methods in
Accepted 22 August 2015
differing environmental contexts is challenging. This research aims to design a forensic bioreactor that
Available online 7 September 2015
can account for environmental factors known to impact decomposition, specifically temperature,
moisture, physical damage from animals, burial depth, soil pH, and organic matter content. These
Keywords:
forensically relevant environmental variables were characterized in a soil science context. The resulting
Artificial aging
Bioreactor
metrics were soil temperature regime, soil moisture regime, slope, texture, soil horizon, cation exchange
Bone capacity, soil pH, and organic matter content. Bioreactor chambers were constructed using sterilized
Decomposition thin-walled polystyrene boxes housed in calibrated temperature units. Gravesoil was represented using
Postmortem interval mineral soil (Ultisols), and organic soil proxy for Histosols, horticulture mix. Gravesoil depth was
Soil analysis determined using mineral soil horizons A and Bt2 to simulate surface scatter and shallow grave burial
respectively. A total of fourteen different environmental conditions were created and controlled
successfully over a 90-day experiment. These results demonstrate successful implementation and
control of forensic bioreactor simulating precise environments in a single research location, rather than
site-specific testing occurring in different geographic regions. Bone sections were grossly assessed for
weathering characteristics, which revealed notable differences related to exposure to different
temperature regimes and soil types. Over the short 90-day duration of this experiment, changes in
weathering characteristics were more evident across the different temperature regimes rather than the
soil types. Using this methodology, bioreactor systems can be created to replicate many different
clandestine burial contexts, which will allow for the more rapid understanding of environmental effects
on skeletal remains.
ß 2015 Elsevier Ireland Ltd. All rights reserved.

1. Introduction methods for PMI estimation from skeletal remains continue to

Accurate estimation of the postmortem interval (PMI) provides
context to forensic investigations. The PMI defines the time
elapsed between death and discovery of human remains [3].
Methods for estimating PMI of recovered remains with soft tissue
present, which can be observed for up to 1285 average degree days
(ADD) or the equivalent of a PMI of approximately six months or
less, have received more attention than methods applicable after
complete skeletonization has occurred [4]. However, precise
* Corresponding author. Tel.: +1 864 656 4140; fax: +1 864 656 4466.
E-mail address: harman2@clemson.edu (M.K. Harman).
elude researchers [1–4]. Change in citrate concentration is a novel
method being developed in forensic research that utilizes the
biochemistry of bone specimens to estimate PMI [1,2]. Currently
there are no methods that provide a robust methodology for
evaluating the impact of environmental variables on changes in
biochemical properties of bone in a burial context.
Skeletal remains are exposed to diverse microenvironments
that result in varied bone weathering patterns and decomposition
processes. Different areas on a single bone can exhibit different
stages of bone weathering and decomposition depending on
whether or not a given bone area has ground contact (soil
exposure) [5,6]. Skeletal remains recovered from a burial context
with full soil contact provide particular challenges for estimating

http://dx.doi.org/10.1016/j.forsciint.2015.08.014
0379-0738/ß 2015 Elsevier Ireland Ltd. All rights reserved.
 M.A. Dunphy et al. / Forensic Science International 257 (2015) 242–251
PMI; however the unique properties of the soil at a burial site also
provide the opportunity to examine environmental variables for
modeling the decomposition process in different geographic
locations. Benninger et al. [7] suggested that soil-based methods
are potentially useful for estimating PMI over extended time-
frames. Methods for systematically examining the effects of soil
and related environmental variables on decomposition are lacking
and are the focus of the current study.
In order to develop methods for estimating PMI that are not site
specific to a given geographic region and based on case study
observations, a careful experimental design approach is needed.
Variables known to impact the rate of decomposition must be
isolated and evaluated independently. Environmental modeling
databases provide information about the most important variables
related to decomposition, specifically temperature, moisture, and
soil properties for nearly every location in the United States and for
most worldwide locations. In this manner, the impact of a
particular set of variables and environmental characteristics can
be evaluated to better understand changes in the biochemical
properties of bone over time [2]. By gaining an understanding of
these factors, methods proposed for estimating PMI from buried
skeletal remains can be broadly applied across regions with
different environmental conditions.
Decomposition is categorized into five phases based on the
physicochemical and bacterial environments: autolysis, putrefac-
tion, liquefaction, dissolution, and skeletonization [8]. Soft tissues
break down due to microbial decomposition and natural cell death
occurs during the first four phases, while bone diagenesis occurs
due to mechanical breaking, decalcification, and dissolution in
acidic soils or water during dissolution and skeletonization phases.
Taphonomy uses the study of ecology to identify environmental
factors that can lead to body decomposition after death [9]. The
primary taphonomic factors include: water movement, tempera-
ture regime, organic matter content, fluoride and carbonate
concentration, duration of internment in soil, soil pH, cryoturba-
tion, microbial activity, release of fat, and mineralogy [10,11].
Using a mechanistic approach to understand the distinct variables
in this complex system, these taphonomic factors can be
systematically controlled in a forensic bioreactor to empirically
determine their impact on bone biochemistry.
Bioreactors are technical systems that enable reproducible and
controlled operating conditions and support biologically, chemi-
cally, or physically active processes. Bioreactors are generally
defined as devices in which biological and/or biochemical
processes develop under closely monitored and controlled
environmental and operating conditions (e.g. pH, temperature,
pressure, oxygen content, nutrient supply, waste removal) [12–
14]. In the current study, the term ‘‘forensic bioreactor’’ describes a
technical system used to monitor, control and manipulate
environmental factors that impact decomposition. Carter et al.
[14] successfully implemented a forensic temperature incubation
system over 28 days for the assessment of rat cadaver decomposi-
tion with respect to temperature (15–29 8C) and microbial activity.
In a shorter term study, Abdel-Maksoud [15] implemented an
archeological temperature incubation system over 13 h for the
assessment of bone specimens with respect to temperature (200–
300 8C). Both of these studies attempted to identify an environ-
mental mechanism for the changes seen in cadaver specimens
[14,15]. Carter et al. [14] found that temperature acted as an
environmental mechanism and increased microbial activity, which
is known to impact decomposition. Abdel-Maksoud [15] found
that temperature as an environmental mechanism impacted
coloration and crystallinity in such a way that artificially aged
bone samples could be matched to archeological bone samples.
Based on the results of these studies, the simulation of other
environmental mechanisms and evaluation of their impact on
243
decomposition would aid in understanding how cadaveric samples
progress in burial conditions over time. Through the design of a
forensic bioreactor system capable of controlling existing param-
eters known to impact decomposition, many different burial
scenarios could be considered side by side.
Design parameters for existing bioreactor systems are well
described, thus a system can be designed to incorporate discrete
variables relevant to forensic taphonomy. Common design
parameters include: size of the apparatus, temperature, sterility,
oxygenation, ability to apply sensors, use over time, transparency
of the bioreactor, easy placement of the sample, and easy
replacement of the medium [17]. Given the chemical properties
that soil imparts to the system, soil becomes the medium for a
forensic bioreactor [11]. Similarly, as microbial activity is a
hallmark of decomposition processes, the bioreactor can be used
for its intended purpose – the cultivation of microbial populations
[12]. In evaluation of the apparatus, chamber type, sample type
and sample size must be considered.
The overall purpose of the current study is to design and
develop a forensic bioreactor as a method for controlling key
environmental variables affecting bone decomposition, specifically
temperature and soil, in order to systematically evaluate their
impact on bone biochemistry over time. A synergistic collaboration
between relevant disciplines is pursued, including bioengineering
to define the biosystems control of the bioreactor, forensic
anthropology to define factors impacting bone decomposition,
and soil science to define the soil types and related environmental
variables (Fig. 1). The specific objectives are: (1) to identify the
environmental factors most important for modeling burial settings
in an outdoor context that are useful for forensic anthropology; (2)
to discretize environmental factors of an outdoor forensic burial
setting using soil science data; and (3) to design a forensic
bioreactor that can be easily monitored to mimic an outdoor burial
setting.
2. Materials and methods
2.1. Defining a baseline of environmental factors through forensic
taphonomy
Human remains in clandestine burials are interred in gravesoils.
These soils define the environments that forensic specimens are
exposed to during decomposition and lead to the acceleration or
Fig. 1. Identifying the synergy between collaborating fields: anthropology,
bioengineering, and soil science.
244 M.A. Dunphy et al. / Forensic Science International 257 (2015) 242–251
deceleration of natural decomposition processes [3,8,18]. The
factors contributing to body decomposition after death can be
divided into several categories, including physical (temperature,
water movement, animal factors), chemical (mineral content, pH,
dissolved gases), microbial (microbial activity, organic matter
content), and interment time (Table 1). Many of these factors are
related to the composition of the soil and can be categorized into
discrete types using soil science taxonomies. Using a subset of
these environmental descriptions, Vass [3] proposed a mathemat-
ical model for estimating PMI that focuses on temperature,
moisture, pH, and partial pressure of oxygen.
2.2. Identification of measurable environmental factors from
anthropology and soil science (Table 1)
Temperature is the most well studied physical factor associated
with the rate of decomposition and PMI research [3,14,15].
Specifically, temperature can lead to an increase in microbial
activity resulting in accelerated decomposition during the first four
stages [14]. It can also lead to evaporation of available moisture in a
given soil environment, thus decreasing microbial activity.
Considering both temperature and moisture together, arid
environments can lead to mummification, while tropical environ-
ments can lead to skeletonization [10,19]. In the case of
skeletonized remains, temperature can also lead to measurable
alterations in bone histological morphology [18]. Temperature
varies depending on region and elevation and can be measured
using instantaneous or continuous temperature probes [9,20].
Temperature variations over time are considered in accumulated
degree days, whereby the mean daily temperature is summed over
the entire duration of exposure. Soft tissue decomposition, and
thus the onset of skeletonization, is considered to occur after 1285
accumulated degree days [4].
Water movement and moisture content are also directly
associated with rate of decomposition [2,19,21]. Decomposition
accelerates in wetter environments, is slowed in lower moisture
environments when activity and motility of microbial populations
will be restricted; and is halted entirely in excess moisture
environments that stop microbial activity [19]. Another way that
moisture changes decomposition rates is through the formation of
pores in bone that allow exposure to microbial activity [12]. For
skeletonized remains, moisture can lead to cracking or splitting of
the bone [2]. Moisture varies dependent on regions and soil
moisture can be evaluated using soil moisture regimes [9].
Table 1
Relationship between relevant taphonomic factors and metrics within soil science
based on existing literature. Variables are categorized as physically, chemically, or
microbially impacting decomposition.
Taphonomic factors Soil science metrics References
Physical
Temperature Soil temperature regime [2–4,9,10,12,14,15,18–21]
Water Soil moisture regime
Slope
Texture
Animal factors N/A
Burial depth Soil horizon
Chemical
Mineral content Cation exchange capacity [3,8,11,16,18,22]
Ion content values
Soil pH Soil pH
Dissolved gasses Soil horizon
Microbial
Microbial activity Organic carbon content [8,10,14,18,19,21]
Organic matter Organic matter content
Alternately, it can be directly measured locally using a soil
moisture probe or dry and wet weight comparisons [9].
Disruption by animals can physically alter the exposure of
human remains through processes such as trampling, gnawing,
and digestion [10]. Increased exposure of hard and soft cadaver
tissues to environmental temperature, water movement, and
microbial activity accelerates decomposition [21]. These variables
are not directly relatable to soil science metrics; however, they can
be accounted for during sample preparation.
Considering chemical factors, the gravesoil present at a burial
site defines a cadaver’s chemical environment. The minerals
commonly found in soil that can influence decomposition include:
calcium, magnesium, sodium, potassium, silicon, aluminum, and
iron [11] These minerals, specifically the cation Ca2+, can compete
with hydrogen ions found in the soil for sites on hydroxyapatite
during the decomposition process and lead to increased or
decreased mineral phase dissolution of bone [16].
The pH of soil measures the concentration of H+ ion of a solution
on a logarithmic scale. Soil pH influences the solubility of
hydroxyapatite, the primary inorganic component of bone
[16,22]. As soil drops below a pH of 6, hydroxyapatite can be
reduced more quickly and release calcium and phosphorous ions. If
pH becomes alkaline, reduced rates of dissolution are seen [11]. In
addition, pH influences the activity of microorganisms, such as
fungi or bacteria [18]. The pH of soil can be measured using soil
analysis or a soil pH probe in the field.
Dissolved gasses impact the type of microbial decomposition
present. Oxygen levels are increased in soil when a burial site is
formed due to aeration of soil [8]. Depending on final grave depth,
clandestine burials can be aerobic with primary decomposition
occurring through proteolysis or anaerobic with bacterial fermen-
tation as the primary decomposition mechanism [3,8]. These levels
can be estimated based on presence of soils from different horizons
in the burial to estimate initial aeration of gravesoils, as well as
through laboratory testing.
Microorganisms are found in soil organic matter, which consists
of metabolic byproducts from decomposition as well as soil
humates [18]. Microbial decomposition augments the autolysis
phase of decomposition, which in turn increases the amount of soil
organic matter present in a gravesoil. Although bacteria, microflo-
ra, and fungi metabolism are known to impact the postmortem
microstructure, research suggested that some additional cofactor
must exist to account for the variation seen in the postmortem
intervals [21]. Temperature, moisture, aeration and soil type
directly impact the rate of decomposition through these processes
by altering amount of microorganisms and metabolism of these
organisms [8,10,14,18,19].
2.3. Discretization of bioreactor inputs based on environmental
factors
Soil temperatures can be discretized based on the soil
temperature regime for defined geographic regions. Soil
temperature regimes have been mapped to allow visual
assessment for a region (Fig. 2). These regimes consider the
mean annual temperatures, as well as the comparison between
mean summer and mean winter temperature, at a soil depth of
50 cm [20]. Soil temperature maps can be obtained for many
global locations.
There are twelve soil orders [20,23]. These are categorized
based on variables such as geographic region (temperature and
moisture parameters), mineral and ion content, texture, and
organic matter content [23]. These soil orders have been
mapped to allow visual assessment of their prevalence per
geographic region (Fig. 3). All twelve soil orders can be found in
the United States; however, some such as Ultisols are more
 M.A. Dunphy et al. / Forensic Science International 257 (2015) 242–251 245

Fig. 2. Temperature regimes of the United States depicted on the U.S. map. The four primary moisture regions: frigid, mesic, thermic and hyperthermic.
Image provided can be found at USDA/NRCS. http://www.nrcs.usda.gov/wps/portal/nrcs/detail/soils/use/maps/?cid=nrcs142p2_053998.

prevalent in the Southeast region of the U. S. than others. With
respect to forensic taphonomy, each of these soil orders can be
defined based on mineral content, cation exchange capacity
(CEC), texture, and organic matter content. The higher the cation
exchange capacity, the more cations the soil can hold. Soil
cations include calcium, magnesium, potassium, ammonium,
hydrogen, and sodium. Soil anions include chlorine, nitrate,
sulfate, and phosphate. These cations and anions make up the
mineral content of a soil, which is dictated by the parent
material. The soil texture provides information regarding the
type and thus the soil’s ability to retain moisture or the ability of
water to move through soil. In the instance of a horticulture
proxy soil for an organic soil, this value is not available since it
lacks minerals. Organic matter content affects the presence of
microorganisms in the soil and soils can be classified based on
their measured percentage of organic matter. Soil horizons with
less than 20% organic carbon content are commonly considered
mineral soils. Most of soil orders are mineral soils with only a
few exceptions [20].
Soils layers are defined as soil horizons and are characterized
based on additions and alterations of the parent material. There
are master horizons: organic materials (O), zone of humus

Fig. 3. Soil classification in the United States depicted on the U.S. map. Image provided by USDA/NRCS.
All twelve soil orders can be found in the U.S.; however, prevalence varies. http://www.nrcs.usda.gov/Internet/FSE_MEDIA/stelprdb1237749.pdf.
246 M.A. Dunphy et al. / Forensic Science International 257 (2015) 242–251

accumulation and eluviation (A), zone of maximum eluviation (E),

zones of maximum illuviation (B), parent material (C), rock (R)
[20]. Lower-cased letters are often used to indicate more specific
properties. For example the following horizons and subcategories
were identified at the Simpson Agricultural Station site in
Pendleton, SC shown in Fig. 4: A, Bt1, Bt2, Cr (Fig. 5). The horizon
with accumulation of silicate clay (t) is described as the Bt horizon
in Fig. 5.

2.4. Preparation of bioreactor

Bioreactor chambers were designed using 8.5 cm  12 cm 

6 cm polystyrene boxes. Each chamber was thoroughly cleaned,
dried and then sterilized using an ethylene oxide autoclave at
120 8C for 30 min. Based on the bioreactor input discretization
described above, the experimental design included the following
variable conditions: four temperature regimes (frigid, mesic,
thermic, hyperthermic); three soil types including two soil
horizons of mineral soil and an organic soil (mineral – Ultisol A
horizon, mineral – Ultisol Bt2 horizon, organic – horticulture mix);
and two bone types (whole rib, thin section of rib).
Mineral soil was obtained at the Simpson Agricultural Station in
Pendleton, SC from a Hiwassee sandy loam having a 10–15% slope
(Fine, kaolinitic, thermic Rhodic Kanhapludults). A kaolinitic soil
consists of layered silicate mineral clay (kaolinite). A thermic soil
temperature regime is in the range of 15–22 8C. Due to the relative
unavailability of organic soils in the southern United States, a
commercially available horticulture mix (Vigoro1 all-purpose
potting mix) was used as a proxy for an organic soil. Horticulture
mix soils generally incorporate high organic matter soils to
promote plant growth.
Temperature regimes for the bioreactor were simulated using
the following procedures. Frigid and mesic conditions were
achieved using separate laboratory refrigerators with variable
temperature controls (Isotemp, Thermo Fisher Scientific Inc.)
Thermic conditions were achieved by placing an insulated heating
system set to 35 8C inside of an environmental freezer set to 5 8C.
Finally, a hyperthermic environment was achieved using a
laboratory incubator (Isotherm, Thermo Fisher Scientific Inc.) set
to 37 8C for the duration of the experiment.
A total of fourteen different environmental conditions were
created to simulate environmental factor combinations and are
shown in Table 2. Eight of these chambers were used for thin
section of rib bone tested in both mineral and organic soils for all
four temperature conditions. For the remaining six chambers, the
frigid and hyperthermic temperature conditions were utilized to
Fig. 5. Soil depths (cm) and horizons for collection of Ultisol at Simpson Agricultural
Station (A, zone of humus accumulation and eluviation; Bt1 and Bt2, zones of
maximum illuviation of clay (kaolinite); Cr, saprolite, which is a weathered rock).
Soil Map Unit: HaD, Hiwassee sandy loam, 10–15% (Fine, kaolinitic, thermic Rhodic
Kanhapludults) is the whole taxonomic name.
simulate the effects of temperature extremes. Four of these
held whole rib bones buried in mineral soil and organic soil, and
two of these chambers simulated surface burial of thin sections
of rib bone buried in A horizon mineral soil. All bioreactors
were run for 90 days with bone samples collected at 30-day
intervals.
2.5. Selection of specimens for use in forensic bioreactor (Fig. 6)
The impact of soil properties on forensic bone was simulated
through burial of uniform bone samples in the bioreactors’ soil.
Fresh-frozen porcine spare rib bones were obtained from a local
grocer in Greenville, SC and soft tissues were promptly removed
and discarded. Porcine rib bone cross-sections (2–4 mm) were
cut using a water-cooled band saw, rinsed in filtered water to
remove residual blood, and air dried. Four 8 cm whole ribs were
placed into four bioreactor environments (one rib each) and 120
thin sections were placed into a total of ten bioreactor
environments (12 sections each). All bones were covered

Table 2
Experimental set up of the bioreactors’ conditions was based on temperature
regime (abbreviated as F, M, T, H for frigid, mesic, thermic and hyperthermic,
respectively), soil horizon (abbreviated as A for surface grave A horizon and B for
shallow grave depth in Bt2 horizon), specimen type (abbreviated as W for whole
and S for sections), and soil type (abbreviated as Min for mineral and Org for
organic). The naming scheme and contents for the main categories of forensic
bioreactor conditions are: S-Min-B) rib bone sections in mineral soil at shallow
grave depth; W-Min-B) whole rib bone in mineral soil at shallow grave depth; S-
Min-A) rib bone sections in mineral soil at surface grave depth; S-Org) rib bone
sections in organic soil; and W-Org) whole rib bone in organic soil.

Temperature Forensic bioreactor conditions

regime
S-Min-B W-Min-B S-Min-A S-Org W-Org

Frigid (0–7 8C) S-F-Min-B W-F-Min-B S-F-Min-A S-F-Org W-F-Org

Mesic (8–14 8C) S-M-Min-B S-M-Org
Thermic S-T-Min-B S-T-Org
(15–22 8C)
Hyperthermic S-H-Min-B W-H-Min-B S-H-Min-A S-H-Org W-H-Org
(>22 8C)
Fig. 4. Ultisol soil collection site at Simpson Agricultural Station, Pendleton, SC, USA.
 M.A. Dunphy et al. / Forensic Science International 257 (2015) 242–251
such that all surfaces were exposed to soil conditions. Bone
specimens were removed from each bioreactor at predeter-
mined time intervals of 30, 60 and 90 days after burial and
grossly assessed. Whole bones were removed and three thin
sections were cut off one end before replacing the whole bone
back into the soil of its respective bioreactor. For bones
previously cross-sectioned, three thin sections were simply
removed from each bioreactor at each time interval.
2.6. Monitoring of bioreactor conditions
Soil temperature was allowed to equilibrate for one week prior
to placement of bone samples. Once at equilibrium, air tempera-
ture was measured using a digital thermometer. Frigid and
hyperthermic condition simulators were monitored by continu-
ous digital readout recorded weekly. Mesic and thermic condi-
tions simulators were measured using a digital thermometer
(Accumet Basic A815, Fisher Scientific) also measured weekly. The
temperature conditions were monitored for the duration of the
experiment.
Soil conditions were evaluated prior to addition of soil to
bioreactor chambers at the Clemson Soil Analysis facility. Soil
analysis of mineral soils at horizons A and Bt2 and organic soil were
conducted in triplicate, including basic evaluation as well as
determination of organic matter content.
3. Results
3.1. Specification of forensic environmental factors with respect to soil
science metrics
Based on the systematic review of taphonomic, soil science
and forensic literature, a series of variables were determined
and aligned with appropriate soil science metrics (Table 1).
Physical metrics were most directly related to soil taxonomy
Fig. 6. Whole (A, B) and sectioned (C, D) porcine spare rib bones before exposure to the bioreactor environments and subsequently buried in organic (B) and mineral (D) soils
in the polystyrene chambers.
247
and mapped soil variable regimes. Soil taxonomy identifies soil
orders and horizons, which then allow for analysis of burial
depth. Mapped soil variable regimes, for example the soil
moisture regime data provided averaged global moisture data
based on location; whereas, soil slope and texture are obtained
through site analysis and subsequent soil analysis, thus
providing temporal site specific data. Similar maps are available
for temperature. Animal factors, specifically the alteration of
human remains, can be taken into account during sample
processing, and thus simple measurement can quantify this
condition. Based on this information, all four physical tapho-
nomic factors were able to be applied to bioreactor design and
were selected and controlled throughout a study.
Chemical metrics were most directly related to data obtained
through soil science analysis using field and laboratory work. Soil
horizon information related to depth of burial, which also
corresponds to aerobic or anaerobic decomposition, was docu-
mented on-site (Fig. 5). The remaining chemical data metrics were
obtained through laboratory soil science analysis and included soil
pH, mineral content (CEC and soil ion content values). Cation
Exchange Capacity is the total capacity of a soil to hold
exchangeable cations. Mineral content and soil pH can be easily
determined before, during, or after bioreactor design and use,
depending on the design. Soil analysis can be destructive or non-
destructive, again depending on the experimental design. In
contrast, presence of dissolved gasses can be easily anticipated in
natural conditions; however, disturbance of the soil will aerate it
upon placement in a bioreactor system and thus this variable is not
a suitable bioreactor control.
Microbial metrics were related to organic matter and organic
carbon values. Organic matter was directly measured through soil
laboratory analysis to obtain numerical values. Among the relevant
taphonomic factors, microbial activity is most affected by other
outlined factors, and thus direct measurement would be an
effective dependent variable for a study; however, it was not
248 M.A. Dunphy et al. / Forensic Science International 257 (2015) 242–251

considered a controllable environmental factor for the purposes of

this study. In contrast, soil organic matter can be measured
throughout the study through the use of soil laboratory analysis.

3.2. Discretization of soil science metrics

Based on the review of potential inputs relevant to forensic

anthropology (Table 1), taphonomic factors and soil science
metrics were discretized and successfully incorporated into the
bioreactor design, including soil temperature; soil type; and burial
depth.
Soil temperature was discretized using documented soil
temperature regimes present in the United States (Fig. 2). The
main temperature regimes found in the United States are frigid Fig. 7. Final bioreactor system consisted of a temperature unit, bioreactor chambers,
(0–7 8C), mesic (8–14 8C), thermic (15–22 8C), and hyperthermic and environmental components (soil type, forensic specimen).
(>22 8C). These four temperature ranges were used to discretize
temperature conditions for bioreactor design through the selection
of a target temperature in each range.
Soil type was discretized through the use of soil taxonomy than the 74.5% organic matter content of the organic soil.
based on organic matter content. Soils above 20% organic carbon Similarly, the calcium ion content of the Ultisol horizon A and B
content are considered organic, and of the twelve soil orders, were 108.0 and 38.0 mg/kg respectively, and both were
Histosols are the primary organic soil order. Thus, soil organic significantly lower (t-test, p < 0.001) than the 1373.0 mg/kg
matter was discretized into two primary categories, organic and calcium ion content of the organic soil. This increase is likely due
mineral soils. Using the map provided in Fig. 3, soil type can also be to soil enrichment for horticulture purposes; however, in
discretized based on region and soil order to allow control of soil this forensic bioreactor application such increases could be
properties. used to gain insight into the effect of calcium content on
Burial depth was discretized based on soil horizon. Soil horizons decomposition.
are specific to locations and define the state of altered parent
material as it moves from rock to soil form. An example of soil 3.4. Gross assessment of bone sections using weathering
horizons measured at the Simpson Agricultural Facility is shown in characteristics
Figs. 4 and 5. In this Ultisol, the soil has been discretized into four
specific horizons: A, Bt1, Bt2, and Cr. For the purposes of forensic Results of gross assessment of bone sections using weathering
research, a surface scatter was defined as located in the A horizon, characteristics revealed notable differences related to exposure to
thus specifying the exposed soil conditions. Similarly, a shallow different temperature regimes over the 90-day experiment. Bone
grave burial at 50 cm was defined as a part of the Bt2 horizon. Thus, specimens from the frigid temperature regime retained less
inclusion of these relevant soil horizons simulated these two weathered characteristics including a natural ivory color appear-
different burial conditions within individual bioreactor chambers ance, more dense and moist bone, and the presence of bone
(Table 2). marrow tissue (Figs. 9–11). In the hyperthermic temperature
regime, the bone specimens exhibited more weathered character-
3.3. Evaluation of forensic bioreactor inputs istics including reduced density (e.g. light weight), trabecular bone
with less bone marrow tissues present, chalky appearance for the
The final bioreactor design incorporated a temperature unit
based on the four discretized temperature regimes (frigid, mesic,
thermic and hyperthermic), with internal components consisting
of two different discretized soils (mineral and organic) including
two soil horizons to simulate surface scatter and shallow grave,
and both sectioned and whole bone specimens simulating skeletal
remains in gravesoil (Fig. 7).
Results from the temperature monitoring verified that the
simulated temperature conditions in each bioreactor experienced
little variation over the experiment duration (Fig. 8). The
bioreactors suitably simulated the predefined temperature
regimes of frigid (3.8  0.8 8C), mesic (8.8  0.9 8C), thermic
(16.0  1.0 8C), and hyperthermic (37.5  0.1 8C) (Table 3).
Results from analysis of soil confirmed that discrete
characteristics of mineral and organic soils at different burial
depths were achieved (Table 4). Average soil pH was 4–6 for all
soils analyzed, with lowest pH in the Ultisol horizon A and
highest pH in the horticulture mix. Other measured parameters
showed discrete differences between the soils. The CEC of Ultisol
at horizon Bt2 was lowest at 4.9 meq/100 g. CEC for Ultisol at
horizon A was 8.1, and the CEC of the horticulture mix was 16.4.
There were stark differences between mineral and organic soils
for calcium ion content and organic matter content. The organic Fig. 8. Recorded temperatures from the simulated temperature conditions in each
matter contents of the two Ultisol horizons were 14.1% and bioreactor showed minimal temperature variations within each regime throughout
12.1%, and both were significantly lower (t-test, p < 0.001) the experiment duration.
 M.A. Dunphy et al. / Forensic Science International 257 (2015) 242–251 249

Table 3
Temperature regimes of the United States compared to bioreactor environment
temperatures created in the laboratory and monitored throughout the 90-day
period of this study.

Soil temperature Regime Simulated Standard

regime range [8C] average [8C] deviation [8C]

Frigid 0–7 3.8 0.8

Mesic 8–14 8.8 0.9
Thermic 15–22 16.0 1.0
Hyperthermic >22 37.5 0.1

cortical bone, reduced bone grease present, and a general brittle

quality throughout. The greater degree of weathering was
expected based on the increase in temperature regime. Differences
in the degree of weathering across soil types were less distinct. The
difference between the frigid and hypothermic temperature
regimes were similar in both the mineral and organic soil types,
although soil staining was more prominent for bone specimens in
the mineral soils at colder temperatures. Over the short duration of Fig. 9. Bone sections after 90 days exposure to mineral soil from the Bt2 horizon (S-
this experiment, the differences in soil type do not appear to Min-B) in the frigid, mesic, thermic and hyperthermic temperature regimes. Visual
differences included a progressively lighter color, loss of bone grease and a general
impact the rate of weathering as strongly as differences in
increasingly brittle quality from the frigid to hyperthermic temperature regimes.
temperature. These results confirm previous studies [5,6,14,19] Scale bar for each image is incremented in 1 mm tick marks.
related to the variables that impact the rate of bone weathering
and demonstrate the usefulness of this experimental design to
control one variable in order to document the effect of another on
the rate of weathering.

Fig. 10. Bone sections exposed to organic soil (S-Org) for 60 days and 90 days in the frigid and hyperthermic temperature regimes. Visual differences included a darkened
color, loss of bone grease, and reduced density (e.g. lightweight) in the hyperthermic conditions. All images were acquired under similar lighting conditions. Scale bar is
10 mm in length.
250 M.A. Dunphy et al. / Forensic Science International 257 (2015) 242–251

Table 4 4. Discussion
Analysis of soils used in the bioreactor design prior to introduction of bone samples.

Soil analysis – averages Standard The study describes an interdisciplinary approach to forensic
deviation anthropology and develops a forensic bioreactor system based on
Ultisol – mineral soil – horizon A defined factors known to impact bone decomposition (Fig. 1 and
Texture Coarse sandy loam N/A Table 1). Specifically, environmental factors important to forensic
(65% sand + 20% anthropology and useful for modeling decomposition in an
silt + 15% clay)
outdoor context were defined and then discretized using soil
Soil pH (pH) [–] 4.0 0.1
Cation exchange capacity 8.1 0.6 science data. Finally, these factors were successfully incorporated
(CEC) [meq/100 g] into an easily monitored forensic bioreactor prototype that
Organic matter (OM) [%] 14.1 0.7 mimicked an outdoor setting. In this manner, the bone samples
Calcium ion content [mg/kg] 108.0 27.1
were treated as a tissue undergoing dynamic changes when
Ultisol – mineral soil – horizon Bt2 exposed to a soil medium. This research provides a foundation for
Texture Clay (24% sand + 18% N/A bone decomposition models and advances forensic anthropology
silt + 58% clay)
related to estimating PMI for skeletal remains.
Soil pH (pH) [–] 5.4 0.1
Cation exchange capacity (CEC) 4.9 0.3 The environmental variables most commonly related to
[meq/100 g] decomposition were identified and related to the corresponding
Organic matter (OM) [%] 12.2 1.2 soil science metrics (Table 1). Of these variables, moisture, burial
Calcium ion content [mg/kg] 38.0 5.5 depth, mineral content, soil pH, and organic matter content were
Horticulture mix – organic soil identified as pre-experiment parameters. These variables were
Texture N/A N/A identified through existing soil science literature, many of which
Soil pH (pH) [–] 5.8 0.1 are easily assessable through the use of online databases, allowing
Cation exchange capacity (CEC) 16.4 0.8
[meq/100 g]
for the appropriate selection of soil to fit the researchers need. In
Organic matter (OM) [%] 74.5 3.0 addition, these variables can be quantified before, during or after
Calcium ion content [mg/kg] 1373.0 87.0 the experiment depending on the objectives of the research. Using
existing databases, temperature and soil order variables were
discretized to allow testing of specific environmental conditions

Fig. 11. Whole bones were sectioned after exposure to mineral soil (W-Min-B) for 60 days and 90 days in the frigid and hyperthermic temperature regimes. Visual
changes included a chalky whiter color for cortical bone, reduced density (e.g. light weight) and loss of bone grease evident in the cancellous bone in the hyperthermic
conditions. Note, soil residues have been removed to better visualize the bone specimens. All images were acquired under similar lighting conditions. Scale bar is
10 mm in length.
 M.A. Dunphy et al. / Forensic Science International 257 (2015) 242–251
across a global spectrum of potential environments (Figs. 2 and 3).
In this manner, a bioreactor system capable of simulating a broad
range of temperature and soil related environmental conditions
was developed and evaluated.
Other variables identified here that were not simulated in this
study can be discretized using similar means. Soil moisture can be
discretized based on the soil moisture regimes also provided by the
USDA, and measured using soil analysis facilities or directly
monitored using sensors within the bioreactor chambers. Mineral
content and soil pH can also be selected based on soil order using
available soil classification maps (Fig. 3) or through analysis of
local soils. Each soil order has specific attributes such as
predominant minerals that are commonly denoted as sub-
classifications.
The bioreactor systems designed in this experiment were able
to simulate four different temperature conditions within 1 8C of
the target temperature (Fig. 8 and Table 3), two extremes of soil
organic matter content (Table 4), and two types of forensic burial
scenarios (Fig. 7). In addition, test samples were selected that
accounted for two levels of exposure that could be anticipated in a
clandestine burial of human remains, namely whole bone or bone
sections (Fig. 6). Bone sections exposed to the conditions simulated
in the bioreactors exhibited gross weathering characteristics,
which revealed notable differences related to exposure to different
temperature regimes and soil types. Bioreactor chambers were
easily sterilized prior to research, as well as constructed of
disposable materials to facilitate post-experiment disposal.
Through the use of clear, latching lids, each chamber was also
easily accessible for collection of samples over the course of the
experiment. This study was limited in the number of variables that
could be incorporated into the system. Modifications to this system
would allow for inclusion of additional variables with additional
sensors as needed for monitoring, as well as make the system
easier to build in non-university laboratories.
Future bioreactor systems could incorporate a complete seal to
allow the creation of a constant atmospheric environment. This
type of design would allow for the control of exposure to gasses
such as oxygen and carbon dioxide, which can directly determine if
decomposition will be driven by anaerobic or aerobic organisms. In
addition, a sealed system would allow for the implementation of a
desired moisture scheme. A moisture scheme could involve
addition of water at a constant rate over the experiment to test
water movement, or the continuous measurement of moisture
levels with addition of water if levels drop below a desired level. It
would also be important to utilize glass rather than polystyrene
chambers. Glass can be sterilized using simpler methods than
plastics, which would make this type of system easier to use.
Ultimately, this research has shown the potential for collabora-
tions between forensic research, soil science, and bioengineering to
develop a system that allows the simulation of many environments
in one research location. Through the alignment of existing
variables with those measurable in other fields, forensic research
can focus on quantitative evaluation of research questions. This
system is particularly valuable for the testing of new research
251
methods in forensic taphonomy, especially related to the analysis
of skeletal remains and the impact of global environmental
conditions.
Acknowledgements
This work was supported by institutional funding provided by
Clemson University as Bioengineering faculty start-up funds and
Calhoun Honors College Undergraduate Departmental Honors
research funds.
References
[1] H.P. Schwarz, K. Agur, L.M. Janz, A new method for determination of postmortem
interval: citrate content of bone, J. Forensic Sci. 55 (2010) 1516–1521.
[2] F. Kanz, C. Reiter, D.U. Risser, Citrate content of bone for time since death
estimation: results from burials with physical characteristics and known PMI,
J. Forensic Sci. (2013) 1–8.
[3] A.A. Vass, The elusive universal post-mortem interval formula, Forensic Sci. Int.
204 (2011) 34–40.
[4] A.A. Vass, W.M. Bass, J.D. Wolt, J.E. Foss, J.T. Ammons, Time since death deter-
minations of human cadavers using soil solution, J. Forensic Sci. 37 (5) (1992)
1236–1253.
[5] A.K. Behrensmeyer, Taphonomic and ecologic information from bone weathering,
Paleobiology 4 (2) (1978) 150–162.
[6] M.A. Janjua, T.L. Rogers, Bone weathering patterns of metatarsal v. femur and the
post-mortem interval in Southern Ontario, Forensic Sci. Int. 178 (2008) 16–23.
[7] L.A. Benninger, D.O. Carter, S.L. Forbes, The biochemical alteration of soil beneath a
decomposing carcass, Forensic Sci. Int. 180 (2008) 70–75.
[8] B.B. Dent, S.L. Forbes, B.H. Stuart, Review of human decomposition processes in
soil, Environ. Geol. 45 (2004) 576–585.
[9] M. Tibbett, D. Carter, Soil Analysis in Forensic Taphonomy: Chemical and Biologi-
cal Effects of Buried Human Remains. International Library of Archaeology, CRC
Press, 2008.
[10] D.H. Ubelaker, Taphonomic Applications in Forensic Anthropology, Forensic
Taphonomy: The Postmortem Fate of Human Remains, CRC Press Inc., 1997,
pp. 77–90.
[11] F.D. Pate, J.T. Hutton, K. Norrish, Ionic exchange between soil solution and bone:
toward a predictive model, Appl. Geochem. 4 (1989) 303–316.
[12] R. Pörtner, S. Nagel-Heyer, C. Goepfert, P. Adamietz, N.M. Meenen, Bioreactor
design for tissue engineering, J. Biosci. Bioeng. 100 (2005) 235–245.
[13] I. Martin, D. Wendt, M. Heberer, The role of bioreactors in tissue engineering,
Trends Biotechnol. 22 (2004) 80–86.
[14] D. Carter, D. Yellowlees, M. Tibbett, Temperature affects microbial decomposition
of cadavers (Rattus rattus) in contrasting soils, Appl. Soil Ecol. 40 (2008) 129–135.
[15] G. Abdel-Maksoud, Comparison between the properties of ‘‘accelerated-aged’’
bones and archaeological bones, Mediterr. Archaeol. Archaeom. 10 (2010)
89–112.
[16] E.M. White, L.A. Hannus, Chemical weathering of bone in archeological soils, Am.
Antiq. 48 (1983) 316–322.
[17] A. Kodde, Development of a Bioreactor for Bone Tissue Engineering, Eindhoven
University of Technology, 2003, pp. 1–65.
[18] A.M. Child, Towards an understanding of the microbial decomposition of archae-
ological bone in the burial environment, J. Archaeol. Sci. 22 (1995) 34–40.
[19] D.O. Carter, D. Yellowlees, M. Tibbett, Moisture can be the dominant environ-
mental parameter governing cadaver decomposition in soil, Forensic Sci. Int. 200
(2010) 165–174.
[20] US Department of Agriculture, NRCS, Keys to Soil Taxonomy, Washington,
DC,11th ed., 2010.
[21] L.S. Bell, M.F. Skinner, S.J. Jones, The speed of post-mortem change to the human
skeleton and its taphonomic significance, Forensic Sci. Int. 82 (1996) 130–140.
[22] S. Weiner, H.D. Wagner, The material bone: structure-mechanical function rela-
tions, Annu. Rev. Mater. Sci. 28 (1998) 271–298.
[23] F.E. Damann, A. Tanittisong, D.O. Carter, Potential carcass enrichment of the
University of Tennessee Anthropology Research Facility: a baseline survey of
edaphic features, Forensic Sci. Int. 222 (2012) 4–10.