The spiders in question are mygalomorphs, an ancient lineage of heavy-bodied, long-lived burrowing spiders that includes tarantulas and their relatives. Five species of trapdoor spider are known to be endemic to California’s coastal dunes: Aptostichus miwok, Aptostichus ramirezae, Aptostichus simus, Aptostichus stephencolberti, and Cryptocteniza kawtak. The last of these, described by Bond and colleagues in 2020 as an endangered living fossil along the Central Coast, underscores how precarious these relict populations are. Each spider constructs a silk-lined burrow capped with a hinged trapdoor, and each burrow is effectively a census unit: the team assumed one spider per occupied burrow, a reasonable assumption given that these spiders are solitary and territorial for most of their lives.

Counting animals that spend their entire lives underground presents an obvious methodological challenge, and the researchers solved it with a deceptively simple design. At each site, they mapped a long, thin sampling strip of roughly 26,000 square meters — 30 meters wide and hundreds of meters long — and overlaid it with a continuous grid of one-square-meter quadrats. From that grid, they randomly selected 150 plots per site using computer-generated random numbers, eliminating the human tendency to sample where burrows are easiest to find. Surveyors were trained to distinguish occupied burrows, identified by their thick, multilayered, bright white silk that remains elastic when handled, from abandoned burrows, whose silk turns gray to black, becomes fragile, and often loses its door entirely. This standardization matters because mygalomorphs are not the only silk-producers in the dunes, and misidentification would inflate the counts.

The raw numbers are striking. At Sunset State Beach, occupied burrow density came out at 0.287 burrows per square meter — about 29 trapdoor spiders per 100 square meters — which, extrapolated across roughly 121,406 square meters of potential habitat, yields the estimate of 35,206 spiders. Moss Landing was denser still: 0.453 burrows per square meter, or just over 45 spiders per 100 square meters, across approximately 68,797 square meters of habitat north of the jetty road, for an estimated 30,959 individuals. The single densest quadrat, at Moss Landing, contained 17 occupied burrows crammed into one square meter of sand. Remarkably, the mean number of occupied burrows did not differ significantly between the two sites, indicating that overall burrow densities were statistically similar even as the spatial stories behind those densities diverged dramatically.

And diverge they did. At Sunset State Beach, burrow presence was strongly and positively associated with vegetation. A Fisher’s exact test on the combined dataset found a significant relationship between plant presence and burrow presence, with an odds ratio of 6.29 — meaning a quadrat with plants was more than six times as likely to contain a burrow. A negative binomial model, chosen because burrow counts were overdispersed relative to the mean, confirmed that plant presence significantly predicted burrow abundance. Nearest-neighbor analysis drove the point home: quadrats with burrows sat an average of just 0.21 meters from the nearest plant, compared with 1.00 meters for quadrats without burrows, a difference confirmed by a Wilcoxon rank-sum test. The spiders at Sunset, in short, are building their homes in the shadow of dune plants — even though the vegetation itself was randomly distributed across the site.

At Moss Landing, the pattern inverted. No significant vegetation–burrow association was detected by any test: the Fisher’s exact odds ratio was a nonsignificant 1.37, the Spearman correlation between plant and burrow presence was essentially zero, and plant presence failed to predict burrow abundance in the negative binomial model. Yet burrows were still spatially clustered — Moran’s I, a measure of spatial autocorrelation, indicated significant positive clustering of burrow presence at both sites. The crucial difference is that at Moss Landing, the clustering of burrows was decoupled from the clustering of plants. Both burrows and vegetation were aggregated there, but the two aggregations did not overlap in any statistically meaningful way, and burrow-present quadrats were not significantly closer to vegetation than burrow-absent ones. Something other than plants is organizing the Moss Landing population.

The authors propose several ecological mechanisms for this contrast. At Sunset, where a single species, Aptostichus stephencolberti, appears to dominate, the strong vegetation association is consistent with an ideal free distribution, in which individuals preferentially occupy the best microhabitats — in this case, areas where plant roots may physically support burrow walls in loose aeolian sand. Dune plants also concentrate arthropod prey, including the tenebrionid dune beetle Coelus, which the team consistently found beneath plants and whose remains turned up inside trapdoor spider burrows. At Moss Landing, by contrast, the higher burrow density likely reflects a multi-species assemblage in which competition for space and resources, density-dependent processes, limited dispersal, and microclimate constraints shape where spiders settle. Juvenile burrows clustered near adult burrows suggest that spiderlings disperse only short distances from their mothers, a life-history trait that would generate spatial clustering independent of any plant cue.

The timing of the surveys adds an important caveat to the population estimates. The counts were conducted in July and August, precisely when male trapdoor spiders abandon their burrows to wander in search of mates — and when other studies of mygalomorph populations have documented dramatic seasonal swings in burrow density, from as low as 0.01 to 0.19 individuals per square meter within a single summer. Maternal care complicates the picture further: the team found large female burrows where females and up to 200 spiderlings retreated to the deepest chambers when disturbed, and related work on the dune-obligate wolf spider Geolycosa wrightii has shown that females temporarily seal their burrows during brooding, making active burrows harder to detect. The true populations may fluctuate considerably across the year, and the estimates should be read as seasonal snapshots rather than permanent censuses.

What makes the study genuinely consequential is its conservation message: the same species, living in the same kind of habitat, obeys different spatial rules at different sites. At Sunset, where vegetation predicts both burrow occurrence and abundance, trampling, invasive species, or coastal development that strips plant cover could disproportionately destroy habitat suitability — making native dune vegetation restoration a direct investment in spider conservation. At Moss Landing, where burrow density is higher but vegetation is irrelevant to burrow placement, the priority shifts to protecting the physical integrity and contiguity of the dune system itself, preserving soil stability, microclimate, and the spatial structure that density-dependent processes have built. A one-size-fits-all dune management plan, the authors argue, would fail both populations.

The broader stakes are considerable. California’s 3,427 miles of coastline once supported vast, continuous dune systems, but decades of agricultural, residential, and commercial development have fragmented them into disjunct islands — a process of coastal squeeze that traps endemic dune species between the rising Pacific and the hardened terrestrial back beach. Trapdoor spiders are poor dispersers with restricted gene flow, long life spans, and extreme habitat specificity, exactly the profile that pushes habitat specialists from vulnerable to extinction. Yet dune conservation has historically fixated on charismatic umbrella species — snowy plovers, dune plants, butterflies — while burrowing arthropods like Aptostichus stephencolberti have been assumed to benefit passively from protections timed around bird nesting seasons and plant blooming periods. Those closures rarely coincide with the July and August mating wanderings of male spiders, leaving dispersing juveniles exposed when beaches reopen and visitors slide down dune faces, crushing burrows and burying their doors. If these spiders are, as the spatial evidence suggests, ecosystem engineers whose burrows alter soil aeration, moisture retention, and sediment stability — and if shorebirds increasingly depend on dune invertebrates as sea level rise erodes intertidal foraging grounds — then the unseen 66,000 spiders of Sunset and Moss Landing may be holding up far more of the dune ecosystem than their numbers alone imply. Baseline censuses like this one are the first step toward noticing them before they are gone.

Subject of Research: trapdoor spiders

Article Title: The distribution, spatial clustering, and density of trapdoor spiders in two California coastal dune ecosystems

Article References: Keller, M. F., Dickinson, K. A., Jochim, E. E., Quayle, I. L., Campos, J. E., Briggs, H. R., Wiggins, C. R., & Bond, J. E. (2026). The distribution, spatial clustering, and density of trapdoor spiders in two California coastal dune ecosystems. Frontiers in Arachnid Science, 5, Article 1913353. https://doi.org/10.3389/frchs.2026.1913353

Image Credits: AI Generated

DOI: 10.3389/frchs.2026.1913353

Keywords: trapdoor spiders, California coastal dunes, mygalomorph spiders, population census, spatial ecology, conservation, Aptostichus stephencolberti, ecosystem engineers, coastal squeeze, Frontiers in Arachnid Science, distribution, spatial