PPC123 | TECHNICAL

Dr Sophie Wulff of Arctech Innovation explores how climate change, resistance and invasive species are reshaping pest control, and why scent science and data-driven detection will be critical to staying ahead of the next wave of pest threats.

You smell. Every single one of us does. Some more than others. That might sound flippant, but smell has a bad reputation for a reason. It’s tied to disgust, decay and danger. That reaction is a survival mechanism. If something smells wrong, we instinctively avoid it because, historically, that kept us alive.

But smell also has a positive side. It’s a hidden language that humans and animals use constantly, often without realising it. There’s a famous experiment where men were asked to wear the same t-shirt for several nights, after which women were asked to smell them and choose the most appealing scent. The women consistently preferred the scent of men who were genetically most different from them. Smell, it turns out, plays a quiet but powerful role in behaviour.

Where I work, we’re unlocking that hidden language. And it matters, because the future of pest detection is data-driven. Some of the most effective tools we need have been right under our noses all along.

Pest volatility is the real challenge

Before getting into scent science, it’s important to address what I see as one of the biggest challenges facing pest management right now. Pest volatility.

We’re seeing increasing changes in pest patterns, and those changes are accelerating. Climate change, global travel and insecticide resistance are all driving forces behind it.

Take the Asian tiger mosquito. It can transmit diseases such as dengue and yellow fever and has already been detected breeding in parts of Kent and the South East. At the moment, UK winters are still cold enough to prevent it overwintering, so it’s classed as detected but not established.

But it is now established in France, having moved north from the Mediterranean.

What makes it particularly risky is egg survival. The eggs can survive desiccation. They can be laid on something like a tyre in southern Europe, transported to the UK, and with just a small amount of rainfall, hatch into breeding adults.

Rodents are another example. We’ve lost winter kill. Harsh winters used to reduce rat populations, but that’s no longer reliable. Since the 1980s and 1990s, rats have increasingly bred right through December and January, leading to spring population surges in cities like London and Manchester.

Then there’s resistance. Bed bugs have developed genetic knockdown resistance, thicker cuticles and behavioural adaptations that reduce the effectiveness of some sprays. German cockroaches have evolved glucose avoidance, rendering certain baits ineffective. Global travel spreads these resistant populations quickly, particularly through transport hubs.

Invasive species are also pushing further into the UK. Asian hornets threaten honeybees. Oak processionary moths are now common in London parks and are moving north. Tropical ants such as pharaoh ants and ghost ants are thriving in heated urban environments.

And with warmer temperatures come shorter breeding cycles. Insects are cold-blooded, so their metabolism and reproduction speed increase with heat. A two-degree rise can mean one to five extra generations per season. House flies are a good example. At extreme temperatures, eggs that normally hatch in 20 hours can hatch in eight, and larvae can become adults in four days. This is what unpredictability looks like in real terms.

Why detection matters more than ever

So how do we take control of that unpredictability? The answer, whether we like it or not, is data-driven pest control. I can’t pretend to know how to run a profitable pest control business. That’s your expertise. But all roads right now are leading towards better data.

Knowing what species you have, exactly where it is, when it arrived and how many there are changes everything. Sometimes one pest is too many. Sometimes it’s population dynamics that matter. Accurate detection puts you in a position to respond efficiently and proportionately.

The industry is moving from reactive to predictive. Traditionally, you visit a site, find the infestation and deal with it. That’s time-consuming and often means you’re chasing a problem that’s already established.

New detection methods, including AI-powered sensors, monitor sites 24/7. They don’t just tell you there’s a pest. They tell you when it arrived. Catching a single scout rat or one bed bug before it breeds can prevent major clean-out jobs and repeated visits.

We’re also seeing a shift towards digital proof of absence. Companies are starting to charge for the absence of pests, not just for killing them. That’s a big change in business models. Detection supports non-toxic monitoring, keeps high-value contracts active and provides audit readiness and compliance assurance.

Accurate detection also allows targeted treatment. Instead of blanket control, technicians can act precisely. You become a sniper rather than a carpet bomber. That has obvious implications for chemical resistance and integrated pest management.

Smell as an underestimated science

This is where scent science comes in.

Smell has always been used in detection. Dogs are an obvious example. A dog’s olfactory brain is around 40 times larger than ours. I was involved in a UK study at the London School of Hygiene and Tropical Medicine looking at canine detection of Covid. We collected odour samples from around 4,000 participants and identified a very specific scent associated with infection, even in asymptomatic cases. But insects are even better sniffers than dogs, just in a different way.

Dogs are high-definition scanners. They build complex scent pictures and have strong scent memory. Insects are more like laser-focused sensors. Their receptors are fewer but far more specialised. A male silkworm moth can detect a single molecule of a female’s pheromone and respond instantly by flying upwind.

Insects don’t explore with smell. They react to it. Their olfactory system is a set of “if-then” triggers. If pheromone, then mate. If CO2, then bite. If glucose, then eat. That precision is what makes scent so powerful.

“Insects don’t explore with smell, they react to it. That precision is what makes scent so powerful.”

Learning from insect chemistry

Chemicals are the language insects use. Pheromones communicate within a species. Kairomones benefit the receiver, like mosquitoes detecting human CO2. Allomones benefit the sender, such as defensive odours. Synomones benefit both, such as plants attracting parasitic wasps to kill caterpillars.

These chemical signals are already used in pest management, particularly as attractants. One example my team has worked on extensively is bed bug aggregation pheromones.

Research has shown that bed bug faeces contain aggregation pheromones that attract other bed bugs to refuge sites. In laboratory studies, bed bugs cluster strongly around these odours. When hungry, they follow CO2 and human odours. When looking to hide, they follow aggregation pheromones.

By identifying the specific chemical components and releasing them in the correct ratios, we can replicate that signal. The result is an attractant that works with the human host rather than competing with it. These traps are particularly effective for low-level infestations and for pre- and post-treatment verification.

Replicating the insect brain with AI

Odours are complex mixtures. Traditional gas sensors can detect single chemicals like CO2, but odours are more like fingerprints or musical scores. What’s been missing is the ability to interpret those complex signals. That’s where AI comes in.

We’re using machine learning to replicate neural networks found in insect brains. Sensor arrays collect volatile organic compound data, which is transmitted to the cloud and decoded by algorithms trained to recognise specific pest odour patterns.

Each device generates rich data: where, when, what and how many. Combined over time with historical and meteorological data, this allows us to predict and prevent outbreaks rather than simply react to them.

Remote monitoring has had its ups and downs, but it’s improving rapidly. Whether through motion, cameras or scent detection, it’s becoming an essential tool for managing unpredictability.

Where this leaves the industry

We’re not quite at fully predictive, climate-adaptive heat mapping yet. But that’s where the industry is heading.

Pest volatility isn’t going away. Detection is becoming as valuable as treatment. Data is becoming a commercial asset. And scent science, long underestimated, may be one of the most powerful tools we have to stay ahead.

The future of pest control is about making the unknown known. And sometimes, the answer really has been right under our nose all along.

What’s coming next for pest detection?

Scent-enabled AI detection
At Arctech Innovation, work is underway on scent-based detection systems that use AI to interpret complex pest odour profiles. Rather than simply flagging activity, these tools aim to identify the species present, pinpoint when it arrived and, over time, provide deeper biological insight such as sex or reproductive status.

24/7 remote monitoring sensors
Sophie also highlights the growing role of continuous monitoring using motion sensors, cameras and environmental detection. The value lies in knowing exactly when a pest appears, enabling earlier intervention and avoiding established infestations.

Digital proof of absence
Detection technology is enabling a shift in how pest services are delivered and charged for. Monitoring systems are increasingly being used to demonstrate the absence of pests, supporting compliance, audit readiness and long-term contracts rather than reactive call-outs.

Advanced attractant-based traps
Building on scent science, more chemically precise pheromone and odour-based traps are being developed, particularly for pests like bed bugs. These tools are designed for early detection, verification and low-level infestations, working alongside treatments rather than replacing them.

Climate-adaptive heat mapping
Sophie also points to AI systems being developed to combine detection data with environmental and meteorological information. While still emerging, this approach could allow pest professionals to anticipate outbreaks and plan resources more effectively.