A few weeks ago, a reporter asked me whether relatively high western monarch butterfly counts again this year mean that the population has recovered. Although it is good news that the population counts are higher than last year, I couldn’t reply that the population has recovered. Since the start of the Western Monarch Count, the “average” rate of change of overwintering monarch counts in California is a 5% decline per year. We also don’t yet have enough years of high counts to know whether increases during the past two years are likely to persist.
For insects, yearly population change can vary widely
Populations of butterflies, like almost every species, grow (or decline) in proportion to their current abundance. Changes in abundance are proportional because it takes parents to produce offspring. For example, under relatively good conditions, a monarch butterfly in one winter might have two great-great-grand-daughters that survive migration to the next winter. Under relatively bad conditions, a monarch butterfly in one winter might have only 0.5 great-great-grand-daughters that survive to migration to the next winter. If you start with 10,000 butterflies and have a good year, you end up with 10,000 x 2 = 20,000 butterflies. If the next year is a bad year, you end up with 20,000 x 0.5 = 10,000 butterflies.
Over many years, the multiplicative average (known as the “geometric mean”) yearly rate of change predicts the change in total abundance. On this kind of multiplicative scale, it becomes clear that the low Thanksgiving Counts in 2018, 2019 and 2020 were really low, compared to other years, whereas the recent high numbers are fairly average compared to the years from 2000-2017. And, since the first year of Western Monarch Count in 1997, the geometric mean rate of change has been a yearly multiplier of 0.95, in other words, a 5% decline per year.
As we monitor more sites, it becomes harder to compare old and new data
The estimate of 5% decline per year does not take into account the fact that we are surveying more overwintering sites now than we were in the 1990’s. In the 1990s and 2000s, volunteers monitored ~100 sites each year. In the past ten years, the number of sites monitored has increased from ~100 to 250+. This change in survey effort probably means that the real rate of decline is a bit more than 5%.
However, not all sites are equally suitable for monarchs. Many of the largest sites have been consistently monitored for many years. Many of the newer sites we are now monitoring — thanks to greater volunteer effort — have few or no butterflies. We are also counting newly recognized sites such as those on private property where butterflies have either showed up for the first time or had never been observed previously simply because no one was looking for them before. Therefore, correcting our estimate of the yearly rate of change for survey effort is a complicated statistical analysis. If you are curious about the method, we published this kind of analysis for western monarchs from 1981-2016 in a scientific journal.
It will take many more years before we know if the population increase is a blip, or a trend
Given two relatively good years in a row, it’s tempting to think that the rate of change in monarch populations might have suddenly switched from decreases to increases. However, this kind of change is hard to detect for insects like monarchs, because for butterflies — like many insects — the yearly rate of change varies a lot from year to year. When something varies a lot, it takes a while to know the overall trend.
Think about driving down a highway — if you see two green lights in a row, do you assume they are all green and keep going through the next ones without looking? Probably not. If you see 10 or 20 green lights in a row, though, you might start to think that the lights are timed to match the traffic. And if you drive the same route every day and always encounter only green lights, you might have enough experience to know they are all likely to be green. In a similar way, it will take more years to know if this is the beginning of a positive trend or just a blip in the usual bounciness of butterflies.
Conclusion: For a stable western monarch population, we need numbers to return to the low millions
One approach is to ask, given variation in population size, how large would the western monarch butterfly population need to be to avoid crashes to very low numbers. Given past variation in abundance, we would need about 1 million monarch butterflies on average (by which I mean the geometric mean) for at least ten years to know we have a population that is unlikely to get below 50,000 butterflies. For reference, there were once low millions of monarch butterflies overwintering on the California coast in the 1980’s, and this is the kind of abundance I would like to see in the future.
So, we know the population has increased again this year, and that is great news! We don’t know if it is a one-time pair of green lights, or a new trend toward a population that is increasing, at least on average. But we do know that the population is still much smaller than it was in the 1980’s and 1990’s. In short, it is a time to celebrate the higher counts this year, and hope that these increases continue. But, of course, we can do more than just hope. We can act to protect and restore monarch butterfly populations and habitat in the West, in order to make future increase as likely as possible.