Harvester ants fly on the rain, not the date

If you keep half an eye on the desert Southwest through the summer, you've probably heard that harvester ants fly "with the monsoon." That's true, but it undersells how literally true it is. This isn't a loose seasonal association, ants flying sometime in the general vicinity of a wet month, so much as a hard dependency: the flight doesn't happen until the rain does, wherever and whenever that turns out to be.

That distinction matters because the monsoon itself is a front, not a single date, arriving at different places on different days, some years earlier and some later, and in places drier than others.

The North American monsoon is the seasonal shift that brings the Southwest's summer thunderstorms, and the standard reference for its timing, Adams and Comrie's 1997 paper in the Bulletin of the American Meteorological Society, puts the whole event at roughly June through mid September. It doesn't switch on everywhere at once. Moisture builds along the western slopes of the Sierra Madre Occidental in Mexico first, then pushes into New Mexico and southeast Arizona in early July, and the pattern fully matures across the wider Southwest by mid July, once a high-pressure ridge sets up over the Four Corners region and starts pulling moist air in from the southeast. By mid September the cycle typically loses its energy and drier conditions return.

That means, even within a single monsoon season, the rain a harvester ant colony is actually waiting for lands on a rolling schedule that runs roughly west and north through the summer. A colony near the Arizona-New Mexico border can be sitting under useful rain while one further west and north is still bone dry.

Pogonomyrmex is a genus of seed-harvester ants spread across the Chihuahuan, Sonoran and Mojave deserts, and across nearly all of that range, mating flights have only been documented happening in summer, in direct response to those seasonal rains. That's the finding Helms and Helms Cahan state plainly in the opening of their 2010 paper on the genus, and it sits on top of decades of prior field observation going back to Hölldobler's 1976 study of Pogonomyrmex mating behaviour, the paper that's still the standard reference for how the genus behaves around a flight.

The logic isn't mysterious. A newly mated queen has to dig into soil and hold moisture in a sealed chamber while she raises her first brood on nothing but her own reserves, taking in no food from outside. Try that in dry, baked desert ground and she has no chance. Wait for the ground to actually be wet, and the odds tip back in her favour. So the whole genus, across three deserts and a lot of variation in exactly which weeks are wettest, converges on the same rule: don't fly until the rain has fallen.

The cleanest demonstration of that rule comes from a colony that never got the chance to follow it.

In parts of the western Mojave Desert, summer rain barely happens at all. Almost all the precipitation there falls in winter and early spring instead. Helms and Helms Cahan went looking for Pogonomyrmex rugosus there and, on 18 March 2009, near Mojave in Kern County, California, found exactly what the genus's usual rule would predict once you swap in the region's actual rainy season: three of 48 colonies surveyed over a four-hour period had alates streaming out and taking off, at 12:30, 13:00 and 14:00 local time. A dealate female was found walking the following day at a second site 33 km, or 20 miles, south, and a foundress was excavated from a starting nest three days after that. A neighbouring seed-harvester, Messor pergandei, had already had its first big flight of the year at the same site on 17 March.

That's a flight four to six months earlier than P. rugosus manages anywhere else in its range, and the authors are careful about what it does and doesn't prove. Either the western Mojave population has fully converged on the winter-and-spring-rain timing that other desert species there already use, or the colonies still produce alates on the usual summer schedule and simply hold them in the nest, unflown, until the following winter's rain finally gives them an opening. Both explanations point at the same underlying mechanism: it's the rain being tracked, not a season as such, and where the region's rain moves to a completely different part of the year, the flight moves with it.

The rain dependency covers getting a queen into the ground. What happens after that varies even within Pogonomyrmex.

Most of the genus founds fully claustrally, sealed in and living on reserves alone, but Johnson documented an exception in Pogonomyrmex californicus, which founds semi-claustrally instead: a queen who has to leave the chamber and forage while raising her first brood, rather than living entirely off what she brought down from the flight. It's the same point made elsewhere on this site about founding strategy varying by species rather than by genus, and it means the advice for a Mojave P. rugosus queen and a P. californicus queen from the same general region isn't automatically the same, even though both of them only got that far because rain let them fly in the first place.

The monsoon's job isn't finished the moment a queen mates and comes down. It has to keep delivering, or the flight was wasted.

Johnson's long-term study of Veromessor pergandei, a related Sonoran Desert seed-harvester, put a number on exactly how often that happens. Across 20 years of monitoring in central Arizona, successful recruitment, meaning young colonies actually showing up and surviving, occurred in only 4 of them. A statistical model using nothing but total rainfall from January through June correctly predicted which years those would be 17 times out of 20. Queens dug out of the ground after two weeks of dry conditions were measurably more dehydrated than queens still inside a wet natal nest, which is the mechanism behind the pattern rather than just a correlation: dry ground after the flight kills founding queens directly, not just indirectly by preventing the flight in the first place.

That's a sobering ratio for anyone picturing every rainy monsoon evening as a sure new generation of ants. Most years, in this part of the desert, the rain that triggers the flight isn't followed by enough rain afterward to carry queens through founding. The flight is a bet the colony has to make anyway, because waiting for certainty would mean never flying at all.

None of this is really about Pogonomyrmex alone. It's a fairly complete demonstration of something true more broadly, and easy to lose track of if you only ever check a forecast for a fixed date: the trigger is the weather event, not the week it usually happens in. A dry monsoon pushes a flight late or thins it out. A wet one earlier than usual pulls it forward. In a place with no reliable summer rain at all, the flight can end up somewhere else on the calendar entirely, as the Mojave data shows about as clearly as a single field observation can.

If you're anywhere in the Southwest watching for harvester ants this month, the practical version is simple: track the rain in your specific area rather than a generic date, because the same species a hundred miles away, under a different piece of the monsoon, may already be finished or not have started. Ants don't read the barometer goes into what does and doesn't move the needle on a flight more generally, and fully claustral or not covers what a queen from any of these species actually needs once she's down and out of the rain.

Dealate scores flight odds hour by hour from live weather for the species active near you, rain included, rather than pinning hopes on a date that a desert doesn't actually keep to. A reason to look up after a storm has passed, not a guarantee of what you'll find.

Helms, K.R. & Helms Cahan, S. (2010). Divergence in mating-flight patterns of the seed-harvester ant Pogonomyrmex rugosus (Hymenoptera: Formicidae) in the western Mojave Desert. Myrmecological News 13: 15 to 17. Source of the field observations near Mojave, California, and the "only known to occur in summer in response to seasonal rains" statement for the genus.

Hölldobler, B. (1976). The behavioral ecology of mating in harvester ants (Hymenoptera: Formicidae: Pogonomyrmex). Behavioral Ecology and Sociobiology 1: 405 to 423. The foundational study of Pogonomyrmex mating behaviour that later work, including Helms & Helms Cahan, builds on.

Adams, D.K. & Comrie, A.C. (1997). The North American Monsoon. Bulletin of the American Meteorological Society 78(10): 2197 to 2213. Source of the monsoon's typical June to mid-September timing and its early to mid-July arrival across Arizona and New Mexico.

Johnson, R.A. (2002). Semi-claustral colony founding in the seed-harvester ant Pogonomyrmex californicus: a comparative analysis of colony founding strategies. Oecologia 132(1): 60 to 67. Cited here for the title's finding, that founding strategy varies within Pogonomyrmex rather than being fixed for the whole genus.

Johnson, R.A. (2021). Desiccation limits recruitment in the pleometrotic desert seed-harvester ant Veromessor pergandei. Ecology and Evolution 11: 294 to 308. Source of the 20-year recruitment data, the discriminant-analysis figure linking rainfall to successful colony establishment, and the dehydration measurements in founding queens.