Bugmap — home

How we know

Here’s how we build every estimate, bug by bug, and where we’re least sure of them.

A typical year, not this year

Every page shows a typical year, built from climate averages for 1991–2020. It isn’t a live count or a forecast. A warm spring starts a season early; a dry summer can thin it out. Use it to know what to expect, not what’s happening outside today.

Two kinds of estimate

How many. Mosquitoes are the one bug we estimate by number, because field traps count them at sites across the country and the estimate is fit to those counts. For mosquitoes, darker means more, and one city can fairly be called worse than another.

When. For every other bug we estimate the season: the months it’s out and when it peaks. Darker means further into the season, not more insects. Ticks are sampled in the field too, and their seasons are fit to that sampling, but nothing we tried could predict how many ticks a place has, so we don’t say.

Each bug is measured against itself. The darkest square for one bug isn’t the same number of insects as the darkest for another, which is why no page ranks one bug against the next. And we say how busy bugs are, never how likely you are to catch anything from them. For health questions, see the CDC on mosquitoes and on ticks.

Bug by bug, and how sure we are

Every bug carries one of three labels. Well checked means fit to field sampling and tested against places the estimate never saw. Partly checked means its timing matches independent records but its numbers don’t. Least certain means it rests on published biology and nothing has tested it.

Mosquitoes

A city’s mosquito reading is the worst of three kinds.

All mosquitoes, out in the country (well checked). Every kind together, in the parks, woods and wetlands around a city. Fit directly to carbon-dioxide trap counts from 39 field sites run by the National Ecological Observatory Network, using climate and land cover, and tested by predicting each site with that site left out. Its wettest places, like the Everglades and the Okefenokee Swamp, are wetter than any site it learned from, so their readings are the least certain. One common western mosquito, Culex tarsalis, has no estimate of its own yet; it’s counted here.

House mosquitoes (partly checked). The northern and southern house mosquito (Culex pipiens and Culex quinquefasciatus). We apply laboratory measurements of how their breeding and biting change with temperature (Shocket and others, 2020) to each city’s climate, hour by hour. Their seasons line up with trap counts from the field sites, but those sites are in wild country where house mosquitoes are rare, so how many there are in a city hasn’t been checked. Rainfall and standing water aren’t in this estimate, only temperature.

Asian tiger mosquito (least certain). Aedes albopictus, from laboratory temperature studies (Mordecai and others, 2017), shown only where it’s been recorded or is likely to live (Monaghan and others, 2019). Those county records run through 2016–17, so its recent spread north is missing.

Ticks

Blacklegged (deer) ticks and lone star ticks (well checked, for timing). Where each is found comes from county surveillance records in the CDC’s ArboNET Tick Module. The season is fit to tick-drag sampling at the same 39 field sites, and tested by leaving each site out: the peak lands within a month of the sampling at more than eight sites in ten. How many ticks a place has is another matter. Nothing we tried predicted it, so we treat every place in range alike and show the season only. One known miss: in the East, adult blacklegged ticks are also out in autumn and early spring, and the estimate doesn’t show it. Ticks keep no particular hours; they wait on vegetation whenever it’s warm enough.

Other ticks, including the American dog tick and the Rocky Mountain wood tick, aren’t estimated yet.

Fire ants

Imported fire ants (least certain). Found where the US Department of Agriculture’s federal fire ant quarantine is in force. Their reading is the share of the day warm enough for workers to forage, from field studies of foraging and soil temperature (Porter and Tschinkel, 1987; Vogt and others, 2003). It says when they’re active, not how many mounds there are. Nothing independent has tested it.

Flies and midges

Horse flies and deer flies (partly checked): a season from temperature and land cover, checked against public observation records once those are corrected for when people are outdoors looking. No-see-ums (least certain) rest on published biology; the observation records for them turned out to be unusable.

Black flies (least certain) breed only in running water. Their season comes from temperature, once a year in spring and early summer; where they live comes from a national map of permanent streams, counted more where the land is hilly enough for the water to run fast. An earlier version used ponds and marsh instead, called them scarce where they’re notorious, and was taken down until this replaced it.

Wasps and hornets, chiggers and cockroaches

All three are least certain: a season drawn from the temperatures each is known to need, a habitat weighting from land cover, and a climate limit where one applies. Wasps and hornets share one estimate because yellowjackets, paper wasps and hornets all live the same year: a queen starts a nest in spring, it grows all summer, and frost ends it. Hornet records do follow that shape, scarce until August and highest in September and October. The cockroaches are the outdoor kinds, American and smokybrown, which need mild winters and hot summers; this says nothing about roaches living indoors. Nobody samples any of the three systematically, and public observation records couldn’t check them: wasps get noticed around houses in autumn, roaches get photographed indoors in winter, and almost nobody photographs a chigger.

Stink bugs

Least certain. What’s drawn is the brown marmorated stink bug’s house season: they look for shelter as days shorten, move onto and into buildings on warm afternoons once nights turn cool, and leave again in spring. The thresholds are round numbers matched to that described calendar. Public records of them peak in the same month as the estimate in each part of the country, but those records also carry when people are out looking, so they can’t check it properly.

Fireflies

Partly checked. Flashing fireflies come out after a set amount of warmth has built up since winter: the peak is near 800 degree-days above 10 °C, from a 12-year field study in Michigan. That puts the peak in May along the Gulf, June across the middle of the country and July in the north, which is what public records of the big dipper firefly show once they’re corrected for when people are out looking.

Where stink bugs and fireflies live

Neither has an official range. The stink bug’s comes from public observation records on GBIF: its share of all insects recorded in every one-degree square, against its share nationally. Dividing by all insects is what stops this from being a map of where naturalists live. Only records released for any use, with credit, are counted, and a square needs more than a stray sighting or two.

Fireflies are drawn by summer rain instead. Flashing fireflies need damp ground, which is why they thin out west of the Plains, and there are too few freely licensed records to map them directly. Of the squares where flashing fireflies are on record, 97% get enough summer rain to be in range here.

Where a city begins and ends

A city here is its whole metro area, as the Census Bureau draws it. House and tiger mosquitoes are read over the built-up part of that area; ticks and countryside mosquitoes over the rest. The time of day comes from each city’s daily temperature cycle, each bug’s known active hours, and its sunrise and sunset. A state page is its metro areas side by side, not the whole state.

The map is drawn from an estimate made about every five miles across the country, smoothed into bands. A city’s own page averages its whole metro area, so a spot on the map can read a little differently from its city, and a spot far from any city has had less to check it against. Tap any spot on a zoomed-in map and you get that spot’s year, from the same estimate the map is drawn with.

A park is its whole boundary, as the National Park Service draws it, and all of it counts as country: a park page leaves out the bugs of houses and yards. Where a park climbs 3,000 feet or more, its page also reads each bug band by band up the slope, from the estimate before it’s smoothed, so a valley floor and the ridge above it keep their difference.

When we’re wrong

You know your own back yard better than a climate record does. Where an estimate is plainly wrong for a place you know, that’s worth more to us than another data source, and when one turns out to be wrong we fix it or take it down, as we did with black flies until we could do them properly. Tell us on the About page.

Sources

  • Climate: PRISM Climate Group, Oregon State University. 1991–2020 monthly normals, about 4 km.
  • Field sampling: National Ecological Observatory Network (NEON), operated by Battelle and funded by the National Science Foundation. Mosquitoes sampled from CO2 traps (DP1.10043.001) and ticks sampled using drag cloths (DP1.10093.001), RELEASE-2026.
  • Tick ranges: ArboNET Tick Module, CDC (2025 county surveillance).
  • Fire ant range: USDA APHIS federal imported fire ant quarantine.
  • Mosquito temperature responses: Shocket and others (2020); Mordecai and others (2017).
  • Tiger mosquito range: Monaghan and others (2019), county records and modeled presence.
  • Fire ant foraging: Porter and Tschinkel (1987); Vogt and others (2003).
  • Observation records for fly seasons: Global Biodiversity Information Facility (GBIF).
  • Land cover: Annual National Land Cover Database 2025, USGS.
  • Elevation: PRISM Climate Group 4 km digital elevation model, from the USGS National Elevation Dataset.
  • Streams: HydroRIVERS (Lehner and Grill 2013). Bugmap incorporates data from the HydroSHEDS version 1 database which is © World Wildlife Fund, Inc. (2006-2022) and has been used herein under license. WWF has not evaluated the data as altered and incorporated within Bugmap, and therefore gives no warranty regarding its accuracy, completeness, currency or suitability for any particular purpose. Portions of the HydroSHEDS v1 database incorporate data which are the intellectual property rights of © USGS (2006-2008), NASA (2000-2005), ESRI (1992-1998), CIAT (2004-2006), UNEP-WCMC (1993), WWF (2004), Commonwealth of Australia (2007), and Her Royal Majesty and the British Crown and are used under license. The HydroSHEDS v1 database and more information are available at hydrosheds.org.
  • Cities: U.S. Census Bureau metropolitan areas (2024), population estimates (Vintage 2025), 2020 urban areas and state boundaries.
  • Parks: National Park Service, Land Resources Division unit boundaries.
  • Park photos: Wikimedia Commons, public domain or Creative Commons, each credited on its park’s page.