Wales turns to DNA to identify dogs behind livestock attacks

Can DNA recovered from attacked livestock identify the particular dog responsible when its owner denies involvement? That is the question behind a Welsh evidence-gathering project—and the answer requires more than finding traces of canine DNA.

Farmers are receiving testing kits, Dyfed-Powys Police has assigned specialist crime-scene support, and researchers at Liverpool John Moores University will examine samples and develop collection guidance. Together with police powers introduced earlier this year, the work is intended to connect material recovered from an attacked animal to an identifiable dog. BBC Wales reported the latest developments on 29 October. ¹

The distinction is important: this is an effort to strengthen investigations, not an announcement that a particular attack has been solved. The available project material does not establish a measured rate for recovering usable profiles or identifying individual dogs, nor does it document a DNA-supported prosecution. That does not mean there have been no results; it means a dependable success rate has not been demonstrated in the evidence available here. ^{2 4 6}

For farmers, the stakes are already clear: injured or dead animals, financial losses and the distress of discovering an attack. North Wales Police Rural Crime Team constable Ian Nicholls described reports as “constantly coming in”. The approximate Wales-wide police count comes without a published counting methodology, so it should not be mistaken for a comprehensive census. ¹

NFU Mutual estimated that attacks cost insurers more than £300,000 in Wales in 2025, according to BBC Wales. Separately, Farmers’ Union of Wales material reported an NFU Mutual estimate of £385,000 for the value of livestock severely injured or killed that year, 23% above 2024. Those figures describe different measures—insurance costs and livestock value—and cannot be added together. Nor does the 23% rise establish an increase in the number of attacks. ^{1 7}

The investigative problem is not always proving that an animal was attacked. It is connecting that attack to a specific dog, particularly when its owner disputes responsibility. Wales wildlife and rural crime coordinator Rob Taylor identified that situation as a central use for the sampling power. ¹

The research is directed toward canine short tandem repeat, or STR, profiles. STRs are repeating stretches of DNA used in genetic profiling. Recovering a suitable profile from material on attacked livestock, then comparing it with a sample from a dog, is the route toward individual identification. Merely detecting canine DNA does not complete that task. ^{3 5}

A related laboratory method, qPCR, measures the amount of canine DNA in a sample. The published assay research describes that measurement as a quality-control step that helps prepare DNA for subsequent STR analysis—not as a test that independently identifies an individual dog. Research specifically addressing recovery of canine STR profiles from livestock-attack samples is described in the supplied record as a preprint, not peer-reviewed evidence establishing an operational success rate. ^{3 5}

In practical terms, the whole chain matters: suitable material must be recovered from the livestock, reach the laboratory in usable condition, yield a profile and support a comparison. Police gaining access to a suspect dog’s sample addresses one part of that chain. It does not guarantee that the attacked animal will provide enough usable evidence for the other side of the comparison.

The collection work predates this October development. Ten police forces in England and Wales participated in a DNA-collection trial in June 2024. Early Evidence Kits were developed to help recover offending-dog DNA from attacked livestock. That was an earlier trial, not ten forces newly joining the current Welsh effort. ^{9 6}

The present work brings kit distribution together with police support and university analysis. But distributing equipment and establishing a dependable collection procedure are different achievements. The project descriptions available here do not supply a complete handling protocol, a maximum interval between attack and collection, a laboratory turnaround time or a kit-distribution total. An exact deadline or promise of results would go beyond what has been established. ^{2 4 6}

University researcher Dr Nick Dawnay’s longer-term ambition also makes the remaining work clear: “Looking longer term, we'd also like to see whether it would be possible for farmers or vets to collect DNA evidence themselves if a crime scene investigator can't get there in time.” Farmers are already receiving kits, but that statement does not establish that a universally validated farmer-collection service is operating. ¹

That is a consequential distinction in a rural investigation. A kit may help preserve an opportunity to gather evidence; it is not itself a result. The value of this project will depend on whether collection and analysis consistently turn that opportunity into a useful comparison.

The legal change is already in force. The Dogs (Protection of Livestock) (Amendment) Act 2025 received Royal Assent on 18 December 2025 and took effect in England and Wales on 18 March 2026. March was commencement, not the date the legislation received Royal Assent. The changes give police powers to take samples from dogs in livestock-attack investigations. ^{13 14 1}

The law also replaced the former £1,000 maximum fine with an unlimited maximum. Courts can therefore impose a penalty above the old ceiling, but the punishment depends on the case. Neither taking a sample nor having the power to impose a larger fine establishes responsibility for an attack. ¹⁴

Much of the practical value may arise before a courtroom is involved. Taylor said 90% of attacks were resolved outside court, with owners often apologising, compensating farmers and attending an approved online responsible dog ownership course. His figure comes without a defined reporting period or case denominator. It is not a DNA success rate or a conviction rate, and it does not show that every remaining case reaches prosecution. ¹

Better identification evidence could nevertheless matter greatly in those discussions: accepting responsibility should not have to rest solely on an admission, and a disputed allegation deserves evidence. Wales now has stronger sampling powers and additional help collecting that evidence. What remains to be demonstrated is how reliably material from an attacked animal can identify an involved dog—and support a usable outcome when responsibility is contested.