Porcine (Sus scrofa domesticus) bone was utilized for this experiment. Fresh porcine samples were collected from a local abattoir (Staffordshire Meat Packers, UK). These samples were fresh on the day of collection and immediately taken to the laboratory at Cranfield University, Defence Academy of the United Kingdom, Shrivenham, UK. The pigs were all aged 4 to 6 months old and were bred for produce at the abattoir farm. There were twelve femora, twelve scapulae, and twelve individual ribs that were sampled for both summer and winter seasons. A scalpel was used to deflesh the bone of all soft tissue. Once defleshed, the bones were rinsed with de-ionized water. Five of each bone type were cut into approximately 3 × 0.5 × 0.5-cm fragments to allow for chemical analysis (analytical samples) whilst six of each type were kept complete for observational analysis (observation samples). Control samples (fresh frozen bone) from each bone element were obtained for comparison. These control samples were used for chemical analysis. One of each bone type of both observational and analytical samples was placed under shade. All the samples and fragments were weighed before and after the experiment. At the end of the observational study, the samples were stored in an – 18 °C freezer and thoroughly thawed prior to the FTIR analysis.
Long bones (femur) and flat bones (scapular and ribs) were sampled to analyze differences between types of bones if bleaching was to occur.
Weather data was collected daily between the summer months of 21 June 2017 and 22 September 2017; and the winter months between 21 December 2017 and 22 March 2018. UV levels, temperature, precipitation, humidity, and wind velocity were collected at three points during the day (6 am, 12 pm, and 6 pm) for the entire length of both seasons using the online weather database (Accuweather). The weather data that was collected was used to explore a correlation coefficient by measuring the UV and calculating accumulated degree days (ADD) and examining if there is a pattern of correlation against bleaching levels.
The samples were weighed (g), and the cortical thickness of the samples was measured before and after exposure to the outdoor environment to observe mass reduction. Photographs were taken at each interval to document the observational findings and record changes in the positioning, colouring, and weathering of the bones.
Chemical analysis was carried out using a Bruker Alpha Platinum ATR FTIR. A scan resolution of 4 cm−1 and 16 scans were employed for data collection, within a range of 2500–400 cm−1. Physicochemical modifications to the mineral and organic components of the bone were investigated for the control and experimental samples. Spectragryph 1.2 was used to calculate the area of the v3 phosphate (1200–900 cm−1), amide (1750–1600 cm−1), and v3 carbonate (890–850 cm−1) bands. Carbonate to phosphate and amide I to phosphate ratios were calculated to assess changes to the carbonate within the bone mineral and collagen, respectively, and the splitting factor v4 was calculated to assess changes to the crystallinity index. These physicochemical variables were used to compare control samples against experimental samples at different intervals of sun exposure to confirm whether bone changes had taken place due to bone bleaching.
A CM700d spectrophotometer was used to analyze the colour space values (L*A*B) to determine whether a pattern of colour change can be quantified. These were taken with an aim to obtain a fixed scientific value for the colour changes that were being observed.
To monitor visual colour changes, an observational scoring system was developed. For each section of bone, a score of 1–5 (number value) was given at each 3-week interval, 5 being complete white and 1 being a natural bone colour. A fully bleached bone equated to an ‘off-white to white’ colouration with no areas of brown or yellow staining across the entirety of the bone. This allowed for quantitative measure.
Thus, the bone colour bleaching score was assessed on the following:
Natural bone colour—beige with a tint of brown
Cream—cream beige in colour with tints of yellowing
Off-white—cream with no tints of yellow or brown
Brilliant white—bright white
Desert white—–bright white with drying/flaking
Sections of bone were observed on their coloured appearance and given an observational score between 1 and 5 as above. The scores were then added to give a total for each bone. The sections of bone that were scored were as follows:
Femur—medial and lateral condyle, patellar surface, medial and lateral epicondyle, medial and lateral distal shaft, anterior and posterior distal shaft, medial and lateral proximal shaft, anterior and posterior proximal shaft, anterior and posterior mid-shaft, medial and lateral mid-shaft, and the head and neck
Scapular—supraspinatus fossa, infraspinatus fossa, posterior surface area, glenoid cavity, spine, and neck
Rib—external shaft, internal shaft, and distal and proximal rib ends
Analytical sample preparation
A total of 30 porcine femur samples were randomly selected for chemical analysis. Five bleached samples from the summer, the simulated desert, and from the winter experiment were taken as well as five control samples. The sections were washed with de-ionized water and cut into 1-mm slices using a diamond cutting blade. The slices were then milled into a fine powder using a Retsch mixer mill and sieved using 106-μm mesh sieve. Bone colour was assessed using CIE (International Commission on Illumination) L*A*B colour scale with a CM700d spectrophotometer.
Outdoor taphonomy facility
The samples were placed in a supine position at a taphonomy facility at Cranfield University Shrivenham. The facility was approximately 90 m2 with a 1.5-m highly meshed fence. For the open sun-exposed area, the area was approximately 6 m2. Samples that were placed over the summer period were put out on the first day of summer (21 June 2017) and were collected on the last day of the season (22 September 2017). This method was also employed for the winter samples. A chicken mesh fencing was placed around this area to reduce scavenging. The soil was a light sedimentary sandy type. The soil pH was taken before and after the experiment. A wooden board was placed over an area of grass and covered with soil from the surrounding area which allowed the bones to be at the same level of UV exposure and reduced the green algae staining that is often found on bleached bone in these types of environments [1, 6]. The shaded area was covered with a canopy of approximately 6 m2. Additionally, a plastic crate was used to cover the samples to minimize exposure to sunlight, which ensured temperature regulation and ventilation. Photographs and sketches were taken at each collection to assess the level of bleaching.
Simulated desert environment
Additionally, a high-UV environment that replicated a desert was created and placed within the laboratory to see if higher UV levels have greater impact on bleaching than the expected levels observed within the UK. An Arcadia 10% UVA UVB reptile light bulb was used for UV radiation, with a UV index of approximately 9–10. For further information on UV indexes, refer to . For observational analysis, seven femora and seven rib samples were utilized for this experiment, with no scapulae. For chemical analysis, 24 femora and 24 rib fragment samples were placed inside the tank. The samples were placed on Prorep Leo Life reptile sand, and a 100-w heat lamp was placed into the tank to replicate the infrared temperatures. Samples were placed inside a reptile tank for a period of 8 weeks, with the fragments examined every week.
Control samples were placed in a supine position into a box, on a layer of sand and placed in a dark cupboard where there was limited access to UV exposure.
The outdoor facility was situated in an open-field area to collect accurate bleaching sun exposure. The limitations of exposing the bones to the elements are that it was not possible to control other weathering factors that could have interrupted the experiment. This became noticeable during winter when storms were present, disrupting the position of the samples. Precipitation would also have affected the positioning of the bones on the platform. Scavenging was another limitation as although the samples had been fenced, some samples were taken during the winter months. This affected the overall reliability of the results and made it difficult to assess the percentage of bleaching that occurred out of all samples.
The limitations for the simulated desert environment were the reliability of the equipment. The UV lighting could have been better equipped. The UV bulb that was used was designed for reptile tanks and so did not reach an expected peak UV reading after the first interval that would be expected in the desert. As the samples were situated in a tank, they did not get the same outdoor exposure to other weathering factors that would occur in the desert, such as colder night temperatures, rain, or sandstorms.
Porcine (Sus scrofa domesticus) bone was utilized for this experiment for ethical purposes. A limitation of this is the microstructural bone differences between human and pig . However, porcine has provided a firm foundation for research in taphonomy for many years and a recent comparative study  concluded porcine to be an adequate substitute when human bone is difficult to obtain.