Data Sources, Organizations & Sponsors
Every field guide stands on the shoulders of the people and institutions who build and maintain the datasets it draws from. Your Ecological Address assembles data from 20+ scientific databases, citizen science platforms, research networks, and AI services into a single ecological profile. The curated places catalog draws on six major ecological research networks spanning 1,000+ field stations, reserves, and protected areas across six continents. None of this would exist without the extraordinary generosity of the organizations below — most of whom make their data freely available for research and education.
This page is our way of saying thank you.
Terrain & Geography
Mapbox provides the visual and spatial backbone of the field guide. Their Terrain-RGB tiles deliver global elevation data at sub-meter precision, which we parse for slope, aspect, and topographic position. Mapbox GL JS renders the interactive map. Their geocoding and Search Box APIs convert place names to coordinates and identify nearby points of interest — parks, natural areas, and landmarks that give a location its human identity. Mapbox has been extraordinarily generous with their developer platform.
Geology
Macrostrat is a platform for geological data exploration built by the Macrostrat Lab at UW–Madison. Their API provides global bedrock geology at remarkable detail — not just “sedimentary rock” but the specific formation, its lithological composition, depositional environment, and age in millions of years. When the field guide tells you that you’re standing on Miocene Columbia River Basalt, that story comes from Macrostrat.
Climate & Weather
Open-Meteo provides the climate backbone of the field guide. Their API delivers both real-time modeled conditions and the historical archive that powers our 30-year normals, trend analysis, and monthly water balance calculations. The archive data draws on ECMWF’s ERA5 and ERA5-Land reanalysis — the gold standard for gridded climate reconstruction. Open-Meteo has been an exceptional partner, generously donating 60 days of customer API access to support this project during development.
The ERA5 reanalysis produced by ECMWF combines decades of satellite, radiosonde, surface, and ocean observations into a consistent global climate record. It is the foundation for the 30-year normals, precipitation patterns, and evapotranspiration calculations in the Climate History and Water Balance cards. We access ERA5 data through the Open-Meteo Historical API.
Weather Underground’s global network of personal weather stations provides hyperlocal ground-truth conditions — the actual temperature at a real station near you, not a model estimate. The field guide queries nearby stations, validates their readings against the Open-Meteo modeled baseline, and rejects malfunctioning stations automatically. When you see “observed at” in the Current Conditions card, that’s a WU station talking.
This lightweight API returns the Köppen climate classification for any coordinate on Earth. Köppen zones — Csb (warm-summer Mediterranean), Dfb (warm-summer humid continental), BWk (cold desert) — are the ecologist’s shorthand for “what kind of place is this?” A single classification code tells you more about what can live at a location than pages of weather data.
Ecoregions & Biogeography
The EPA’s ecoregion framework, originally developed by James Omernik, divides the United States into ecologically meaningful regions based on geology, landforms, soils, vegetation, climate, and land use. Level III provides broad landscape context; Level IV reveals the local ecological character. When the field guide identifies your location as “Willamette Valley Foothills” within the “Willamette Valley” ecoregion, that hierarchical precision comes from EPA.
RESOLVE’s Ecoregions 2017 provides a global classification of terrestrial biodiversity — 846 ecoregions nested within 14 biomes and 8 biogeographic realms. This is the dataset that works everywhere on Earth, telling you whether you’re in Temperate Broadleaf Forest of the Palearctic or Tropical Moist Forest of the Neotropics. Served through UNEP-WCMC’s ArcGIS service.
Land Cover & Vegetation
NLCD is the authoritative land cover dataset for the United States — classifying every 30-meter pixel as forest, grassland, cropland, developed land, wetland, water, or barren ground. It tells you what is actually on the ground right now, as opposed to what could be there under natural conditions. The difference between NLCD and ecoregion is the difference between reality and potential.
This global dataset derives land cover from ESA’s Sentinel-2 satellite imagery at 10-meter resolution — three times finer than NLCD. It provides land cover classification for locations outside the United States where NLCD is unavailable, ensuring the field guide can characterize land surface anywhere on Earth.
LANDFIRE’s Existing Vegetation Type layer goes far beyond generic land cover. Where NLCD says “forest,” LANDFIRE tells you which forest — “North Pacific Dry Douglas-fir Forest and Woodland” or “Columbia Plateau Steppe and Grassland.” These classifications follow the NatureServe Ecological Systems framework, connecting satellite-derived vegetation maps to the ecological literature. The field guide snaps coordinates to LANDFIRE’s 30-meter grid to ensure consistent results.
NatureServe’s Ecological Systems framework is the scientific classification behind LANDFIRE’s vegetation types. Their conservation status rankings (G1–G5 global, S1–S5 subnational) provide a parallel assessment to the IUCN Red List, identifying imperiled species and ecosystems that may not carry formal IUCN designations. When the Conservation Status card shows “NatureServe Imperiled,” it draws from this system.
Biodiversity & Species
iNaturalist is the foundation of the field guide’s biodiversity data. Their API provides research-grade species observations for plants, fungi, birds, mammals, reptiles, amphibians, insects, arachnids, and mollusks — a breadth of taxonomic coverage no single specialist database can match. The field guide queries iNaturalist for total species counts, seasonal activity patterns (month-filtered observations), and conservation-listed species with IUCN and NatureServe status. Dynamic radius widening ensures meaningful results even in sparsely observed areas. iNaturalist represents one of the most remarkable achievements in citizen science: millions of people worldwide documenting biodiversity, one observation at a time.
eBird is the world’s largest biodiversity citizen science platform, with over 100 million observations submitted annually. The field guide queries eBird for recent observations (last 30 days) and notable/rare species (last 14 days) near any location. eBird data fills the “what’s been seen here lately” gap that static databases can’t address — it tells you not just what species could be here, but what birders have actually confirmed in the field this month.
BirdWeather operates a global network of continuously-recording acoustic monitoring stations equipped with BirdNET AI (developed at the Cornell Lab of Ornithology and Chemnitz University of Technology). These stations identify bird species by their vocalizations around the clock, capturing nocturnal and crepuscular species, dawn chorus composition, and birds calling from dense cover that visual observation would miss. The field guide queries BirdWeather’s GraphQL API for nearby station detections — a genuinely new kind of ecological data that didn’t exist five years ago.
IUCN Red List assessments — Critically Endangered, Endangered, Vulnerable, Near Threatened — are the global standard for species conservation status. The field guide accesses IUCN status through iNaturalist’s built-in conservation data layer, which annotates species observations with their formal conservation designations. The presence of IUCN-listed species at a location is one of the strongest signals of ecological value.
Land Protection
PAD-US is the authoritative inventory of protected areas in the United States, including national parks, wilderness areas, wildlife refuges, state parks, conservation easements, and private conservation lands. The field guide queries the PAD-US ArcGIS FeatureServer for any US coordinate, returning all overlapping protected area designations with their GAP biodiversity management status (1–4), IUCN category, managing agency, ownership, and public access level. PAD-US is public domain data maintained by the USGS Gap Analysis Project.
The World Database on Protected Areas is the most comprehensive global database of terrestrial and marine protected areas, with information on over 240,000 sites updated monthly. For international coordinates, the field guide queries the UNEP-WCMC ArcGIS FeatureServer for overlapping protected area polygons, returning designation, IUCN category, governance type, management authority, establishment year, and area. Together, PAD-US (US) and WDPA (international) provide the Land Protection card with global coverage — any coordinate on Earth receives its protection status.
Curated Places — Research Networks & Directories
The field guide’s curated places catalog — over 1,000 biological field stations, nature reserves, and research sites worldwide — is built from authoritative directories maintained by these organizations. Each place is enriched with AI-assisted research and linked to its affiliated networks.
OBFS is the professional association for biological field stations worldwide. Their member directory provided the founding dataset for the curated places catalog, with station names, coordinates, institutional affiliations, and website links. OBFS stations represent the core infrastructure of field-based ecological research — places where scientists live and work in direct contact with the landscapes they study.
The UC Natural Reserve System is the world’s largest university-administered reserve system, protecting over 756,000 acres across 42 reserves from coastal tide pools to alpine peaks, from old-growth redwood forest to Mojave Desert. UC NRS reserve descriptions set the standard for the quality and depth of place documentation that the enrichment system aspires to. The reserve system’s commitment to research, education, and public service is a model for field station networks worldwide.
The LTER Network, funded by the National Science Foundation since 1980, supports long-term ecological research at sites representing major US ecosystem types — from Arctic tundra to coral reefs, tallgrass prairie to old-growth forest. LTER sites contribute foundational knowledge about how ecosystems change over decades and centuries. Site coordinates and metadata were imported from lternet.edu.
NEON is a continental-scale observation facility funded by the National Science Foundation, operating 81 field sites with standardized sensor arrays, automated sampling systems, and airborne remote sensing platforms. NEON’s open data portal provides freely available ecological measurements at unprecedented spatial and temporal resolution. Site coordinates and metadata were imported from the NEON Data API.
The Nature Conservancy manages a global network of nature preserves protecting critical habitat and biodiversity. TNC preserve data was harvested from Wikipedia’s catalog of Nature Conservancy preserves, providing coordinates, state locations, and Wikipedia article links for each site. TNC’s preserves represent some of the most ecologically significant private conservation lands in the United States.
ILTER is the global network of long-term ecological research programs, coordinating site-based research across 44 member countries. ILTER site data was imported from the DEIMS-SDR registry, bringing the curated places catalog to global coverage — from Svalbard to Antarctica, the Daintree to the Negev, the Chilean Andes to the Seto Inland Sea.
DEIMS-SDR is the registry for ecological research sites worldwide, maintained by the eLTER research infrastructure. It catalogs over 1,000 sites with coordinates, ecosystem descriptions, measured parameters, and persistent identifiers. The field guide imports site coordinates and metadata from the DEIMS-SDR API and stores each site’s DEIMS UUID for cross-referencing with the broader European ecological research infrastructure.
Reference & Context
Wikipedia’s Geosearch API finds articles geotagged near any coordinate — providing cultural, historical, and ecological context that enriches the AI narratives. The Wikimedia Commons image repository supplies species photographs for bird identification in the eBird drawer. Wikipedia’s coverage is uneven but often surprisingly deep for natural areas, parks, and geological features.
AI & Narrative Intelligence
Anthropic’s Claude language models are the interpretive voice of the field guide. All ecological data from the cards above is assembled into a structured profile and sent to Claude with instructions to interpret relationships, not recite numbers. The Naturalist, Scientist, and Teacher personas each receive the same data but are prompted to read it through different lenses — sensory and story-driven, technical and hypothesis-generating, or wonder-driven and age-appropriate. The teaser blurbs that appear in the Identity panel are generated by Claude Haiku; the full narratives use the same model with extended context. Claude was also the primary development collaborator for the entire system — a genuine thinking partner from architecture through deployment.
A Note on Open Data
This project exists because these organizations chose to make their data accessible. Every API call in the field guide reaches a service that someone built, documented, and maintains — often on limited budgets, often for the public good. The research networks that maintain field station directories — OBFS, LTER, NEON, ILTER — represent decades of institutional commitment to place-based science. Citizen science platforms like iNaturalist and eBird thrive on participation. Open-source weather services like Open-Meteo survive on community support. Government scientific agencies like USGS, EPA, and ECMWF depend on public investment in basic research. The UNEP World Conservation Monitoring Centre maintains the global protected areas database as a public good for conservation.
The field guide is a reader of their work. The credit belongs to them.