Every November, volunteers fan out along the California coast to count western monarch butterflies clustered in eucalyptus and Monterey pine groves from Bolinas to San Diego. In 2020, they found fewer than 2,000 – a number so low that some researchers privately discussed whether the population had crossed a point of no return. Two years later, counters tallied more than 300,000, a swing so dramatic it made national news. By the 2023 season, the count had dipped again.
That whiplash captures something important about the state of butterflies in North America. A federal analysis published by the U.S. Geological Survey, drawing on data originally reported in the journal Science, found that butterfly populations across the continent have fallen roughly 22% over the past two decades – a figure spanning hundreds of species and driven by habitat loss, pesticide use, and climate disruption. The western monarch, with its well-documented California colonies and its boom-and-bust recent history, offers one of the sharpest case studies of what those numbers look like on the ground and whether conservation can still bend the curve.
A continental decline, measured species by species
The 22% figure is not a rough estimate or an extrapolation from a single site. USGS researchers and their collaborators synthesized professional surveys and community science records – including data from the North American Butterfly Association and iNaturalist – to build trend lines for individual species across regions. Some butterflies declined far more steeply. Others held steady or even increased. But the overall trajectory pointed clearly downward, making the analysis one of the most comprehensive assessments of insect loss ever conducted on the continent.
The causes are layered. Agricultural expansion and urban development have erased grasslands and meadows that once supported larval host plants and adult nectar sources. Neonicotinoid and other systemic insecticides, applied at landscape scale, can reduce butterfly survival even at sub-lethal doses. And rising temperatures have shifted bloom times, migration windows, and overwintering conditions in ways that many species cannot track quickly enough. No single pressure explains the full 22% drop, and the relative weight of each factor varies by species and region – a complexity that makes policy responses harder to design but no less urgent.
Western monarchs: a population on the edge
Western monarchs migrate from breeding grounds across the interior West to a narrow strip of coastal California, where they overwinter in roughly 200 known groves. The Western Monarch Thanksgiving Count, organized by the Xerces Society for Invertebrate Conservation, has tracked these colonies since the 1990s and provides the longest continuous dataset for the population.
The numbers tell a stark story. In the late 1990s, counters regularly recorded more than a million monarchs. By 2018 and 2019, the count had plummeted to roughly 30,000 and then below 2,000 – a decline exceeding 99% from historical highs. The partial rebound to more than 300,000 in 2021 and 2022 showed the population could still surge when conditions aligned, but scientists cautioned that a single good year does not constitute recovery. Year-to-year swings of this magnitude suggest a population operating without a buffer, where one bad drought or one poorly timed cold snap could push numbers back toward crisis levels.
A peer-reviewed population viability analysis published in Frontiers in Ecology and Evolution in 2019 formalized those concerns. The model, built by researchers including Cheryl Schultz of Washington State University, used demographic data and threat scenarios to estimate the probability that the western monarch population would fall below a quasi-extinction threshold. Under baseline conditions – meaning existing pressures continued without significant new intervention – the probability of quasi-extinction rose sharply over a 50-year window. But the same model showed that targeted actions, particularly protecting and restoring milkweed-rich breeding habitat and reducing pesticide exposure at key sites, could substantially lower those odds.
That dual finding – high risk paired with a clear conservation pathway – has made the Schultz et al. study a touchstone in policy debates. It has been cited by the U.S. Fish and Wildlife Service, the International Union for Conservation of Nature, and dozens of conservation organizations working on pollinator protection.
Eastern monarchs face parallel risks
On the other side of the Rocky Mountains, eastern monarchs migrate to oyamel fir forests in central Mexico, where they cluster in colonies measured by the hectares of canopy they occupy. A USGS Open-File Report titled “Non-negligible near-term risk of extinction to the eastern migratory population of monarch butterflies” updated extinction-risk estimates using overwintering data from 2006 through 2022.
The report calculated explicit probabilities of population collapse over 10- and 20-year windows, using occupied forest area as a proxy for population size. Translating hectares into actual butterfly numbers requires assumptions about clustering density that vary by year and site condition, but the directional finding was unambiguous: the eastern population faces a non-trivial probability of falling below viable levels within a generation. Combined with the western PVA, the report reinforces a continental pattern. Monarchs on both flyways are in trouble, and the threats – habitat loss, agricultural chemicals, climate instability – overlap substantially.
What remains unresolved
Several gaps limit the precision of current projections. The western monarch PVA was built on data available through 2018, meaning it does not incorporate the near-collapse of 2020, the rebound of 2021-2022, or the climate extremes – record heat waves, prolonged drought, and atmospheric river flooding – that have reshaped California’s coastal ecosystems in the years since. No updated institutional model incorporating those conditions has appeared in the peer-reviewed literature as of early 2026.
The Xerces Society’s annual counts remain the best real-time indicator for the western population, but they have not yet been folded into a formal federal population viability update. That leaves a gap between what field observers are seeing and the modeling frameworks that inform policy. Until new analyses integrate recent data, statements about long-term recovery or collapse in the western population carry meaningful uncertainty.
Federal protection for monarchs also remains in limbo. The U.S. Fish and Wildlife Service has repeatedly deferred a listing decision under the Endangered Species Act. Without a formal listing, habitat protections depend on voluntary programs – such as USDA Conservation Reserve Program contracts that incentivize milkweed planting – and state-level rules whose effectiveness is difficult to measure at continental scale. Meanwhile, no primary study in the current literature isolates the relative contribution of specific pesticide applications to the multi-species butterfly decline documented by the USGS, so the chemical piece of the puzzle remains frustratingly imprecise.
Where conservation is gaining ground
Despite the uncertainties, the science points to concrete actions that work. Milkweed restoration in the western monarch’s breeding range – across Nevada, Utah, Oregon, and inland California – directly addresses the habitat bottleneck identified in the Schultz et al. PVA. Programs run by the Xerces Society, the Monarch Joint Venture, and state wildlife agencies have planted millions of native milkweed stems in recent years, targeting agricultural margins, roadsides, and public lands where monarchs are known to breed.
Protecting overwintering groves is equally critical. Many of the California coastal sites where monarchs cluster are on private land or in small municipal parks vulnerable to development, tree removal, or neglect. Local land trusts and city governments in places like Pacific Grove and Pismo Beach have taken steps to manage groves for monarch habitat, but the patchwork nature of these efforts means that a single landowner’s decision can eliminate a site used by tens of thousands of butterflies.
For the broader butterfly decline, the USGS analysis suggests that the same forces helping monarchs – habitat restoration, reduced pesticide intensity, and climate-adaptive land management – would benefit many of the hundreds of species losing ground. Butterflies are not just charismatic; they are pollinators, food sources for birds and other wildlife, and sensitive indicators of ecosystem health. Their decline signals problems that extend well beyond any single species.
How to weigh the next headline
Monarch stories tend to arrive in two flavors: alarm and hope. A bad winter count generates headlines about impending extinction; a good one prompts declarations of recovery. Neither framing captures the reality, which is that a population swinging between near-zero and a few hundred thousand is not stable – it is volatile, and volatility in a shrinking system is a warning sign, not a reassurance.
The strongest evidence available as of spring 2026 shows that North American butterflies have declined markedly over two decades, that both eastern and western monarch populations face elevated near-term extinction risks, and that conservation actions can meaningfully shift the odds. The open questions are not about whether a problem exists but about how fast it is unfolding, which pressures matter most in which regions, and which investments – per acre protected, per dollar spent – will deliver the greatest return.
For anyone trying to judge the next round of numbers, the most useful question is not “up or down?” but “compared to what?” A single year’s count matters less than whether it fits within a recovering trend or a declining one. And the models that matter most are not the ones that predict a date of extinction but the ones that identify which levers – habitat, chemicals, climate – move the needle furthest when pulled.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
