Reconstructing sudden ambient temperature changes in death time estimation from temperatures in two closed compartments using a Bayesian double-exponential approach
Year: 2026
Type: article
Abstract: Traditional methods of time of death estimation in forensic casework mostly assume a constant ambient temperature $$T_a$$ , which can lead to significant errors. Reconstructing ambient temperature changes at a crime scene a posteriori can therefore be an important task. As a first step, we reconstruct a single sudden drop in $$T_a$$ at unknown time $$t_0$$ using a posteriori temperature measurements $$T_Z(t)$$ , $$Z \in \{X, Y\}$$ , in two closed compartments X , Y (boxes, cupboards,..) at the crime scene. As a reliable reconstruction based on a simple Newtonian cooling model turned out to be not straightforward in a previous study, we apply a Bayesian framework with a double-exponential cooling model for $$T_Z(t)$$ . The approach is tested in two experimental scenarios in a climate chamber under controlled conditions. The results show a reduced reconstruction error in either $$T_a$$ , or $$t_0$$ , or both, compared to a simpler approach using Newtonian cooling and maximum likelihood estimation. The study thus represents a proof of concept giving hints for further research.
Source: Scientific Reports
Institutions: Jena University Hospital, Zuse Institute Berlin
Language: English
Subfield: Insect Science
Domain: Life Sciences
Open Access status: gold
Estimated APC paid: $2,690
Awards: G1300450862, G6982419574
OpenAlex ID: W7234380656
