kbc@keithborer.co.uk +44 (0)191 332 4999
Home Services Our Team News FAQ Vacancies

Sniffing out the Truth: The use of Canines when looking for Ignitable Liquids

August 25 2022

Sniffing Out The Truth

Related Services

Fire Investigation

A conviction for Arson with intent to endanger life can result in a long custodial sentence. Consequently, it demands that appropriate evidence is adduced by the prosecution to ensure the basis of a conviction is sound.

In a number of arson cases we encounter, it is suggested by the prosecution that ‘accelerants’ have been used in order to aid the fire’s ignition and/or to help the fire to grow and spread more quickly. This assertion has been used to support the charge of intent.


These days, the terminology ‘ignitable liquid’ is preferred, as the use of the word ‘accelerant’ both implies an intention by the alleged fire starter to accelerate a fire (which might not be the case) and ignores the possibility that other non-liquid materials can be used as an “accelerant”, such as balls of crumpled up newspaper.

Despite a raging hot and prolonged fire, ignitable liquid residues often survive in the debris and in protected areas, for example under skirting boards or in carpet underlay. Specially trained canines and their handlers are frequently called to search for these residues.

Sensitive noses

It is widely acknowledged, in the scientific literature and the fire investigation community, that such canine teams can be a valuable asset to the fire investigation process. The dogs are trained to detect all of the commonly used liquids such as petrol, white spirit, lighter fluid etc, as well as other less common hydrocarbon compounds and mixtures. Dogs can detect low level odours beyond the human detection range and they can quickly and effectively search a large area for ‘indications’ of these.  As a result, these dogs assist in targeting areas for the collection of debris or items which may contain ignitable liquids. They are an effective and efficient presumptive test to be deployed in conjunction with laboratory analysis.

Confirmatory testing

The problem we see is a lack of follow up when the dog finds something of interest. After all, a dog can’t tell you how sure they are; they can only indicate an area where they think the remains of an ignitable liquid are still present. As might be expected, sometimes they will get it right, and sometimes wrong. This is why there is a decades old, standard forensic process where, once a dog has indicated a potential location, a suitably trained person, such as a CSI, takes samples from the relevant areas to submit for laboratory analysis for definitive confirmation (or otherwise) of an ignitable liquid.

We still see cases where this is not done – either samples are not taken following canine indications, or they are taken but not sent for analysis. Why this happens is unclear, but the canine’s indication alone might be the key piece of the evidence presented by the prosecution as to why they believe an accelerant was used.


The recommendation for laboratory confirmation is set down in the USA publication NFPA 921 ‘Guide for Fire and Explosion Investigations’ (version 2021)[1] produced by the National Fire Protection Association (USA). This organisation, amongst others, has been responsible for developing and promoting the use of the ‘Scientific Method’ across the global fire investigation community, including here in the UK. NFPA 921 is accepted as an authoritative text for fire investigation and a core reference text by the UK’s fire investigation community.

The following factors need to be considered when making use of canines at fire scenes:

Canines (and their handlers) cannot be certain what has actually been ‘indicated’.

Ignitable liquid residues often have no smell detectable to humans. A dog cannot give evidence but even if they could, they could not confirm they found an ignitable liquid, only they thought one was present.  Nor can a human handler verify the dog’s indication at the time.  Unless a scientist analyses samples from the area(s) indicated by a dog, then even if there is a smell, any finding is speculative and should be considered improper evidence. Section 17.7.7 of NFPA 921 says ‘Any canine alert not confirmed by laboratory analysis should not be considered validated and, accordingly, should not be offered as direct or circumstantial evidence of the presence of an ignitable liquid in a criminal or civil trial’.

It is possible for canines to give a false positive or false negative result.

We have seen cases where dogs have ‘indicated’ an area, yet subsequent lab analysis of samples from that area has not found an ignitable liquid.  Sometimes the handler will argue that the dog’s nose is more sensitive than the analytical method, but fires produce thousands of chemicals in the debris, some of which are of the type that dogs are trained to detect and the dog may struggle to distinguish between them.  In a case such as this, it is not possible to know the ‘true’ answer so, unless there is other supporting evidence of the use of an ignitable liquid, the dog’s indication cannot be relied upon.

We have also seen examples where a very high level of an ignitable liquid produced a false negative in a canine.

Even if an ignitable liquid is confirmed by a laboratory analysis, an innocent explanation for the presence of the substance may exist.

Even when the presence of ignitable liquid residues is confirmed by lab analysis, it is important to consider if there is a reasonable explanation for its presence. For example, petrol residue in a garage amongst tools powered by petrol would not necessarily be unexpected. The presence of petrol in someone’s living room is another matter, but we have had cases where unconventional lifestyles have meant that various rooms in houses were used to store/repair/clean petrol fuelled items, such as motorbikes and lawnmowers. Consequently, the presence of an ignitable liquid alone is not necessarily proof that it was used as an ‘accelerant’ at a fire scene. The presence of ignitable liquids needs to be considered in the wider context.

Ignitable liquid residues can also be inadvertently moved around at a scene, for example on firefighter’s boots or on fire hoses. Also sampling, handling, storage and potential cross contamination of samples whilst in the care of the prosecution are all matters that require careful scrutiny on behalf of the defence. In this regard, an empty (‘control’) bag from the batch used for collecting samples of scene debris or a suspect’s clothing should be submitted to the laboratory to show that the bags have not been previously contaminated.

We occasionally see cases where contamination has occurred or cannot be ruled out as a control bag was not submitted.

False positives

Canines represent a valuable tool in fire investigations, but a canine indication without laboratory confirmation carries a realistic chance of it being a false positive.  If this is then presented as key ‘evidence’, it falls short of an acceptable standard of practice and it brings a huge and unnecessary risk to the Criminal Justice System. By using a biology comparison, if a red stain, located in a key position, was ‘indicated’ to be blood based on a colour test but not investigated further using DNA analysis, this would be considered by the courts to be unacceptable evidence of the presence of blood. Yet this is effectively what occurs when canine ‘indications’ are not confirmed by laboratory analysis.

This article concerns the use of canines to indicate the presence of ignitable liquids at a scene, but the same principals apply when a fire investigator makes use of a portable, handheld “sniffer” which is a device that detects a range of volatile organic chemicals (VOCs) including, but by no means limited to, things like components within petrol residues etc.  Any findings from a ‘sniffer’ indication must also be sampled and analysed at a laboratory to determine if an ignitable liquid residue is present.

KBC Fire investigation experts

Keith Borer Consultants has a team of three proficient fire investigators who routinely assess fire origin, cause and factors involving fire movement and spread, including risk to life. They are instructed by both the prosecution and defence, to consider cases ranging from small, single item fires to multi-million-pound loss fires and multi-fatality fires. If you wish to discuss a case, please contact Jennefer Gray, Alan Henderson or Dr David Schudel on 0191 332 4999 or email kbc@keithborer.co.uk.


[1] National Fire Protection Association (NFPA) 921 Guide for Fire and Explosion Investigations, 2021


Alan Henderson
BSc(Hons), CChem, MRSC

Subscribe to our mailing list