Methodology & Data Sources

Public Gardens Guide is a reference for botanical gardens and arboretums across the United States. This page explains exactly how the guide is built — the source of the data, what we choose to publish and index, and how we handle facts we are not sure about. The goal is a useful answer with visible assumptions, not a black box.

Where the data comes from

Every garden in this guide is drawn from Wikidata, a structured, openly licensed knowledge base maintained by a global community, released under the CC0 public-domain dedication. We query Wikidata for botanical gardens and arboretums (items typed botanical garden or arboretum) and rebuild the guide from that export: name, coordinates, garden type, resolved city/state where Wikidata records one, and an official website where one is recorded.

The current corpus is 752 gardens. We do not add gardens from unsourced trivia pages, forums, or user submissions.

What we do not fabricate

Admission price, opening hours, season, and APGA (American Public Garden Association) accreditation are not in the current Wikidata export for these gardens. We show them only once we have sourced them from a further pass (OpenStreetMap gap-fill, APGA member reconciliation) — until then, a garden page simply omits that field rather than guessing. If a garden page links to an official website, check that site for current hours and admission.

What we publish, and what we index

Every valid garden record — real name, valid coordinates, a resolved state — is built and reachable on the site, so every internal link (state hubs, nearby-garden rails, search) resolves. But we do not send a thin page to Google just to have more pages indexed. A garden page is added to the sitemap and left indexable only when the record carries genuine per-page substance beyond “name + type + coordinates”: a resolved state plus a website or a known city.

Today that is 305 of 752 gardens. The rest stay live and linked — you can browse to them, and they resolve — but carry a noindex, follow signal and are left out of the sitemap until enrichment gives them enough real substance to earn a place in the index. This is the same index-floor approach we use across our reference sites: quality-first, growing the indexed set only as real data is added, never templated mass.

Accuracy and uncertainty

Garden data drawn from a community-maintained knowledge base is inherently incomplete — not every garden has a resolved city or state, and a listed “website” field can go stale. We show what the source supports and avoid overstating it. If a record looks wrong or a garden is missing, tell us at hello@publicgardensguide.co and we will check it against the source.

How often it refreshes

Because Wikidata is continuously edited, the guide improves as new statements are added and existing ones are corrected upstream. Planned enrichment passes — OpenStreetMap opening-hours and admission tags, APGA accredited-member reconciliation, and a broader Wikidata type sweep beyond botanical-garden/arboretum — will widen both the corpus and the indexed subset over time, rather than staying frozen at launch.

Licensing and attribution

The underlying garden facts come from Wikidata and are dedicated to the public domain under CC0. Map rendering uses OpenStreetMap data. Public Gardens Guide is built and operated by Purcell Analytics. For more on the project, see About.