The Felidae Research Atlas
All 45 accepted cat species carry indexed research, and five of them account for 8,853 of the 18,019 indexed studies, 49.1 percent.
We tagged 17,027 of the 18,019 indexed studies of the Felidae, the family of the lion, the tiger and the domestic cat’s wild relatives, 45 accepted species in all, against 51 dimensions shared across domains plus fauna-specific extensions. Unlike the plant families in this series the cats have no unstudied species: every one of the 45 has a literature. What the record shows instead is concentration. Five species hold about half of it, the six species described or split since 2020 hold 75 studies between them, and the amount of research a species carries bears little relation to how threatened it is.
Every species in the family
We load every accepted Felidae species from the Mammal Diversity Database and link each to the indexed literature. A species with no linked paper is recorded as having no indexed paper.
The Felidae hold 45 accepted species in Version 2.5 of the Mammal Diversity Database: the seven big cats of the genus Panthera and Acinonyx, the lynxes, the small cats of the Andes and of southern Asia, and six species that were raised from subspecies or split from them between 2020 and 2026. 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.
No accepted cat species is without indexed research, so the question this atlas answers is not which species are missing from the record but how unevenly the record is spread across them. The smallest literatures belong to the recent splits, which carry between 3 and 20 studies each: Leopardus fasciatus 3, Leopardus pajeros 8, Leopardus garleppi 14, Prionailurus javanensis 15, Leopardus braccatus 16 and Leopardus pardinoides 20. These are not unstudied animals so much as newly separated ones, whose earlier literature sits under the name they were split from.
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, management reports 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 thin.
Concentration of research on five species
Five species carry 8,853 of the 18,019 studies. The genus Panthera alone accounts for 7,454, or 41.4 percent of the corpus.
The puma, the lion, the tiger, the leopard and the jaguar are the five most studied cats, and together they hold 8,853 studies, 49.1 percent of everything indexed for the family. Puma concolor alone carries 2,448 studies, 13.6 percent of the corpus, more than the 29 least studied species combined. The five are large, wide-ranging, conflict-prone and charismatic, and they attract the funding and the field programmes that generate papers.
The concentration is not only taxonomic. It is also a concentration of question: the literature on the five is dominated by population ecology, human-wildlife conflict and movement, while the small cats of South America and southern Asia are represented mainly by distribution records and occasional genetics. The tag vocabulary in section 8 lets the two halves be compared directly.
Research effort by IUCN category
Threat category does not predict research effort. The median Endangered cat carries 59.5 studies; the median Least Concern cat carries 398.
Thirty-eight of the 45 species carry a Red List category in the reference extract used here. Six are the species described or split since the last assessment and have none, and one, Felis chaus, appears in the extract without a usable category. Among the 38, 18 are assessed as threatened, that is Endangered or Vulnerable; no cat is currently listed as Critically Endangered.
Research effort does not follow the threat ranking. The median number of studies per species is 398 for Least Concern, 244.5 for Vulnerable, 129 for Near Threatened and 59.5 for Endangered. The four Endangered species span the whole range of the record, from Panthera tigris with 2,115 studies to Catopuma badia, the bay cat of Borneo, with 32. What drives the count is body size, range size and proximity to people, not conservation status.
The 18 threatened species carry 8,424 studies between them. That is a substantial literature, but it is distributed as unevenly as the rest: the tiger, the lion and the snow leopard account for most of it, and the Andean cat, the bay cat and the flat-headed cat are each below 100 studies.
Studies recovered from common names
A harvest that searches scientific names alone misses papers that only ever write “Iberian lynx”. A judged common-name pass added 1,047 links on 1,039 studies.
Every corpus in this series is built by searching each accepted scientific name and its checklist synonyms. For plants that is close to complete, because botanical papers name their species in Latin. For mammals it is not: a large part of the applied and regional literature on cats uses the English common name and never the binomial, so a scientific-name harvest silently under-counts exactly the field and management studies a conservation reader wants.
We measured the size of that gap before trying to close it. A probe searched the Mammal Diversity Database and Red List common name of each species, filtered the results to whole-word matches, checked which were already in the corpus and asked an independent model whether each unmatched paper was about the species. It put the paper-weighted recall of the scientific-name harvest at 95.1 percent (interval 91.7 to 97.7) and the species-weighted mean at 88.1 percent. The probe is an upper bound for the best-known cats, because the search interface reported a capped sample size rather than a true hit count for 23 of the 55 names.
The harvest itself then ran over the 41 names that are not dominated by a homonym, judged 2,220 candidate papers one at a time, confirmed 1,116 as being about the species and linked them. The corpus now carries 1,047 such links on 1,039 studies, each one recorded with the name that found it and the judge’s reason.
For the Iberian lynx this is the difference between a partial record and a usable one: 145 of its 492 studies, 29.5 percent, were found only under “Iberian Lynx”. The pattern is strongest for species whose literature is regional and applied, and weakest for the big cats, whose papers almost always give the binomial.
Study-site countries
Study sites cluster in the United States, Brazil and India. Half of the corpus names a field site that can be placed on a map.
For where research sits against species ranges, coverage cells and access distances, see section 6; for the same sites set against environmental change and split by decade, section 7.
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 →The United States leads with 1,443 studies that name a study site inside it, followed by Brazil (775), India (592), Mexico (457), South Africa (397), Spain (283), China (307) and Indonesia (312). The ordering follows the big five: North American pumas and bobcats, Brazilian jaguars and ocelots, Indian and South African big cats. Spain enters the list almost entirely on the Iberian lynx, one of the most intensively monitored carnivores in the world, and Nepal on tiger and snow leopard work.
Geographic distribution of study sites
Of the 45 species with known range countries, 0 (0.0%) 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 45 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 Mammal Diversity Database country distribution (v2.5, the checklist the census uses; countries MDD marks as uncertain are left out), so a country counts as range only where the species is recorded in the wild; 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 45 species with range data, 0 (0.0%) 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 (0 of 45 have no indexed research) and species whose papers name no place a geoparser can locate (0 of the 45 studied species).
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 (Chad, Brunei Darussalam, Guinea, Guinea-Bissau, Armenia); they also head the table below.
| Country | Range species | With an in-country study site | Shortfall | Study sites in country |
|---|---|---|---|---|
| Chad | 7 | 1 | 86% | 3 |
| Brunei | 5 | 0 | 100% | 0 |
| Guinea | 5 | 0 | 100% | 0 |
| Guinea Bissau | 5 | 0 | 100% | 0 |
| Armenia | 6 | 1 | 83% | 2 |
| Guyana | 6 | 1 | 83% | 1 |
| Central African Republic | 7 | 2 | 71% | 2 |
| Djibouti | 4 | 0 | 100% | 0 |
| Eritrea | 4 | 0 | 100% | 0 |
| Sierra Leone | 4 | 0 | 100% | 1 |
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 (125°W to 71°W, 17°N to 45°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. The range lists are wild distributions, so a country here holds wild populations of these species with little located field research; read the list as a prompt to check, not a verdict.
| Country | Species in range | Study sites in country |
|---|---|---|
| Turkmenistan | 7 | 1 |
| Central African Republic | 7 | 2 |
| Cambodia | 7 | 2 |
| Niger | 7 | 2 |
| French Guiana | 6 | 1 |
| Guyana | 6 | 1 |
| Armenia | 6 | 2 |
| Iraq | 6 | 2 |
| Suriname | 6 | 2 |
| Brunei | 5 | 0 |
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 Pokhara (Nepal, 29 sites within 100 km), Jundiaí (Brazil, 26 sites within 100 km), Alajuela (Costa Rica, 24 sites within 100 km); 5.7% 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 = Mammal Diversity Database v2.5 countryDistribution, confirmed countries only (entity_places role=range, source mdd; 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; 8,898 of 18,019 studies (49.4%) have one. Density and distance use precise sites only (not a whole country, not a rounded centroid): 3,719 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. 45 of 472 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 0.024 across 472 ecoregions and −0.003 across 1,750 admin1 regions; for drying it is −0.082 across 203 ecoregions and −0.119 across 754 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 | Range species without a site |
|---|---|---|---|---|---|
| Maranhão Babaçu forestsTropical & Subtropical Dry Broadleaf Forests | Brazil | 64.9%81% forested in 2000 | 2 | 0.14 | 8 |
| Borneo peat swamp forestsTropical & Subtropical Moist Broadleaf Forests | Indonesia, Malaysia, Brunei | 63.9%95% forested in 2000 | 1 | 0.15 | 9 |
| Eastern Guinean forestsTropical & Subtropical Moist Broadleaf Forests | Côte d'Ivoire, Ghana, Togo | 63.5%87% forested in 2000 | 0 | 0.00 | 5 |
| Northern Khorat Plateau moist deciduous forestsTropical & Subtropical Moist Broadleaf Forests | Thailand, Laos | 61.3%25% forested in 2000 | 0 | 0.00 | 9 |
| Northern Vietnam lowland rain forestsTropical & Subtropical Moist Broadleaf Forests | Vietnam | 57.6%38% forested in 2000 | 0 | 0.00 | 5 |
| Khangai Mountains conifer forestsTemperate Conifer Forests | Mongolia | 55.5%9% forested in 2000 | 0 | 0.00 | 4 |
| Southern Vietnam lowland dry forestsTropical & Subtropical Dry Broadleaf Forests | Vietnam | 53.7%53% forested in 2000 | 0 | 0.00 | 5 |
| Guinean mangrovesMangroves | Guinea-Bissau, Sierra Leone, Guinea, Gambia, Liberia, Senegal | 53.2%56% forested in 2000 | 0 | 0.00 | 6 |
| Nyanga-Chimanimani Montane forest-grasslandMontane Grasslands & Shrublands | Zimbabwe, Mozambique | 49.3%35% forested in 2000 | 0 | 0.00 | 6 |
| Northeast Brazil restingasTropical & Subtropical Moist Broadleaf Forests | Brazil | 43.6%36% forested in 2000 | 0 | 0.00 | 10 |
| Region | Country | Forest lost | Papers with a site | Per 10,000 km2 | Range species without a site |
|---|---|---|---|---|---|
| Bình Dương | Vietnam | 96.3%37% forested in 2000 | 0 | 0.00 | 5 |
| Bong | Liberia | 95.5%100% forested in 2000 | 0 | 0.00 | 2 |
| Aveiro | Portugal | 93.7%62% forested in 2000 | 0 | 0.00 | 2 |
| Montserrado | Liberia | 90.4%88% forested in 2000 | 0 | 0.00 | 3 |
| Coimbra | Portugal | 90.2%68% forested in 2000 | 0 | 0.00 | 2 |
| Bomi | Liberia | 89.3%99% forested in 2000 | 0 | 0.00 | 3 |
| Melaka | Malaysia | 88.7%83% forested in 2000 | 0 | 0.00 | 10 |
| Margibi | Liberia | 87.3%95% forested in 2000 | 0 | 0.00 | 3 |
| Leiria | Portugal | 87.0%51% forested in 2000 | 0 | 0.00 | 2 |
| Abidjan | Côte d'Ivoire | 85.2%71% forested in 2000 | 0 | 0.00 | 4 |
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 | Range species without a site |
|---|---|---|---|---|---|
| Sudd flooded grasslandsFlooded Grasslands & Savannas | South Sudan, Ethiopia | −10.4%semi-arid | 1 | 0.05 | 4 |
| Gobi Lakes Valley desert steppeDeserts & Xeric Shrublands | Mongolia | −9.6%arid | 0 | 0.00 | 4 |
| Caspian lowland desertDeserts & Xeric Shrublands | Kazakhstan, Russia, Turkmenistan, Iran | −9.5%arid | 0 | 0.00 | 12 |
| EspinalTemperate Grasslands, Savannas & Shrublands | Argentina | −9.4%dry sub-humid | 3 | 0.10 | 7 |
| Khangai Mountains conifer forestsTemperate Conifer Forests | Mongolia | −8.5%semi-arid | 0 | 0.00 | 4 |
| North Arabian highland shrublandsDeserts & Xeric Shrublands | Saudi Arabia | −8.1%arid | 0 | 0.00 | 4 |
| Mediterranean dry woodlands and steppeMediterranean Forests, Woodlands & Scrub | Algeria, Morocco, Libya, Tunisia, Egypt | −8.1%arid | 0 | 0.00 | 7 |
| Great Lakes Basin desert steppeDeserts & Xeric Shrublands | Mongolia, Russia | −7.9%semi-arid | 0 | 0.00 | 9 |
| Pontic steppeTemperate Grasslands, Savannas & Shrublands | Russia, Ukraine, Kazakhstan, Romania | −7.5%dry sub-humid | 11 | 0.11 | 7 |
| Badghyz and Karabil semi-desertDeserts & Xeric Shrublands | Turkmenistan, Afghanistan, Uzbekistan, Tajikistan, Iran | −6.3%arid | 1 | 0.07 | 7 |
| Region | Country | Aridity trend | Papers with a site | Per 10,000 km2 | Range species without a site |
|---|---|---|---|---|---|
| Lakes | South Sudan | −13.0%dry sub-humid | 0 | 0.00 | 5 |
| Biskra | Algeria | −12.8%arid | 0 | 0.00 | 4 |
| Central Equatoria | South Sudan | −12.8%dry sub-humid | 0 | 0.00 | 6 |
| Gabès | Tunisia | −12.3%arid | 0 | 0.00 | 3 |
| San Luis | Argentina | −12.1%semi-arid | 0 | 0.00 | 14 |
| Misratah | Libya | −12.1%arid | 0 | 0.00 | 2 |
| Gafsa | Tunisia | −12.0%arid | 0 | 0.00 | 3 |
| Kalmyk | Russia | −11.3%semi-arid | 0 | 0.00 | 8 |
| Hovd | Mongolia | −11.1%semi-arid | 0 | 0.00 | 4 |
| Sidi Bou Zid | Tunisia | −11.1%arid | 0 | 0.00 | 3 |
Aridity trend 1990 to 2024, cell by cell
3 · When the research happened
The same study sites split by decade of publication. Of the 6,655 located papers, 75.0% were published in the 2010s or 2020s and 54.1% since 2016. The whole corpus of 18,019 papers, located or not, is 72.8% from those two decades, so located research is more recent than the corpus. Of the 528 ecoregions with a located paper, 39 have none published since 2010; of the 963 admin1 regions, 84. 11 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. 6,655 of 18,019 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. 12 precise places (open sea, Southern Ocean, atoll centroids) fall in no ecoregion and 9 in no admin1 region; they are in the cell maps but not in the unit tables. Range species without a site is a country-level test (every unit in a country is credited with the country's whole range list), so it is an upper bound.
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 0.33 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 0.33 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 34.1 papers there (ecoregions) or 1077.7 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. 683 ecoregions in the universe, 513 with a located paper; negative binomial, dispersion theta 0.79, mean by Poisson quasi-likelihood, ridge lambda 300. 5-fold spatial cross-validation blocked by realm: pseudo-R2 0.55 (0.34 without the national-effort covariate), Spearman rho 0.68 out of fold. Labels against the 90 percent interval: 0 below, 655 within, 28 above.
| Ecoregion | Country | Observed | Expected, 90% interval | Tail probability | Label |
|---|---|---|---|---|---|
| Mindanao-Eastern Visayas rain forestsTropical & Subtropical Moist Broadleaf Forests | Philippines | 0 | 14.8490%: 0 to 49 | 0.094 | within |
| Drakensberg Escarpment savanna and thicketTropical & Subtropical Grasslands, Savannas & Shrublands | South Africa | 0 | 14.4590%: 0 to 48 | 0.096 | within |
| Angolan wet miombo woodlandsTropical & Subtropical Grasslands, Savannas & Shrublands | Angola, Congo - Kinshasa | 0 | 11.8390%: 0 to 39 | 0.112 | within |
| Kazakh semi-desertDeserts & Xeric Shrublands | Kazakhstan | 0 | 11.6690%: 0 to 39 | 0.113 | within |
| Snake-Columbia shrub steppeDeserts & Xeric Shrublands | United States | 1 | 23.6890%: 0 to 78 | 0.117 | within |
| Eastern Guinean forestsTropical & Subtropical Moist Broadleaf Forests | Côte d’Ivoire, Ghana, Togo | 0 | 11.0990%: 0 to 37 | 0.117 | within |
| Guianan piedmont moist forestsTropical & Subtropical Moist Broadleaf Forests | Venezuela, Brazil | 0 | 10.5390%: 0 to 35 | 0.122 | within |
| North Western Ghats montane rain forestsTropical & Subtropical Moist Broadleaf Forests | India | 0 | 10.1090%: 0 to 34 | 0.126 | within |
| North Tibetan Plateau-Kunlun Mountains alpine desertMontane Grasslands & Shrublands | China | 0 | 9.8490%: 0 to 33 | 0.128 | within |
| Kuh Rud and Eastern Iran montane woodlandsMontane Grasslands & Shrublands | Iran, Pakistan | 0 | 9.4290%: 0 to 32 | 0.132 | within |
| Ecoregion | Country | Observed | Expected, 90% interval | Tail probability | Label |
|---|---|---|---|---|---|
| Nansei Islands subtropical evergreen forestsTropical & Subtropical Moist Broadleaf Forests | Japan | 27 | 1.1290%: 0 to 4 | under 0.001 | above |
| Ussuri broadleaf and mixed forestsTemperate Broadleaf & Mixed Forests | Russia | 38 | 2.6390%: 0 to 9 | under 0.001 | above |
| Southern Acacia-Commiphora bushlands and thicketsTropical & Subtropical Grasslands, Savannas & Shrublands | Tanzania, Kenya | 123 | 10.2890%: 0 to 34 | under 0.001 | above |
| Western Java montane rain forestsTropical & Subtropical Moist Broadleaf Forests | Indonesia | 58 | 5.0490%: 0 to 17 | under 0.001 | above |
| Serra do Mar coastal forestsTropical & Subtropical Moist Broadleaf Forests | Brazil | 132 | 14.0190%: 0 to 47 | 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. 2,513 administrative regions (admin1) in the universe, 920 with a located paper; negative binomial, dispersion theta 0.38, mean by Poisson quasi-likelihood, ridge lambda 300. 5-fold spatial cross-validation blocked by country: pseudo-R2 0.33 (0.13 without the national-effort covariate), Spearman rho 0.57 out of fold. Labels against the 90 percent interval: 0 below, 2,429 within, 84 above.
| Region | Country | Observed | Expected, 90% interval | Tail probability | Label |
|---|---|---|---|---|---|
| Balochistan | Pakistan | 0 | 18.7190%: 0 to 80 | 0.228 | within |
| Papua | Indonesia | 0 | 14.9790%: 0 to 64 | 0.248 | within |
| Tabora | Tanzania | 0 | 13.5690%: 0 to 58 | 0.257 | within |
| Acre | Brazil | 0 | 13.4390%: 0 to 58 | 0.258 | within |
| Atacama | Chile | 0 | 13.2890%: 0 to 57 | 0.259 | within |
| Punjab | India | 0 | 10.4690%: 0 to 45 | 0.282 | within |
| San Luis | Argentina | 0 | 9.9990%: 0 to 43 | 0.287 | within |
| Somali | Ethiopia | 0 | 9.0290%: 0 to 39 | 0.298 | within |
| Ucayali | Peru | 0 | 8.4890%: 0 to 37 | 0.305 | within |
| Sipaliwini | Suriname | 0 | 8.3890%: 0 to 36 | 0.306 | within |
| Region | Country | Observed | Expected, 90% interval | Tail probability | Label |
|---|---|---|---|---|---|
| Borski | Serbia | 9 | 0.0390%: 0 to 0 | under 0.001 | above |
| Okinawa | Japan | 27 | 0.3790%: 0 to 2 | under 0.001 | above |
| Karlovacka | Croatia | 44 | 1.1290%: 0 to 5 | under 0.001 | above |
| Mtskheta-Mtianeti | Georgia | 6 | 0.0590%: 0 to 0 | under 0.001 | above |
| Mara | Tanzania | 77 | 2.4090%: 0 to 11 | 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 0.68 and 1.38, top decile 1.06; administrative regions (admin1): deciles 3 to 8 between 0.67 and 1.11, top decile 1.07. 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. 6,655 of 18,019 papers in this corpus have a precise trusted study site (36.9 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: 8,146 unit-papers over 683 ecoregions and 7,914 over 2,513 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 300, 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 felidae, 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
What this atlas measures, and what it does not.
The Felidae are the best-covered family in this series. Every accepted species has a literature, half of the studies can be placed at a field site, and all 45 species have at least one study site inside their native range. That is not true of any plant family measured so far, and it reflects both the size of the animals and a century of applied carnivore research.
What the record is not is even. Five species hold half of it. The median Endangered cat carries a sixth of the studies the median Least Concern cat carries. A quarter to a third of the literature on several regional species is reachable only through its common name, and for three more names that search has not yet been run. The atlas is a statement of where the indexed record stands, with the method and the gaps named, rather than a claim that the cats are well studied everywhere.