Native Plant Data: Sources and Method

Where each field comes from

The genus versus species problem

This is the single most abused number in native plant writing. You will very often read that a white oak supports 500 or more caterpillar species. The published research does not say that. It reports counts by plant genus, so the defensible statement is that the genus Quercus hosts 436 caterpillar species in the Eastern Temperate Forests ecoregion. Attaching that figure to Quercus albaspecifically turns a genus level result into a species level claim that nobody measured. Every figure we publish names the genus and says the level out loud, and a species whose genus is not on its state's ecoregion sheet simply shows no figure at all rather than borrowing one.

Selection criteria

1. The species appears as a sample species on an NWF keystone sheet, so it belongs to a genus with a published wildlife value rather than being picked on our judgment.
2. USDA PLANTS returns an exact accepted name match. Eight NWF sample names did not resolve, through typography or superseded synonyms, and were dropped rather than guessed at.
3. USDA records the species in the state, and classifies it native to the continental region that state sits in. The test is region aware: Alaska is checked against Native AK, Hawaii against Native HI, everywhere else against Native L48.
4. Anything flagged introduced, or probably introduced, for that region is excluded. That is what removed Hawaii entirely and what removed two species from Alaska.
5. Within a layer, species are ranked by the genus level caterpillar count published for that state's ecoregion, because keystone genera are where the food web concentrates.

What this data cannot tell you

USDA PLANTS publishes nativity at continental scale and distribution at state scale. It does not publish nativity state by state. For most species that distinction does not bite, but it can: black locust, Robinia pseudoacacia, is classified native to the lower 48 and is recorded in 49 states, yet its genuine native range is a much smaller Appalachian and Ozark core and it behaves as an invasive across much of the rest. So a row on our pages means USDA records this species in your state and classifies it native to your region. It does not mean a botanist has confirmed it is indigenous to your particular county. For that, a state native plant society or a county extension office is the right next step, and BONAP county level maps are the standard reference.

Two further limits worth stating plainly. Ecoregion assignment per state is editorial, since most states span several Level I regions and we assign the dominant one. And we publish no photographs with this dataset, because we could not confirm a public domain or CC0 license per image for the species involved, and an unverifiable license is not worth a picture.

How similar are the state pages

We measured it rather than assuming. Across all 1,225 state pairs the mean Jaccard similarity of the verified species sets is 0.514, with a median of 0.541. Neighboring states run far higher: Maryland and Virginia share 0.914, Vermont and New Hampshire 0.912, Arizona and New Mexico 0.873. That is not a template artifact, it is the biology. Keystone genera are continental by definition, so states that sit in the same ecoregion genuinely share most of their high value native plants. We publish that number here rather than hide it, and the per state differences that do exist, the verified species list, the ecoregion the counts come from, the drought subset and the regional context, are the parts worth reading.

The case for planting, and the case against

The argument for is straightforward. North America has lost more than 3 billion breeding birds since 1970 (Rosenberg et al., Science, 2019). Terrestrial birds feed insects to their young, and a single chickadee brood consumes roughly 6,000 to 9,000 caterpillars before fledging. Desiree Narango, Doug Tallamy and Peter Marra found that Carolina chickadees in yards where non native plants exceeded about 30 percent of the woody plant biomass could not sustain their populations. Roughly 70 percent of the ornamental species sold in United States nurseries over two centuries were not native to the continental US, by a 2022 peer reviewed analysis, so the default yard is built largely from plants the local food web cannot use. And only about 13 percent of the United States has protected status, which is the core of Tallamy's argument that the habitat that could exist is mostly on private land.

The argument against is worth taking seriously too. Arthur Shapiro, a butterfly ecologist at UC Davis who has run one of the longest running butterfly monitoring datasets in North America, has described backyard native planting as a gardening project rather than restoration. The objection is about scale and function: a yard is not a functioning ecosystem, it lacks the area, the disturbance regimes and the species interactions that restoration implies, and some species that use gardens are the ones already doing fine. Others point out that some non native plants do provide nectar, and that habitat loss and pesticide use operate at scales no garden touches.

Our position is the modest one. Planting a native canopy tree and a native border is a defensible, low cost thing to do with the edges of a yard, and the food web evidence behind the keystone genera is solid. Calling it restoration is a stretch. We would rather publish both readings and let you decide than sell you the stronger claim.

Sources

• USDA NRCS, PLANTS Database. Nativity, state distribution and plant characteristics. Read 2026-08-18.
• National Wildlife Federation, Keystone Plants by Ecoregion. Genus level caterpillar and specialist bee counts. Downloaded 2026-08-18.
• Narango, D.L., Tallamy, D.W. and Shropshire, K.J. (2020). Few keystone plant genera support the majority of Lepidoptera species. Nature Communications 11, 5751.
• Tallamy, D.W. and Shropshire, K.J. (2009). Ranking Lepidopteran use of native versus introduced plants. Conservation Biology 23, 941 to 947.
• Fowler, J. and Droege, S. (2020). Pollen Specialist Bees of the Eastern United States, and companion compilations for the Central and Western United States.
• Rosenberg, K.V. et al. (2019). Decline of the North American avifauna. Science 366, 120 to 124.
• Commission for Environmental Cooperation, Ecological Regions of North America. Level I ecoregion system.

Common questions