Forensic sciences: Effects of stress and perturbations on soil communities/materials and methods

Experimental design
The experimental design was composed of one control (soil only) and three different treatments: The treatments and control were realised in three replicates. The treatments were carried out on the surface of flowerpots filled with soil and litter coming from a mixed oak grove situated in the city of Neuchâtel (Switzerland 47°00'00.63 N; 06°56'00.44E, elevation 586m).
 * Pig head
 * Cow manure
 * Pig blood

Pig Head treatment
The first treatment was a dead pig head (Sus scrofa domesticus) laid at the surface of the soil of each pot. The pig has been chosen because of its anatomical, genetical and physiological similarities with human body. Indeed, a study showed that even if the pork is not a perfect model, its decomposition is relatively comparable to a human in many ways. The head part has been chosen because of its fast decomposition process, due to its small size and facial orifices that are easily reachable for insects.

The heads were stored in a metal mesh cage in order to help sampling underneath the carrion by an easy displacement of the head without disturbing the soil and litter. The mesh of the grid was large enough to allow the invertebrate activity working on the decomposition process.



Manure and blood treatments
The second and third treatments were cow manure laid at the surface of the soil of each pot and pig blood spilled on the pots. The aim of these treatments was to evaluate the influence of blood and manure on amoebae communities (which is still unknown) and compare it with the results of pig heads. Indeed, this study attempts to determine if effects of a cadaver on testate amoebae communities are due to the corpse in its entirety or to its body fluids only, and if the body fluids have a more or less pronounced effects than the cadaver.

Study site
All the flowerpots were stored in a greenhouse of the botanical garden of Neuchâtel, undergoing the local weather conditions except rain. The city of Neuchâtel is exposed to a degraded oceanic climate. It’s a climate with seasons, cold winters and hot summers. The experiment lasted one month and took place in May 2015. This month was alternately rainy and sunny and the daily temperatures were situated between 10 and 28 °C. After specially hot sampling days, the flowerpots were watered in order to keep some moisture content. This was done twice.

Sampling analysis
Samples of soil and litter were sampled once a week during one month (T0 = day 0, T1 = day 11, T2 = day 18, T3 = day 25) and then analysed in order to identify and count the living and dead testate amoebae species that they contained. The respiration rate of the soil organisms and the pH of the soil were also measured once a week. The samples were collected right beneath the pig head and manure and at the surface of the pot, from a depth of 0–5 cm. A small amount of soil was collected in each pot in order to limit the disturbance of the pot.

The samples were then placed into plastic bags.

Respirometry
The CO2 flow (µmol m−2 s−1) released by the soil organisms was measured by the Respirometer Li-cor® Biosciences LI-8100A. Before each measurement, the pigs heads and the manure were displaced.



pH measurement
The pH of each samples was measured with the pH meter Extech® instrument Waterproof ExStik®II pH/Conductivity Meter. The measurement was made as follow:
 * 1) After a calibration of the pH meter at 4, 7 and 10 pH, samples and deionized water were mixed in a 1:1 ratio in a sampling cup. Mixingsoilpig.jpg
 * 2) The suspension settled for 3 minutes prior to measurement.
 * 3) The pH meter was put at the surface of the mixture and let there until a stabilization of the pH displayed on the screen.

Testate amoebae extraction and identification
Each sample was treated following the same protocol:
 * 1) 2g of soil were put in a plastic flask with tap water and shaken for approximately 3 minutes.Samplingsoilpig3.jpg
 * 2) The mixture was sieved trough a 200 μm mesh size in order to remove bigger particles.Sievingsoilpig.jpg
 * 3) The filtrate was sieved again trough a 20 μm mesh. The new filtrate was discarded and the 20–200 μm fraction sitting in the filter was rinsed in a falcon tube.Sievingsoilpig2.jpg
 * 4) Tap water was added up to 50ml.
 * 5) The falcon tube was centrifugated at 15 °C and 2500rpm (rounds per minute) for 10 minutes.Centrifugationsoilpig.jpg
 * 6) The supernatant was discarded and approximately 5ml were left in the tube.
 * 7) 50 μl of Rose Bengale was added and the mixture was let sit for 30 minutes. Colorationsoilpig.jpg
 * 8) Water was added up to 45ml and the tube was centrifugated a second time at 15 °C and 2500rpm for 10 minutes.
 * 9) The supernatant was discarded and approximately 5ml were left in the tube.Extractionsoilpig.jpg
 * 10) To finish 1-2 drops of the extraction were put on a slide with 1 drop of glycerol.

The 20 first amoebae found in each replicate were observed, counted and identified with optical microscopes at 200× or 400× magnification. Identification of testate amoebae was achieved using test morphology, composition, size and colour.

The samples were stored in a cold room between the microscopic identifications.

Statistical analysis
Statistical analysis and graphics were made with R.

ANOVAs and Tukey tests were used to test if the influence of treatments and time (and their interaction) on pH, released CO2 flow, living dead/ratio of testate amoebae and Shannon index (species diversity index) of each sample are significant. The Shannon index allows to appreciate the diversity as a number. The Shannon index of each sample was calculated with the R package Vegan.

Pearson correlation tests were used to check if there is a significant correlation between pH and the living/dead ratio of testate amoebae, and between pH and the Shannon index of the samples.

The normality and the homogeneity of variances were tested with Shapiro's and Bartlett's tests because the ANOVA, the Tukey test and the Pearson's correlation test require this two conditions to be fulfilled, but they were not fulfilled in most cases. Nevertheless, ANOVAs, Tukey tests and a Pearson's correlation tests were done because non-parametric tests were too complicated for our competence level in statistical analysis.