Fractional Model and Numerical Algorithms for Predicting Cactus Cochineal (Dactylopius opuntiae) Infestation on Prickly Pear with Quarantine and Treatment Strategies

In the arid landscapes of North Africa, the prickly pear cactus is more than just a plant; it is a lifeline. For centuries, these hardy trees have provided food, fodder, and income for rural communities across Morocco and the wider Maghreb region. However, since 2014, a tiny but devastating insect known as the cactus cochineal has threatened to erase this agricultural heritage. This pest, which feeds on the cactus sap, is difficult to manage because it hides under a protective layer of white, cotton-like wax that shields it from standard treatments. When an infestation becomes severe, covering more than three-quarters of a cactus pad, the plant withers and dies. To combat this, authorities have implemented strict measures, including quarantining entire zones, treating infected plants, and destroying those that are too far gone. Yet, despite these efforts, the pest continues to spread, and nearly three-quarters of the crops in some areas remain infested.

Understanding how this pest moves and how to stop it requires more than just counting dead plants; it requires understanding the hidden history of the infection. Traditional models often assume that a plant's reaction to an infestation happens instantly, like a light switch flipping on. However, nature is rarely so immediate. The effects of a pest attack, the time it takes for a colony to become visible, and the lingering impact of past weather conditions all create a kind of "memory" in the system. A new study by researchers in Morocco and France explores whether using mathematics that accounts for this memory can better predict the spread of the cochineal and improve the strategies used to fight it. By treating the spread of the pest not as a simple, instant reaction but as a process influenced by its past, the researchers aim to refine the tools used by farmers and policymakers to protect the prickly pear.

The researchers built a detailed mathematical model that divides the cactus population into six distinct groups: healthy plants, those secretly carrying the pest before symptoms appear, those showing visible signs of infection, those under quarantine, those receiving treatment, and those that have been destroyed. They then applied a specific type of advanced mathematics, known as a fractional derivative, to this model. In simple terms, this mathematical tool allows the model to remember the past states of the system. Instead of asking only what is happening right now, the model considers how the current situation was shaped by events that happened days or weeks ago. This approach mirrors the reality of the cochineal, whose protective wax delays the effect of treatments and whose colonies take time to become visible to inspectors.

When the team ran their simulations, they found that this memory effect significantly changes the timeline of an outbreak. While the final number of infected plants remained the same regardless of whether the model included memory, the path to get there was different. In the memory-based model, the peak of the infestation arrived later and was slightly less intense than in traditional models. This delay is crucial for management; it suggests that the system reacts more slowly to changes, giving authorities a slightly longer window to act before the situation becomes critical. The study also tested how different control measures, such as quarantining healthy plants or treating infected ones, would perform under these conditions. They discovered that quarantining healthy plants to prevent them from entering the infection cycle was far more effective at reducing the peak of the outbreak than treating plants that were already sick. This aligns with the logic that stopping the spread at the source is better than trying to cure the damage after it has occurred.

To ensure their model was reliable, the researchers tested it against synthetic data, which are computer-generated scenarios that mimic real-world field surveys. They tried to see if the model could correctly identify the specific rates at which the pest spreads and how effective quarantine measures are. The results were encouraging for some factors but revealed a challenge for others. The model could accurately pinpoint the effectiveness of quarantine efforts, a key piece of information for policymakers. However, it struggled to distinguish between the rate at which the pest spreads naturally and the rate at which treatments work when only looking at the total number of infected plants. This suggests that to get the most accurate picture, authorities would need to measure the spread of the pest through direct observation of the insects themselves, rather than relying solely on the count of infected plants.

The study went a step further by asking what the best possible strategy would be if authorities could adjust their efforts dynamically over time. Using a method that finds the most efficient path to a goal, they simulated a scenario where biological controls, such as releasing predatory ladybugs, and administrative controls, like reinforced patrols to enforce quarantine, could be turned up or down as needed. The simulation showed that the most effective approach involves a rapid, intense burst of quarantine measures right at the start of an outbreak, followed by a gradual tapering off as the threat recedes. This strategy could reduce the peak number of infected plants by nearly 72 percent compared to doing nothing. However, the model also uncovered a difficult trade-off. The aggressive quarantine measures that successfully stop the spread also lead to the destruction of more healthy plants. Because the quarantine zone is strict, many plants that are actually healthy but are located within the restricted area are eventually destroyed as a precaution. The researchers found that by adjusting the model to explicitly value the preservation of healthy plants, decision-makers could find a balance between stopping the pest and saving the crop, though some loss of healthy plants remains inevitable under strict quarantine.

Finally, the team applied this optimal control strategy directly to their memory-based model. They found that when the system's memory is taken into account, the control strategy becomes slightly more efficient. The "memory" of the system means that the effects of past interventions linger, allowing the optimal strategy to achieve a marginally better result with the same amount of effort. This suggests that the very nature of the cactus cochineal's slow, cumulative spread can be turned into an advantage if the response is timed correctly. The study concludes that while the mathematical tools used are complex, the insights they offer are practical. By acknowledging that the spread of this pest is not an instant event but a process with a history, authorities can better time their interventions. The research highlights that while no single strategy can eliminate the threat entirely, a combination of early, aggressive quarantine and careful, targeted treatment offers the best chance of protecting the prickly pear industry, provided that the inevitable cost of destroying some healthy plants is accepted as part of the necessary defense.