Development of a Science Base and Open Source Software for Bloodstain Pattern Analysis
Daniel Attinger. Ph.D.
Executive Summary
Bloodstains are deterministic signs of a violent crime. Bloodstains help determine how the crime happened, pointing e.g. to the region of origin of a blood spatter. However, Bloodstain Pattern Analysis is not straightforward because the physical relation between blood impact and the resulting bloodstains is non-linear. Indeed, the formation of bloodstains involves a complex fluid (the blood), a subtle interplay of fluid mechanics, heat and mass transfer, in the presence of a deforming free surface, and impact surfaces with diverse values of roughness and wettability. The US National Academies recently advocated for stronger scientific foundations given “the complex nature of fluid dynamics”.
The proposed research team has addressed the specific challenge in Bloodstain Pattern Analysis (BPA) to determine where a blood spatter originates from for the purpose of reconstructing the bloodletting event. As stated in 1939 in Balthazard et al., ‘‘Le problème [of reconstructing trajectories] est très difficile à résoudre’’. Indeed, reconstructing trajectories is still very difficult today. The methods of strings and of the tangent typically used in today’s crime scenes are first-order approach to reconstruct trajectories, by assuming that the droplets travel in straight lines. By neglecting drag and gravity forces, these approaches induce systematic and poorly quantified uncertainties. Very recently, BPA research has proposed techniques to reconstruct curved trajectories based on probabilistic or statistical methods, but none of these methods can satisfactorily predict curved trajectories caused by drag and gravity. In this research project, the hypothesis that 3D inspection of individual stains provides sufficient information to reconstruct curved trajectories accounting for gravity and drag has been tested.
A method for trajectory reconstruction has been developed, based on the reconstruction of curved trajectories by estimation of impact conditions from three-dimensional measurements of stains. Based on a set of laboratory experiments analog to the situation of a beating, the method is found to be about 4 times more accurate than the method of strings to determine the region of origin of a blood spatter. The device used to inspect stains in 3D has been modified to be portable, possibly to a crime scene. The method however is based on accurately measuring the shape and minute volume of stains, and currently only works on smooth and non-absorbing surfaces.
To estimate the impact conditions from three-dimensional measurements of stains, it is important to know the physical properties of the blood. This project therefore also delivers measurement of the physical properties of porcine blood, a valid substitute of human blood. Measurements are performed both at equilibrium conditions and in dynamic situations close to those in BPA. A set of innovative and simple tools are proposed to characterize e.g. the shear-thinning viscosity, or the relation between drop spreading upon impact and impact conditions.
A numerical simulation tool has also been developed to describe the impact of blood drops, with a specific consideration of the role of wettability.
A review of the relations between fluid dynamics and bloodstain pattern analysis has also been published, with the purpose to bring together the fluid dynamics and the bloodstain pattern analysis communities, as per the recent wish of the National Academies.