The Orchidaceae Research Atlas
Of 31,892 accepted orchid species, 24,503 (76.8 percent) have no indexed paper; six cultivated genera account for 40.6 percent of 26,501 indexed studies.
We tagged 24,890 of the 26,501 indexed studies of the Orchidaceae, the family of Dendrobium, Vanilla and Phalaenopsis, 31,892 accepted species in all, against 46 dimensions shared across domains plus flora-specific extensions, two of them written for this family. The record concentrates on cultivation, trade and the pharmacology of a few medicinal genera, reaches fewer than a quarter of the family's species, and leaves the largest genera almost untouched.
Every species in the family
We load every accepted Orchidaceae species from Kew's World Checklist of Vascular Plants and link each to the indexed literature. A species with no linked paper is recorded as having no indexed paper.
The Orchidaceae hold 31,892 accepted species in Version 16 of the World Checklist of Vascular Plants: epiphytes of tropical canopies, terrestrial orchids of temperate meadows, leafless species that live on fungi, and the handful of genera the horticultural and medicinal trades have made familiar. We link each accepted name, and every name the checklist treats as its synonym, to OpenAlex; tag every study with a panel of domain-expert models whose tags a second model checks; and place each study at its field site rather than its authors' address. The result is a map of what has been studied, where, and what has not.
A count of papers is not a measure of knowledge, and "no indexed paper" means that no record in OpenAlex matched the species or one of its synonyms. Grey literature, herbarium notes and regional journals outside OpenAlex are not seen. What the count does give is a reproducible floor, species by species, for where the indexed record is silent.
Concentration of research on cultivated genera
Six genera that are grown, eaten or sold account for two fifths of the family's indexed studies (10,752 of 25,919). Dendrobium alone carries 4,696.
Orchids are one of the two largest families of flowering plants, but the indexed research reads like a catalogue of a nursery and a pharmacy. Dendrobium (4,696 studies, 17.7 percent of the corpus) is followed by Gastrodia (1,969), Vanilla (1,303), Cymbidium (1,099), Bletilla (1005) and Phalaenopsis (874). The ten most-studied species together appear in 7,949 studies, 30.0 percent of the corpus; the three largest are Gastrodia elata (1,828 studies), Dendrobium officinale (1,559) and Vanilla planifolia (1,124), two of them medicinal herbs of East Asia and one the source of vanilla.
Read by research focus, the record is first taxonomic and then horticultural. Taxonomy and systematics lead (4,529 studies), propagation and regeneration follow (3,743), and phytochemistry (3,091) sits level with ecology (3,015). The ornamental and horticultural trade is named as a socio-economic use in 1,517 studies, the largest socio-economic use recorded in the family; the next, food and nutrition (410 studies, 190 of them on Vanilla or Gastrodia), is a quarter of its size.
Propagation research by pathway
An orchid seed has no food reserve and cannot germinate on its own in nature. How a study gets a plant from that seed is the central question of orchid propagation, so this atlas tags the route, not just the topic.
Orchid seed is dust-like and carries no endosperm. In the wild the embryo must be colonised by a compatible fungus, which feeds it until it can photosynthesise; the swollen intermediate stage is called a protocorm. In the laboratory the fungus can be replaced by a sugar medium (asymbiotic germination), or supplied deliberately (symbiotic germination), or bypassed altogether by regenerating plantlets from tissue through a protocorm-like body (PLB), a callus, or directly as shoots. The existing flora codebook could say a paper was about propagation; it could not say which route. The new propagation pathway dimension does, and it fired on 2,809 studies.
Asymbiotic germination outnumbers symbiotic germination about three to one (1,278 studies against 409). That ratio is a choice the literature has made: the sugar medium is faster and needs no fungal isolate, but a plant raised without its fungus may not establish when it is returned to the wild, where the fungus is still required. The symbiotic route is the one that conservation reintroduction depends on, and it is the minority method. More broadly, 55.8 percent of the 17,561 studies tagged for spatial scale were carried out in a laboratory, greenhouse or collection rather than in the field.
Pharmacology research by genus and evidence stage
5,571 studies record a bioactivity, a compound class or a therapeutic indication. Seven in ten of them concern Dendrobium, Gastrodia, Bletilla or Anoectochilus.
The medicinal literature of the orchids is East Asian and narrow. Of the 5,571 studies that carry a bioactivity, compound-class or therapeutic-indication tag, 3,946 (70.8 percent) are linked to one of four genera used in Chinese materia medica, and Dendrobium alone accounts for 2,032 (36.5 percent). The chemistry studied is correspondingly particular: polysaccharides (1,333 studies) and flavonoids and phenolics (1,340) lead, far ahead of terpenoids (447) and alkaloids (399). Antioxidant activity (978 studies) is the most-tested property, followed by anti-inflammatory (634), anticancer or cytotoxic (513) and neuroactive (466) effects; 321 of the 410 neuroactive studies concern Gastrodia.
Grouped by how far each study got toward evidence in humans, the counts fall sharply at every stage. Each number counts distinct studies at that stage among the 4,500 studies tagged with an evidence stage, so one extract may appear at more than one. 1,725 studies are at the cell- or enzyme-assay stage; 86 report human clinical evidence.
Pollination strategy and fungal dependence
Two things define orchid biology: flowers that often reward no one, and a lifelong dependence on fungi. The atlas tags both mechanisms, and both are small corners of the record.
Pollination is studied in 1,138 of the 25,919 indexed studies, and 748 of those state the mechanism. The new pollination strategy dimension separates flowers that pay their pollinator from flowers that do not: 465 studies record a deceptive strategy (food or generalised deception 334, sexual deception 153) and 315 record a reward (nectar 233, a non-nectar reward such as oil or fragrance 85). Deception, the trait the family is known for, is the better-studied half of this small literature.
The fungal side is larger but still a minority. 1,463 studies carry at least one tag for fungal dependence: a mycorrhizal symbiont (950), a mycoheterotroph that takes its carbon from the fungus rather than from light (665), or symbiotic seed germination (409). Symbiosis and endophytes together are the most-studied species interaction in the family (1,929 studies), ahead of pollination. Set against a corpus of 25,919, however, fewer than six studies in a hundred engage the mycorrhizal relationship on which every orchid depends at germination.
Species with no indexed paper
24,503 of the 31,892 accepted species, 76.8 percent, have no indexed paper. The largest genera are the least covered.
Behind the medicinal and horticultural genera, 24,503 accepted species, roughly three in every four, have no indexed paper at all. Only 7,389 species have one or more, and the distribution among them is steep: 4,600 of those species (62.3 percent) have exactly one paper, and 454 have ten or more. The largest genera are the emptiest. Bulbophyllum has 2,220 accepted species, 82 percent of them with no indexed paper; Epidendrum, 1,901 species, 88 percent; Stelis, 1,342 species, 93 percent; Lepanthes, 1,233 species, 88 percent. Even Dendrobium, the most-studied genus in the family, has 1,622 accepted species and 74 percent of them have no indexed paper: 2,930 of its 4,696 studies (62.7 percent) are linked to its five most-studied species.
These are candidate gaps, not proof of neglect. Many of these species were described recently, from one or a few collections, in revisions that OpenAlex may index under the genus rather than the species. But the pattern holds across the whole family and across every large genus, and it is the pattern a conservation assessor or a propagation laboratory would meet first.
Research effort by IUCN category
Only 2,237 of the 31,892 accepted species carry an IUCN Red List assessment. Among the 1,127 assessed as threatened, 69.0 percent have no indexed paper.
The whole family is listed under CITES: Orchidaceae as a family sits in Appendix II, and some genera, Paphiopedilum among them, in Appendix I. The Red List is far behind the trade convention. 29,655 accepted species, 93 percent of the family, have no IUCN assessment, and 77.6 percent of those have no indexed paper. Among the assessed species the share with no paper is lowest for Least Concern (58.9 percent of 710 species) and Near Threatened (59.8 percent of 122), rises through Critically Endangered (67.2 percent of 293), Vulnerable (68.9 percent of 280) and Endangered (70.0 percent of 554), and is highest for Data Deficient (83.5 percent of 278). Part of this is structural: species are often listed as threatened because they are rare, narrowly ranged or recently described, traits that also reduce how often they are studied. The relationship is a correlation in the indexed record, not a cause.
Trade regulation itself is rarely the subject of the research. CITES is named in 47 of the 25,919 indexed studies, wildlife or biological trade and trafficking is tagged in 114, and overexploitation or harvesting is recorded as a threat in 607. For a family whose every species is trade-listed, the listing is a background fact of the literature rather than a frequent subject of it.
Study-site countries
Study sites cluster in China, India and Brazil. The species with no indexed paper are concentrated in Indonesia, the northern Andes and New Guinea.
For where research sits against species ranges, coverage cells and access distances, see section 9; for the same sites set against environmental change and split by decade, section 10.
The map above is fixed. The interactive version plots the same study sites and lets you filter them by this atlas's own tags, country, year and citations, then read the papers behind any cluster.
Open the interactive map →China leads with 903 studies that name a study site inside it, followed by India (570), Brazil (420), Indonesia (342), Mexico (321) and the United States (284). China's located research is led by taxonomy and molecular work; Brazil's and Mexico's by taxonomy and field survey. Located research concentrates where the family is diverse and where there are institutions to study it in the field.
The species with no indexed paper are elsewhere
Map instead where the species with no indexed paper occur, using each species' native range countries from the same WCVP distribution table, and the picture changes. Indonesia has 6,019 accepted orchid species in its native flora and 5,044 of them (84 percent) have no indexed paper; Ecuador, 3,660 of 4,462 (82 percent); Colombia, 3,649 of 4,578 (80 percent); Papua New Guinea, 2,700 of 2,993 (90 percent). Brazil (1,973 species with no paper) and Peru (1,972) follow. China, first for located research, has 1,642 accepted species in its native flora and 447 of them (27 percent) with no indexed paper; it is not among the ten countries with the most such species.
The two distributions differ. Located research is concentrated in China, India and Brazil; the species with no indexed paper are concentrated in the Malay Archipelago, New Guinea and the northern Andes, the centres of orchid diversity.
Geographic distribution of study sites
Of the 31,777 species with known range countries, 27,230 (85.7%) have no locatable study site inside any of them; the sites that exist cluster in a few cells and close to cities.
1 · Species ranges against study sites
For each of the 31,892 accepted species we compared two country lists: where it is known to occur, and where its indexed papers were carried out. Known range countries come from the Kew World Checklist of Vascular Plants distribution table (TDWG level-3 areas mapped to countries, native and introduced), so a country counts as range wherever the checklist records the species; in-range is a country-level test, generous within a large country and conservative for a species studied only just across a border. Of the 31,777 species with range data, 27,230 (85.7%) have no locatable study site inside any range country. That figure mixes two things that only a paper-level check can separate: species with no research (24,503 of 31,892 have no indexed research) and species whose papers name no place a geoparser can locate (2,548 of the 7,389 studied species). A further 293 species were studied only outside their known range countries.
Range shortfall by country
Each country is shaded by the share of its range species that have no locatable study site inside it: darker means a larger share of the species known to occur there have not been studied in the field there. Unshaded land has no census species in range. The five countries with the most range species lacking a site are labelled (Indonesia, Colombia, Ecuador, Papua New Guinea, Brazil); they also head the table below.
| Country | Range species | With an in-country study site | Shortfall | Study sites in country |
|---|---|---|---|---|
| Indonesia | 6,021 | 346 | 94% | 259 |
| Colombia | 4,580 | 273 | 94% | 96 |
| Ecuador | 4,465 | 234 | 95% | 60 |
| Papua New Guinea | 2,998 | 25 | 99% | 9 |
| Brazil | 2,940 | 492 | 83% | 260 |
| Peru | 2,599 | 160 | 94% | 52 |
| Malaysia | 2,155 | 93 | 96% | 50 |
| Brunei | 1,733 | 3 | 100% | 2 |
| Venezuela | 1,793 | 69 | 96% | 31 |
| Costa Rica | 1,814 | 209 | 89% | 43 |
2 · Where the study sites are
Each cell is one hexagon of the H3 grid, drawn on its true boundary so that the cells tile without overlap; darker cells hold more distinct study sites. Most cells hold one site; the fieldwork of the whole corpus concentrates in a small number of them.
The densest window, enlarged
The dashed rectangle on the map above marks the window that holds the most study sites (10°W to 44°E, 36°N to 64°N). Here it is at a larger scale and at the grid's own resolution 3 (about 12,400 km2 per cell), with its own log ramp.
Many species, few studies
The ten range countries with the most census species and two or fewer locatable study sites from this corpus. Because the range lists include cultivated and museum records, a country can appear here through garden and herbarium holdings rather than wild populations; read the list as a prompt to check, not a verdict.
| Country | Species in range | Study sites in country |
|---|---|---|
| Brunei | 1,733 | 2 |
| Congo (Kinshasa) | 584 | 1 |
| Guyana | 580 | 2 |
| Nicaragua | 544 | 2 |
| Macao | 523 | 0 |
| Hong Kong | 523 | 1 |
| France | 485 | 1 |
| Suriname | 397 | 1 |
| Zimbabwe | 338 | 1 |
| Solomon Islands | 328 | 2 |
3 · How far the sites are from a city
Distance from each study site to the nearest populated place of 100,000 people or more is a rough measure of how reachable the fieldwork was. The busiest access points, counted greedily so that no site is counted twice, are Pathankot (India, 44 sites within 100 km), Kraków (Poland, 36 sites within 100 km), Yogyakarta (Indonesia, 33 sites within 100 km); 7.8% of all sites lie within 100 km of one of the ten busiest.
These are straight-line distances, not travel times; roads, terrain, sea ice and protected-area access are not modelled. Study-site coordinates are geoparsed from abstracts, and region-level places sit at their centroid.
4 · Method note
Sources. known range countries = Kew WCVP distribution table (TDWG level-3 areas mapped to countries, native and introduced; entity_places role=range, source wcvp / wcvp_introduced; no polygons); access points = Natural Earth ne_10m_populated_places_simple, GitHub master 5.2.0-pre (nvkelso/natural-earth-vector, downloaded 2026-09-29) (populated places with 100,000 people or more), public domain; grid = H3 resolution 3 (h3-js 4.5.0), drawn on the true cell boundaries from h3-js and aggregated one level (h3 cellToParent) on the main maps. A study site is a place a paper's abstract names as the location of the work, resolved to coordinates by a gazetteer and kept only when it came from one of the six trusted geocode sources; 6,677 of 26,501 studies (25.2%) have one. Density and distance use precise sites only (not a whole country, not a rounded centroid): 3,582 distinct places.
Caveat. Study-site coordinates come from automated geoparsing; a 40-site manual check found 31 plausible; a 100 to 200 pin human check is planned. Range coverage is a country-level test, so a study in the right country but far from the species' actual distribution still counts as in range.
Provenance. Generated 2026-10-14 from the published database. Figures in this section are recomputed by phase1/spatial_shortfall.js and checked by the figures audit.
5 · Explore these layers
The coverage cells are a layer on the interactive map, switched on when you open it from here, with the study sites and this atlas's tag filters underneath. A second layer, range shortfall by country, shades each country as the map above does. Any country, ecoregion, administrative region or drawn area can be queried against the whole database in the place explorer, which lists the papers inside it and the topics that are thin there.
Research against change
Three questions of the same study sites: where the environment is changing fastest with the least located research, whether research effort tracks change at all (it does not: the correlation is near zero), and when the located research was published. 64 of 532 ecoregions sit in the high-change, low-research quadrant for forest loss.
In this corpus the rate of environmental change and the density of located research are essentially unrelated: Spearman's rank correlation between the two, unit by unit, is near zero. For forest loss it is −0.120 across 532 ecoregions and −0.007 across 1,846 admin1 regions; for drying it is −0.009 across 237 ecoregions and −0.052 across 750 admin1 regions. These are descriptive figures: they do not show that change causes or deters research, and the units differ in area, accessibility and richness. What they do show is where fast change and little located research coincide, which is the list each block below ranks.
1 · Forest loss 2001 to 2024
Change here is the share of the area that was forest in 2000 (tree cover of 30 percent or more) with a loss event between 2001 and 2024, from Hansen Global Forest Change v1.12 at 30 m. Each ecoregion and each first-level administrative region is placed in one of four quadrants: the top third of the universe by forest loss is high change, the bottom third by located research density (papers of this corpus with a precise study site inside the unit, per 10,000 km2) is low research. The high-change, low-research quadrant is the candidate gap list.
Loss is any stand-replacing event, including fire and the rotational harvest of plantations, and regrowth is not subtracted, so a high value can be a working plantation landscape as easily as a cleared forest; the 30 percent canopy threshold makes open woodland "not forest".
| Ecoregion | Country | Forest lost | Papers with a site | Per 10,000 km2 |
|---|---|---|---|---|
| Sumatran peat swamp forestsTropical & Subtropical Moist Broadleaf Forests | Indonesia | 75.7%91% forested in 2000 | 1 | 0.11 |
| Esperance malleeMediterranean Forests, Woodlands & Scrub | Australia | 69.3%21% forested in 2000 | 0 | 0.00 |
| Drakensberg grasslandsTropical & Subtropical Grasslands, Savannas & Shrublands | South Africa, Lesotho | 68.8%7% forested in 2000 | 0 | 0.00 |
| California montane chaparral and woodlandsMediterranean Forests, Woodlands & Scrub | United States, México | 65.7%30% forested in 2000 | 0 | 0.00 |
| Eastern Guinean forestsTropical & Subtropical Moist Broadleaf Forests | Côte d'Ivoire, Ghana, Togo | 63.5%87% forested in 2000 | 1 | 0.05 |
| Western Guinean lowland forestsTropical & Subtropical Moist Broadleaf Forests | Liberia, Sierra Leone, Côte d'Ivoire, Guinea | 62.0%95% forested in 2000 | 0 | 0.00 |
| Northern Khorat Plateau moist deciduous forestsTropical & Subtropical Moist Broadleaf Forests | Thailand, Laos | 61.3%25% forested in 2000 | 0 | 0.00 |
| Naracoorte woodlandsMediterranean Forests, Woodlands & Scrub | Australia | 59.3%18% forested in 2000 | 0 | 0.00 |
| Peninsular Malaysian peat swamp forestsTropical & Subtropical Moist Broadleaf Forests | Malaysia | 58.0%95% forested in 2000 | 0 | 0.00 |
| Northern Vietnam lowland rain forestsTropical & Subtropical Moist Broadleaf Forests | Vietnam | 57.6%38% forested in 2000 | 0 | 0.00 |
| Region | Country | Forest lost | Papers with a site | Per 10,000 km2 |
|---|---|---|---|---|
| Bình Dương | Vietnam | 96.3%37% forested in 2000 | 0 | 0.00 |
| Bong | Liberia | 95.5%100% forested in 2000 | 0 | 0.00 |
| Aveiro | Portugal | 93.7%62% forested in 2000 | 0 | 0.00 |
| Castelo Branco | Portugal | 91.0%35% forested in 2000 | 0 | 0.00 |
| Porto | Portugal | 90.9%33% forested in 2000 | 0 | 0.00 |
| Montserrado | Liberia | 90.4%88% forested in 2000 | 0 | 0.00 |
| Bomi | Liberia | 89.3%99% forested in 2000 | 0 | 0.00 |
| Melaka | Malaysia | 88.7%83% forested in 2000 | 0 | 0.00 |
| Margibi | Liberia | 87.3%95% forested in 2000 | 0 | 0.00 |
| Leiria | Portugal | 87.0%51% forested in 2000 | 0 | 0.00 |
Forest loss 2001 to 2024, cell by cell
2 · Aridity trend 1990 to 2024
Change here is the linear trend of the annual aridity index (precipitation over potential evapotranspiration, P/PET) from 1990 to 2024, as percent change per decade of the 1990 to 1999 mean, from TerraClimate at about 4.6 km; a negative number means drying. Each ecoregion and each first-level administrative region is placed in one of four quadrants: the top third of the universe by drying is high change, the bottom third by located research density (papers of this corpus with a precise study site inside the unit, per 10,000 km2) is low research. The high-change, low-research quadrant is the candidate gap list.
TerraClimate is an interpolated climate surface, so a 35-year trend at 4 km carries station-density and interpolation error, and a relative trend is unstable where the baseline is near zero; the ranking is restricted to drylands (mean P/PET under 0.65) and excludes hyper-arid units (mean under 0.05) for that reason.
| Ecoregion | Country | Aridity trend | Papers with a site | Per 10,000 km2 |
|---|---|---|---|---|
| Southern Andean steppeMontane Grasslands & Shrublands | Argentina, Chile | −13.7%dry sub-humid | 0 | 0.00 |
| Sudd flooded grasslandsFlooded Grasslands & Savannas | South Sudan, Ethiopia | −10.4%semi-arid | 0 | 0.00 |
| Khangai Mountains alpine meadowMontane Grasslands & Shrublands | Mongolia | −9.8%dry sub-humid | 0 | 0.00 |
| Gobi Lakes Valley desert steppeDeserts & Xeric Shrublands | Mongolia | −9.6%arid | 0 | 0.00 |
| Western Australian Mulga shrublandsDeserts & Xeric Shrublands | Australia | −9.5%arid | 0 | 0.00 |
| Caspian lowland desertDeserts & Xeric Shrublands | Kazakhstan, Russia, Turkmenistan, Iran | −9.5%arid | 0 | 0.00 |
| Knysna-Amatole montane forestsTropical & Subtropical Moist Broadleaf Forests | South Africa | −9.2%dry sub-humid | 0 | 0.00 |
| High MonteMontane Grasslands & Shrublands | Argentina | −9.2%arid | 0 | 0.00 |
| Succulent Karoo xeric shrublandsDeserts & Xeric Shrublands | South Africa | −8.6%arid | 0 | 0.00 |
| Khangai Mountains conifer forestsTemperate Conifer Forests | Mongolia | −8.5%semi-arid | 0 | 0.00 |
| Region | Country | Aridity trend | Papers with a site | Per 10,000 km2 |
|---|---|---|---|---|
| Coquimbo | Chile | −13.7%arid | 0 | 0.00 |
| Mendoza | Argentina | −12.8%semi-arid | 0 | 0.00 |
| Biskra | Algeria | −12.8%arid | 0 | 0.00 |
| San Juan | Argentina | −12.7%semi-arid | 0 | 0.00 |
| Gabès | Tunisia | −12.3%arid | 0 | 0.00 |
| Santiago del Estero | Argentina | −12.3%semi-arid | 0 | 0.00 |
| Govi-Altay | Mongolia | −12.1%arid | 0 | 0.00 |
| San Luis | Argentina | −12.1%semi-arid | 0 | 0.00 |
| Misratah | Libya | −12.1%arid | 0 | 0.00 |
| Gafsa | Tunisia | −12.0%arid | 0 | 0.00 |
Aridity trend 1990 to 2024, cell by cell
3 · When the research happened
The same study sites split by decade of publication. Of the 4,791 located papers, 75.3% were published in the 2010s or 2020s and 55.1% since 2016. The whole corpus of 26,501 papers, located or not, is 73.1% from those two decades, so located research is more recent than the corpus. Of the 500 ecoregions with a located paper, 60 have none published since 2010; of the 948 admin1 regions, 112. 10 located papers have no publication year.
The decade with the most located papers, cell by cell
4 · Method note
Change layers. Forest loss 2001 to 2024 = UMD/hansen/global_forest_change_2024_v1_12 (v1.12 (2024 update)), 2001-2024, sampled at 1,000 m (native 30 m), CC BY 4.0; Aridity trend 1990 to 2024 = IDAHO_EPSCOR/TERRACLIMATE (TerraClimate monthly, EE catalogue copy (1958-01 to 2024-12 at run time)), 1990-2024, sampled at 4,638 m (native 4,638 m), TerraClimate: CC0 1.0 (public domain). MODIS: NASA, no restrictions; dryland classes after UNEP 1992. Values are area means of a systematic sample over each H3 cell of the grid (the maps above draw them one resolution coarser, as valid-area-weighted means of the child cells) and each unit outline (RESOLVE Ecoregions 2017 and GADM 4.1, simplified for display), computed in Google Earth Engine; boundary pixels are weighted fractionally. Layers are remote-sensing and reanalysis products, not field measurements.
Research. 4,791 of 26,501 papers in this corpus have a precise trusted study site (the same selection as the density map above); a paper counts once per cell and once per unit it has a site in, and a unit's density is papers per 10,000 km2 of its area. Only units of 1,000 km2 or more are ranked, and only units in a country where a census species is known to occur, so that a city-state's density is not set against a province's. 38 precise places (open sea, Southern Ocean, atoll centroids) fall in no ecoregion and 18 in no admin1 region; they are in the cell maps but not in the unit tables.
Reading it. A near-zero correlation says that, across these units, more change does not go with more or with less located research. It does not say that change deters research or that research prevents change, and units differ in area, accessibility, richness and national research capacity. A unit in the high-change, low-research quadrant is a candidate gap, not a proven one: papers whose text names no locatable site are invisible here, and the geoparsing caveat of the section above applies.
Provenance. Generated 2026-10-14 from the published database and the Earth Engine cache. Figures in this section are recomputed by phase1/research_vs_change.js, drawn by phase1/gen_atlas_change_section.js and checked by the figures audit.
5 · Explore these layers
These layers are on the interactive map as forest loss 2001 to 2024 and aridity trend 1990 to 2024, cell by cell with the value and the located papers on hover, and a decade selector that limits the study-site dots to one decade of publication.
Expected research Beta
How much located research would each place be expected to carry, given its size, species richness, access, environmental change, biome and national research effort, and where the record falls short of that expectation. Cross-validated pseudo-R2 0.55 for ecoregions and 0.54 for administrative regions. Beta: a first model, stated with its interval and its limits.
Beta This section fits a model of research attention: the number of papers of this corpus with a precise study site inside each ecoregion and each first-level administrative region, predicted from the unit's area, the number of census species whose known range countries include the unit (country level, so it separates countries rather than places within one), its distance to the nearest city of 100,000 people, forest loss and the aridity trend, its biome, realm and latitude band, and how much located research the rest of its country carries. The model predicts where research tends to go. It does not predict ecological importance, threat or need, and a unit with less research than expected is a candidate gap: a place where the record is thinner than in comparable places. It is never proof of neglect. Fit quality is measured out of sample by 5-fold spatial cross-validation (whole realms or whole countries held out): pseudo-R2 0.55 for ecoregions and 0.54 for administrative regions (admin1). Each unit carries a 90 percent prediction interval from the negative-binomial count spread about its expectation. Because located research is strongly clustered (theta under 1), that interval is wide: a unit with no located paper falls below it only when the model expects more than 137.6 papers there (ecoregions) or 197.7 papers there (administrative regions (admin1)), and in this run no unit of either type does. The rankings therefore order units by the tail probability, the chance of a count at or under the observed one given the expectation, and carry the interval label alongside.
1 · Ecoregions: expected against observed Beta
Fit. 774 ecoregions in the universe, 482 with a located paper; negative binomial, dispersion theta 0.54, mean by Poisson quasi-likelihood, ridge lambda 100. 5-fold spatial cross-validation blocked by realm: pseudo-R2 0.55 (0.53 without the national-effort covariate), Spearman rho 0.61 out of fold. Labels against the 90 percent interval: 0 below, 741 within, 33 above.
| Ecoregion | Country | Observed | Expected, 90% interval | Tail probability | Label |
|---|---|---|---|---|---|
| Purus-Madeira moist forestsTropical & Subtropical Moist Broadleaf Forests | Brazil | 0 | 16.2390%: 0 to 61 | 0.156 | within |
| Orissa semi-evergreen forestsTropical & Subtropical Moist Broadleaf Forests | India | 0 | 15.6790%: 0 to 59 | 0.159 | within |
| Northern Vietnam lowland rain forestsTropical & Subtropical Moist Broadleaf Forests | Vietnam | 0 | 15.0890%: 0 to 57 | 0.162 | within |
| Tapajós-Xingu moist forestsTropical & Subtropical Moist Broadleaf Forests | Brazil | 0 | 14.7190%: 0 to 56 | 0.165 | within |
| Solimões-Japurá moist forestsTropical & Subtropical Moist Broadleaf Forests | Colombia, Peru, Brazil | 0 | 12.8390%: 0 to 49 | 0.177 | within |
| California coastal sage and chaparralMediterranean Forests, Woodlands & Scrub | United States, Mexico | 0 | 11.4990%: 0 to 44 | 0.187 | within |
| Sumatran peat swamp forestsTropical & Subtropical Moist Broadleaf Forests | Indonesia | 1 | 25.1890%: 0 to 95 | 0.190 | within |
| California montane chaparral and woodlandsMediterranean Forests, Woodlands & Scrub | United States, Mexico | 0 | 10.7290%: 0 to 41 | 0.194 | within |
| Da Hinggan-Dzhagdy Mountains conifer forestsTemperate Conifer Forests | China, Russia | 0 | 10.2590%: 0 to 39 | 0.198 | within |
| East Central Texas forestsTemperate Broadleaf & Mixed Forests | United States | 0 | 10.0090%: 0 to 38 | 0.201 | within |
| Ecoregion | Country | Observed | Expected, 90% interval | Tail probability | Label |
|---|---|---|---|---|---|
| New Caledonia dry forestsTropical & Subtropical Dry Broadleaf Forests | New Caledonia | 13 | 0.0490%: 0 to 0 | under 0.001 | above |
| Mascarene forestsTropical & Subtropical Moist Broadleaf Forests | Réunion, Mauritius | 13 | 0.1190%: 0 to 1 | under 0.001 | above |
| Puerto Rican moist forestsTropical & Subtropical Moist Broadleaf Forests | Puerto Rico | 20 | 0.4390%: 0 to 2 | under 0.001 | above |
| Comoros forestsTropical & Subtropical Moist Broadleaf Forests | Comoros, Mayotte | 5 | 0.0790%: 0 to 1 | under 0.001 | above |
| Veracruz dry forestsTropical & Subtropical Dry Broadleaf Forests | Mexico | 61 | 3.5690%: 0 to 14 | under 0.001 | above |
The "more than expected" list doubles as a geocoding audit: a small unit with tens of papers against an expectation under one is as likely a place-name collision, a herbarium or a field station as a genuine concentration of fieldwork.
2 · Administrative regions (admin1): expected against observed Beta
Fit. 2,620 administrative regions (admin1) in the universe, 884 with a located paper; negative binomial, dispersion theta 0.50, mean by Poisson quasi-likelihood, ridge lambda 300. 5-fold spatial cross-validation blocked by country: pseudo-R2 0.54 (0.45 without the national-effort covariate), Spearman rho 0.62 out of fold. Labels against the 90 percent interval: 0 below, 2,546 within, 74 above.
| Region | Country | Observed | Expected, 90% interval | Tail probability | Label |
|---|---|---|---|---|---|
| Qinghai | China | 0 | 21.6790%: 0 to 84 | 0.150 | within |
| Chihuahua | México | 1 | 36.0490%: 0 to 139 | 0.174 | within |
| Acre | Brazil | 0 | 11.2790%: 0 to 44 | 0.206 | within |
| Kalimantan Utara | Indonesia | 0 | 10.6590%: 0 to 42 | 0.211 | within |
| Amazonas | Colombia | 0 | 10.2690%: 0 to 40 | 0.215 | within |
| Sonora | México | 1 | 22.4590%: 0 to 87 | 0.219 | within |
| Riau | Indonesia | 1 | 22.0390%: 0 to 85 | 0.221 | within |
| Bolívar | Colombia | 0 | 8.9990%: 0 to 35 | 0.229 | within |
| Ucayali | Peru | 0 | 8.8790%: 0 to 35 | 0.231 | within |
| Lombardia | Italy | 0 | 8.7990%: 0 to 34 | 0.232 | within |
| Region | Country | Observed | Expected, 90% interval | Tail probability | Label |
|---|---|---|---|---|---|
| Sud | New Caledonia | 13 | 0.0490%: 0 to 0 | under 0.001 | above |
| Gotland | Sweden | 12 | 0.3390%: 0 to 2 | under 0.001 | above |
| Kangwŏn-do | North Korea | 7 | 0.1990%: 0 to 1 | under 0.001 | above |
| Andaman and Nicobar | India | 34 | 1.8290%: 0 to 8 | under 0.001 | above |
| Graubünden | Switzerland | 18 | 1.0890%: 0 to 5 | under 0.001 | above |
The "more than expected" list doubles as a geocoding audit: a small unit with tens of papers against an expectation under one is as likely a place-name collision, a herbarium or a field station as a genuine concentration of fieldwork.
3 · Calibration
Out of fold, by decile of expected count (observed / expected): ecoregions: deciles 3 to 8 between 0.58 and 1.29, top decile 0.93; administrative regions (admin1): deciles 3 to 8 between 0.42 and 0.98, top decile 1.12. The middle deciles are the best calibrated; the top decile is under-predicted in every model, because the busiest places (long-term field stations, provinces with a herbarium or a university) exceed what any smooth covariate model gives. That is one more reason the residuals are candidate gaps and candidate artefacts rather than findings.
4 · Method note Beta
Response and exposure. 4,791 of 26,501 papers in this corpus have a precise trusted study site (18.1 percent); the response per unit is the number of those papers with a site inside it (point in polygon, or the nearest unit within 25 km, as the place explorer counts), so a paper with sites in two units counts once in each: 5,706 unit-papers over 774 ecoregions and 5,550 over 2,620 administrative regions (of 847 and 3,686 units in all). The exposure is log(area), with log(area) also entering as a slope so the data set the area elasticity.
Covariates. log_area_km2, log_richness, log_access_km, forest_loss_pct, forest_loss_missing, aridity_trend_pct_per_decade, aridity_trend_missing, log_national_loo_density, national_loo_undefined, biome, realm, lat_band; continuous covariates standardised, categorical levels with fewer than 10 units pooled. Protected-area share is not available (no protected-area table in the database) and is stated as the natural next covariate. Richness is a country-level proxy: every unit of a country is credited with the national list.
Model and validation. Negative binomial (NB2) GLM with a log link and a ridge penalty on the standardised slopes (ecoregions lambda 100, administrative regions (admin1) lambda 300); the mean estimator and lambda are chosen on out-of-fold Poisson deviance over the same spatial folds; the dispersion is estimated by maximum likelihood. 5-fold spatial cross-validation blocked by realm (ecoregions) or country (admin1), never random rows; a held-out realm's own dummy is unseen in training. The 90 percent interval is the 5th and 95th percentile of NB(expected, theta) and carries no parameter uncertainty, so it is if anything too narrow. Zero rule: a unit with no located paper and an expectation under 0.5 is never labelled below and never ranked.
Limits. Located papers only: papers whose text names no locatable site are invisible here, and the geoparsing error rate of the sections above applies. The national-effort covariate carries most of the between-country signal; the fit without it is printed above. Terrestrial units only. This is a Beta layer: the numbers are honest as stated, and they will move when the geocoding, the corpus or the covariates change.
Provenance. Generated 2026-10-14 from the published database (SELECT only) and the change layers' cache. Refresh with node phase1/expected_research.js orchidaceae, redraw with phase1/gen_atlas_expected_section.js; every figure in this section is checked by the figures audit.
5 · Explore this layer
The same units are on the interactive map as the beta layer "Expected vs observed", ecoregions or administrative regions, with observed, expected, interval and label on hover, and a link from each unit to the place explorer. The place explorer shows the same figures for a selected unit under "Beta: expected research".
Summary of the record
Three imbalances in where orchid research has and has not gone.
This atlas is one view of a complete, queryable corpus. A single genus's propagation record, a country's orchid research, or a shortlist of threatened species with no indexed paper to survey is each a query away, on the same reproducible foundation. The next section turns one of these gaps into a ranked list.
Where to look next
The sections above measure what has and has not been studied. This section ranks the species with no indexed paper by whether their genus already has a working propagation protocol, to indicate where ex situ conservation could start.
Propagation protocols transfer within genera: a germination medium or a regeneration route worked out for one Paphiopedilum is the obvious starting point for the next. So we score each genus by the share of its studied species that carry at least one propagation-pathway tag, and list the species in that genus with no indexed paper. A genus with many protocols and many unstudied species is where a conservation laboratory has the most to borrow from. The table keeps genera with at least 15 studied species.
| # | Genus | Studied species with a propagation-pathway study | Species with no indexed paper | Of which assessed as threatened | Examples to start with |
|---|---|---|---|---|---|
| 1 | Cattleya | 61% (50/82) | 161 | 0 | C. acuensis, C. adelinae, C. adrienne |
| 2 | Paphiopedilum | 52% (38/73) | 72 | 17 | P. argus (EN), P. cornuatum (CR), P. gigantifolium (CR) |
| 3 | Vanda | 52% (28/54) | 43 | 1 | V. javierae (EN), V. aliceae, V. amoena |
| 4 | Cyrtopodium | 50% (13/26) | 27 | 0 | C. blanchetii, C. braemii, C. caiapoense |
| 5 | Cymbidium | 42% (30/71) | 40 | 0 | C. acuminatum, C. aliciae, C. ballianum |
| 6 | Cypripedium | 40% (18/45) | 17 | 6 | C. farreri (EN), C. ludlowii (EN), C. margaritaceum (EN) |
Paphiopedilum, the slipper orchids, is the case the ranking surfaces most clearly: the genus is in CITES Appendix I, 38 of its 73 studied species have a propagation study, and 17 of its 72 species with no indexed paper are assessed as threatened. The ranking draws only on the propagation record of each genus's studied relatives and the Red List status of the rest, and surfaces that from those two facts alone.
A starter survey list
Species with no indexed paper drawn from the top genera, threatened species first: a concrete place a propagation laboratory or a taxonomist could begin. Grouped by genus.
Each name is an accepted species with no indexed paper, sitting in a genus whose studied relatives have a propagation record. Whether any protocol transfers to it is determined in the laboratory.