The research, led by Samuel Appleyard-Sanderson of the University of Bristol together with colleagues, set out to test a concept known as targeted selective treatment, or TST. Instead of blanketing an entire herd with anthelmintics, TST treats only the animals carrying the most clinically significant parasite burdens, leaving the rest untreated. The idea was originally developed to slow the evolution of drug resistance in gastrointestinal parasites, by maintaining a reservoir of drug-susceptible worms on the pasture. But ecologists have proposed a second benefit: untreated animals produce uncontaminated dung, which could act as a refuge and recolonisation source for the flies and beetles that breed in cattle faeces and perform the vital ecosystem service of breaking it down.

To test this idea, the team built a spatially explicit simulation of a temperate pasture, represented as a 20-by-20 grid of 400 cells, each capable of holding a single dung pat. The model tracked three functional guilds of dung-inhabiting insects: dung-breeding flies such as the yellow dung fly Scathophaga stercoraria and sepsid flies, endocoprid beetles that live and feed within the pat, and paracoprid beetles such as Onthophagus and Geotrupes species that haul dung underground to breed in the soil beneath. Each guild was parameterised with life-history data drawn from published field and laboratory studies, capturing differences in fecundity, development time, lifespan, dispersal and, crucially, sensitivity to anthelmintic residues at different life stages.

The biological contrasts between the guilds are stark. Flies develop from egg to adult in roughly 20 days, lay around 95 eggs per clutch and live only about 24 days, but they remove very little dung individually and are highly sensitive to ivermectin at every life stage. Beetles are slower and longer-lived, with development periods ranging from about 45 days for endocoprid Aphodius species to 100 days for paracoprids, and much lower fecundity, but they are the workhorses of dung removal, with paracoprid larvae capable of processing nearly half a gram of dung per individual per day. In the model, larvae in contaminated pats suffered complete mortality, adult flies entering contaminated pats died outright, and adult beetles experienced variable mortality, while dung removal by larvae in contaminated pats was halved to reflect documented sublethal effects.

Each simulation ran for 1,000 daily timesteps, representing roughly five continuous grazing seasons. For the first 500 days, the insect communities established themselves without any drug exposure. Treatment then began, with newly deposited pats assigned as contaminated or clean according to the TST level, which ranged from zero, meaning every animal was treated, to one, meaning no animals were treated. The researchers ran ten replicate simulations for every combination of TST level and stocking density of 5, 10 or 15 animals per hectare, producing 330 individual runs in total. The model deliberately assumed that residues in dung remained at full strength until the pat degraded, a conservative choice that maximises lethal and sublethal effects over time.

The results reveal a genuinely complicated picture. As the proportion of untreated animals increased, total insect community density first dipped slightly and then climbed steeply, driven overwhelmingly by flies, which went from being the scarcest guild under full treatment to the most abundant when no animals were treated. Endocoprid beetles also increased, though more slowly. Paracoprid beetles, however, declined gradually as TST rose, a counterintuitive outcome the authors attribute to the breadth of species aggregated into that guild, which may have produced overgeneralised parameters. They note that paracoprid populations are naturally smaller than fly populations, so a community dominated by flies at maximum TST may actually resemble a more natural state rather than a degraded one.

Stocking density proved to be a critical modifier of these patterns. At the lowest density of 5 animals per hectare, insect numbers rose exponentially with increasing TST. At 10 animals per hectare, densities dipped slightly up to a TST of 0.5 before climbing sharply. At the highest density of 15 animals per hectare, insect numbers were actually greatest under full treatment, declined as TST increased, and only recovered partially as untreated levels approached one. The authors suggest this may reflect density-dependent competition: abundant dung at high stocking density initially supports large insect populations, which then rapidly degrade pats and trigger overcompensatory competitive exclusion. They caution that the densities modelled represent the upper end of UK grazing intensity and may be most applicable to intensive or rotationally managed systems.

Perhaps the most encouraging finding concerns dung removal itself, the ecosystem service with the most direct consequences for pasture health. Mean standing pat biomass decreased exponentially as the proportion of untreated animals rose, with reductions of roughly 13 percent at low stocking density, 20 percent at medium density and 31 percent at high density. In other words, leaving more animals untreated meant more insects survived to break down dung, so pats disappeared faster. This aligns with earlier empirical work showing that ivermectin contamination can cut dung removal rates by around a third to a half. For farmers worried that reducing treatment might leave pastures littered with undecomposed faeces, the model offers reassurance that the opposite is more likely to occur.

The study also identified an apparent threshold: reducing the treated proportion of the herd below roughly 50 percent seemed to bring comparatively little additional benefit for beetle populations, although the authors stress that this figure is sensitive to parameter values that remain poorly constrained empirically and should be treated as indicative rather than a precise management target. Fly populations, by contrast, continued to respond positively to reduced treatment well beyond that point, though flies contribute least to dung removal. The authors frame their findings as a theoretical foundation for promoting targeted selective treatment as a compromise between parasite control and biodiversity conservation within precision livestock farming, with direct relevance for agri-environment policy.

The model is not without limitations, which the authors discuss candidly. It includes only three guilds, omitting predators, parasites and the seasonal dynamics that govern real temperate communities, and it assumes constant residue concentrations and simplified colonisation rules, with flies restricted to one-day-old pats and beetles to pats up to five days old. Dispersal was modelled as purely distance-dependent, excluding long-range recolonisation. Residues in reality decline over time, so the model likely exaggerates long-term mortality, though the authors argue this would change the magnitude rather than the direction of their conclusions. What is now needed, they conclude, is empirical field testing of targeted selective treatment across whole farms, which would supply the species-specific dispersal and life-history data required to refine future models. Until then, the simulations offer the clearest picture yet that how, and how many, animals are treated may matter as much as the drugs themselves.

Subject of Research: Ecological effects of targeted selective anthelmintic treatment on dung-inhabiting insect communities in temperate pastures

Article Title: Simulated Targeted Anthelmintic Treatment Maintains Dung Removal but Not Dung‐Inhabiting Insect Communities

Article References: Appleyard‐Sanderson, S., Lurgi, M., Wall, R., & Clements, C. F. (2026). Simulated Targeted Anthelmintic Treatment Maintains Dung Removal but Not Dung‐Inhabiting Insect Communities. Ecology and Evolution, 16(10), Article e74477. https://doi.org/10.1002/ece3.74477

Image Credits: AI Generated

DOI: 10.1002/ece3.74477

Keywords: anthelmintics, ivermectin, targeted selective treatment, dung beetles, dung flies, livestock, ecosystem services, dung removal, simulation model, biodiversity, pasture management, endectocides

Negative density-dependence is the ecological workhorse of population regulation. As more larvae crowd into the same puddle or container, resources such as food and space become scarce, survival falls, and population growth slows. For malaria mosquitoes of the genus Anopheles, this density-dependent mortality during the larval stages is well documented and has long been built into models of mosquito population dynamics. It explains a familiar and frustrating pattern: when insecticides or larvicides kill off a large share of the mosquito population, the survivors suddenly face far less competition, their offspring survive at higher rates, and numbers bounce back toward the previous equilibrium.

Allee effects work in the opposite direction. Named after the ecologist W. C. Allee, they describe situations in which per-capita fitness declines as population density falls. The classic mechanism is mate limitation: when individuals are so rare that they struggle to find each other, reproduction stalls, and small populations can spiral toward extinction. Allee effects are a cornerstone of conservation biology, where they help explain why tiny populations of endangered species often cannot recover even after threats are removed. But for disease vectors, and for malaria mosquitoes in particular, they have received remarkably little attention. Indeed, no Allee effect has ever been empirically demonstrated in Anopheles populations, even though many aspects of mosquito biology, from the swarming behavior of males to the spatial distribution of breeding sites, make mate limitation biologically plausible.

To explore what would happen if such effects do exist, the team built a stochastic, stage-structured simulation model representing a simplified mosquito life cycle. The model tracked individuals through early and late larval stages and into adulthood, and incorporated the two regulatory processes at different points in the life cycle. Negative density-dependence acted on larval survival, mimicking resource competition during development, while a mate-finding Allee effect acted on total fecundity, reducing the number of eggs produced when adult densities were low. By varying the strength of each process independently and in combination, and by simulating different vector control scenarios, the researchers could ask how the interplay between the two forces shapes population persistence, rebound, and extinction.

The first and perhaps most counterintuitive result is that neither process, considered on its own, had a substantial long-term effect on whether the simulated mosquito populations persisted. Negative density-dependence did substantially alter the equilibrium abundance, the level at which populations settle, but it did not, by itself, push populations toward extinction. Allee effects alone, in this formulation, likewise did not dramatically change long-term outcomes. In other words, each regulatory mechanism in isolation leaves the population with a stable attractor that it can return to after disturbance.

Everything changed when the two processes acted together. As the strength of both negative density-dependence and the mate-finding Allee effect increased, the combined effect accelerated population decline and increased the likelihood of extinction. The mechanism is a kind of ecological pincer movement. Density-dependent competition keeps populations from exploding after a control measure removes competitors, while the Allee effect punishes the population for being small. A population knocked down by intervention cannot rely on the usual release from competition to fuel a rapid recovery, and at the same time its remaining adults struggle to find mates. The result is a downward trajectory that neither force could produce alone.

This interaction has direct implications for how control interventions are timed and sustained. The simulations compared sustained interventions with shorter-term campaigns, using immolating larvicides, larvicides that kill larvae, as the example intervention that reduces the larval population. Sustained interventions were able to push mosquito populations down to abundances low enough that Allee effects became important, and once that threshold was crossed, the probability of extinction rose. Short-term interventions, by contrast, allowed populations to rebound, driven by the negative density-dependence as formulated in the model: relieve the larval crowding, and larval survival improves, allowing numbers to climb back. The modeling thus provides a theoretical rationale for a principle that elimination programs have learned empirically, that consistency and duration of control matter as much as intensity.

The study draws on field experience from large-scale larvicidal control in Dar es Salaam, Tanzania, and the authors acknowledge the late Prosper Chaki, who helped explain the data from that program, as well as technicians and scientists at the Ifakara Health Institute. That grounding in real vector control settings matters, because the practical question the model addresses is not abstract. Malaria elimination campaigns across Africa increasingly rely on combinations of insecticide-treated nets, indoor residual spraying, and larval source management, and they face the recurring problem of populations that persist at low density or resurge once pressure eases. If mate-finding Allee effects operate in wild Anopheles populations, then sustained larviciding could be deliberately pushed harder and longer, not merely to suppress numbers but to drive them below the critical threshold where mating failure takes over and extinction becomes self-reinforcing.

The authors are careful about the limits of their findings. Because Allee effects have not been empirically demonstrated in malaria mosquitoes, the study is explicitly a what-if analysis: it evaluates what the implications would be if such effects exist, rather than claiming that they do. The model is a simplified representation of the mosquito life cycle, and the strength of the Allee effect in any real population would depend on details such as mating swarms, dispersal, and the spatial clustering of breeding sites that the model does not resolve. Empirical work to detect Allee effects in vector populations, perhaps through mark-release-recapture experiments or analyses of mating frequency at low density, would be the natural next step. The work was supported by the European Research Council under the European Union’s Horizon 2020 research and innovation programme, with additional support from the Bill & Melinda Gates Foundation.

Even with those caveats, the conceptual payoff is significant. The study shows that less-studied regulatory processes can identify which stages of the vector life cycle are resilient and which are vulnerable, and that the two can be leveraged together. A control strategy that ignores the possibility of Allee effects may underestimate how close a suppressed population is to elimination and abandon interventions prematurely, forfeiting a chance at local extinction. Conversely, a strategy that understands the trade-off can time and sustain its efforts to exploit the vulnerability of sparse populations. As malaria-endemic countries pursue elimination rather than mere control, models like this one suggest that the final, hardest phase of the campaign may depend not just on killing more mosquitoes, but on understanding the ecology of the few that remain.

Subject of Research: Population dynamics of malaria mosquito vectors and the interaction of negative density-dependence and Allee effects in vector control

Article Title: Implications of the trade-offs between negative density-dependence and Allee effects for vector control

Article References: Kipingu, A. M., Kiware, S., Haydon, D. T., Johnson, P. C. D., & Viana, M. (2026). Implications of the trade-offs between negative density-dependence and Allee effects for vector control. Parasites & Vectors. https://doi.org/10.1186/s13071-026-07634-4

Image Credits: AI Generated

DOI: 10.1186/s13071-026-07634-4

Keywords: Anopheles mosquitoes, malaria, vector control, Allee effects, negative density-dependence, larvicides, population dynamics, simulation model, larval source management, disease elimination, Parasites & Vectors, Tanzania