Forensic analysis of gasoline residues in urban soil samples: Consequences for vehicle arson investigation

A substantial proportion of forensic fire investigations involve vehicle arson, where severe thermal damage often limits direct evidence recovery. Consequently, soil samples from the vicinity are frequently analyzed for ignitable liquid residues (ILRs). This study evaluates the forensic reliability of gasoline detection in common urban substrates and defines substrate-dependent detection windows. A total of 100 background soil samples from active parking areas were analyzed by gas chromatography-mass spectrometry (GC-MS); no gasoline residues were detected, indicating a low probability of detectable background contamination (<3.2% at 96% confidence). Controlled spill experiments demonstrated marked variability in persistence: up to 36 h in interlocking stone sand, 5 days in bulk sand, 3 weeks in kurkar, and up to 1 year in asphalt. The addition of water prior to sampling substantially extended detection in porous substrates, increasing the detectable time window in kurkar. Gravimetric analysis showed exponential type evaporation behavior for gasoline, with a transition toward linear profiles following water addition. These results demonstrate that gasoline persistence is strongly substrate-dependent and that moisture conditions substantially influence detectability. They improve interpretation of ILR findings in vehicle arson investigations and support further consideration of targeted pre-sampling water application as a potential approach to enhance ILR recovery from porous substrates.