When wildlife managers return desert bighorn sheep to mountains they vanished from decades ago, the hardest question is not how to release them — it is whether the animals will choose to stay. A study published in the journal Movement Ecology offers a new way to read the answer, by weaving the sheep’s own repeated behaviour directly into the models scientists use to predict where wild animals will live.
Why Habitat Models Keep Missing Something
Traditional resource selection analysis works like a real-estate ranking: researchers compare the places an animal used against the places it could have used, and score terrain, water, forage and cover accordingly. The method is powerful, but it quietly assumes the animal moves through the landscape as if every visit were its first. Real sheep do not. A ewe that found safety on a particular ridge returns to it; a ram that crossed a valley once remembers the route. Behaviour, in other words, has a memory — and the memory shapes the map.
The research team describes their approach in Movement Ecology, integrating what they call recursive behaviour — revisits, returns and repeated use of favoured spots — into standard selection analysis, then testing it on a restocked population of desert bighorn sheep.
What the Sheep Revealed
Once revisiting was counted, the picture of what the sheep valued shifted. Areas near escape terrain — the steep, rocky ground where bighorns outclimb predators — mattered even more than earlier models suggested, because animals returned to them again and again. Seasonal water sources showed a similar pattern: not just visited, but woven into repeated circuits. The result is a more honest map of habitat quality, one that reflects how the animals themselves rank the landscape through their choices over time.
For restocking programmes, the practical stakes are high. Reintroductions fail quietly when released animals drift away into poor habitat or human conflict. A model that predicts not just where sheep can live, but where they will keep choosing to return, helps managers pick release sites and judge early success far more accurately.
A Wider Lesson for Conservation
The same logic applies far beyond one desert species. Elephants retrace ancestral routes, wolves revisit rendezvous sites, and migratory birds return to the same wetland corners year after year. Conservation planning has sometimes treated such patterns — like the shifting availability of key players we analysed in a very different field, football’s structural cost of absences — as noise around the real signal. This study argues the repetition is the signal.
The IUCN has long stressed that reintroduction success depends on understanding species behaviour, not just habitat maps. Tools that formally include behavioural memory give that principle a practical method.
The Bottom Line
Animals do not experience a landscape as a spreadsheet of resources; they experience it as a set of remembered places, trusted routes and known refuges. By teaching habitat models to remember what the sheep remember, this research brings the science of where animals live closer to how animals actually live — and gives restocked populations a better chance of staying home.