How many captures (bar length) and individuals (label) each species contributed.
Brighter bars = caught more times per animal (a "trap-happy" species).
Written automatically from the live data for this site & window, the same numbers the rest of the app is built on.
Who you'll meet
Population signals
How abundant the small mammals are here and how that's changing — a detection-corrected estimate alongside honest indices (MNKA, catch-per-effort), the environmental drivers the population tracks, and whether trapping ran long enough to find every species.
For every co-located driver, we scan lags 0–12 months and keep the strongest correlation with monthly catch-per-effort.
Bars show that best correlation (sign = direction); the label is the lag at which it peaks. The longest bar is the signal this population follows most closely; the others are candidate co-drivers.
Reading the r-value: r runs from -1 to +1 and measures how tightly two things move together: near the ends they rise and fall in near-lockstep, near 0 they barely relate. The sign is the direction : “+” means more driver → more animals, “−” means more driver → fewer.
It is not the percentage of the population explained; that is r² , the square of r (so r = -0.50 means about a quarter, 0.25, of the month-to-month swings line up). With only a handful of months and many lags scanned, read r as a lead worth a closer look , not a proven cause.
A longer bar isn't proof of cause: drivers correlate with each other, and scanning many lags can flag a strong match by chance. Treat this as a ranking of leads to investigate.
Traps miss animals. On NEON's multi-night bouts (pathogen grids run ~3 nights in a row), the recaptures , animals caught more than once in the same bout, tell us how many we didn't catch, so we can estimate the true number present.
The navy line + band is the estimated abundance N̂ with a 95% interval; the grey line is MNKA (minimum known alive). The gap between them is the detection correction.
When a driver leads this population (per the ranking above), its monthly value is overlaid here at that best lag (right axis) so you can watch the abundance track it.
- k ≥ 3 nights → Schnabel estimator; k = 2 → Chapman. Single-night grids can't be estimated (that's what MNKA/CPUE are for).
- p̂ is the per-night detection probability (Model M0): the share of present animals we'd expect to catch on any one night.
- We hide the estimate when there are fewer than 3 within-bout recaptures , too few to be stable.
- N̂ is animals on the sampled grid(s) , not a per-hectare density; a true density needs spatially-explicit capture-recapture (SECR), which a single 10×10 grid per plot can't support, so we deliberately decline it.
The math assumes the population didn't change over the bout and every animal is equally catchable; real animals aren't, so read this as a defensible index, not a census. Schnabel 1938; Chapman 1951; Otis et al. 1978.
MNKA (Minimum Number Known Alive) counts how many individuals were known to be alive each month: caught that month, or before and after. A transparent abundance index (Krebs 1966).
The dotted grey line ( right axis ) is catch per effort : captures per 100 trap-nights . A trap-night is one trap set out for one night, computed from this site's own data (the actual traps set × the nights they ran), so a smaller grid or fewer nights doesn't skew it, and it does not assume a fixed 100-trap grid. Sprung or disturbed traps count as half a trap-night (Nelson & Clark 1973).
It's a relative-abundance index for comparing sites and trends, not a population estimate (it doesn't correct for detectability; that's what the detection-corrected abundance tab is for).
As more trapping bouts accumulate, how many species have been found? When the curve flattens, you've probably found them all. Genus-only IDs ("X sp.") are excluded so an unidentified catch isn't counted as its own species.
The dashed Chao1 line is a bias-corrected minimum estimate of true richness, a floor that includes species not yet caught, shown with a 95% interval. When doubletons (species caught as exactly 2 individuals) are scarce it's unstable and flagged as a lower bound (Chao 1987; Gotelli & Colwell 2001).
Each point is one month: its catch per 100 trap-nights against the value of the selected environmental driver (with your lag applied).
A rising cloud means more animals when the driver is high; the dashed line is an OLS fit. This is the same signal as the correlation banner above, shown as a shape so you can spot thresholds or saturation.
The sex and life-stage breakdown of every handled animal. The number in the middle is the total handled.
Three views of diversity, all in the same intuitive unit, an effective number of species , indexed by q , how much rare species count:
- q = 0 · plain species richness (every species counts equally).
- q = 1 · exp(Shannon) : the effective number of common species.
- q = 2 · inverse Simpson : the effective number of dominant species.
They always shrink (q0 ≥ q1 ≥ q2). When q1 sits close to q0 the community is even ; when it drops far below, a few species dominate .
Abundance = distinct individuals per species, so a much-recaptured animal isn't counted twice. Hill 1973; Jost 2006; Chao et al. 2014.
Captures by species per month, summed across all plots (the site total) by default. Use the plot picker to drill into a single plot.
Values are raw captures (not effort-adjusted). Only species with a handful of captures are drawn, to keep the legend readable.
A violin for each species shows the spread of adult body weights, one mean weight per individual (so a much-recaptured animal isn't counted many times), wide where many animals fall, with a line at the mean. Species with at least 8 weighed adults are shown.
Species are ordered lightest → heaviest (log scale, since a pocket mouse and a woodrat differ ~30×).
If you've opened an individual's dossier, a gold diamond marks where it sits in its species.
The field measurements NEON records, summarised per species: the median with the 5th–95th-percentile range (a typical-adult envelope) and the number of records ( n ).
A small ⚠ marks a measure that contains an extreme value beyond that range, likely a data-entry error (e.g. a 357 g kangaroo rat). The value is kept in the data for your QC but is excluded from the median and the range; hover the ⚠ for the raw min–max.
Adults only : juveniles and subadults are much lighter and shorter, so pooling growth stages would inflate the range and pull the median; restricting to staged adults gives an honest "typical adult size."
Weight and hindfoot length are taken at nearly every handling; tail and ear are measured far less often, so those cells are blank ( — ) for species with fewer than 3 records.
Median + percentile range, not a "mode": for a continuous measurement the median and spread are the honest summary. A species' mean weight and the "heaviest species" call shown elsewhere need ≥8 staged adults, or they're withheld as too noisy. Counts confirmed species-level IDs only.
The longest we confirmed an individual alive : its age at last capture (a conservative age at first capture, plus how long we tracked it). Shown per species where at least 5 individuals were each caught 3+ times .
It's a floor, not a lifespan : animals still alive or that left the grid aren't counted (one not seen again is presumed gone: death or dispersal , never assumed dead), and the record only spans the years sampled here. NEON keeps a tag on an animal for life and doesn't recycle tag numbers, so a multi-year career is a real recapture span , not a tag mix-up; true lifespan is simply longer than what we caught.
Because it's the single longest-tracked individual , species we trap more often reach higher floors just from having more chances, so read each as its own floor, not a ranking of which species lives longest.
A model-based estimate from apparent survival (the gold-standard mark-recapture approach) is deliberately not shown : it needs an offline-validated capture-session model and would over-claim on this data.
Captive maxima (where shown) are from the AnAge database and run several times the wild value, a sanity ceiling, not an expectation.
By calendar month, the share of adults that are reproductively active :
● breeding males (scrotal) and ● reproductive females (pregnant or lactating).
Peaks reveal the breeding season; the dip shows the off-season.
Body-size lab · every individual on one map
Every individual NEON tagged here is a
dot
, placed by its
hind-foot length
(x) and
weight
(y) and coloured by
species
, like a height-vs-weight recruiting board, but for rodents.
Tap a dot
to pin its card; drag the card around, resize it from the corner, and
download the chart with the cards still on it
. Tap
“Open QC history card”
on a card for that animal's every-capture measurements + automatic data-quality flags.
This is a
morphometric / QC map, not a body-condition index
: in these desert rodents hind-foot length barely predicts mass (r ≈ 0.15), so a dot far from its species' cloud flags an unusual or mis-typed record, not “fitness.”
Body-size lab
Every individual NEON tagged here is a dot , placed by its hind-foot length (x) and weight (y) and coloured by species , like a height-vs-weight recruiting board, but for rodents.
Tap a dot to pin its card; drag the card around, resize it from the corner, and download the chart with the cards still on it . Tap “Open QC history card” on a card for that animal's every-capture measurements + automatic data-quality flags.
This is a morphometric / QC map, not a body-condition index : in these desert rodents hind-foot length barely predicts mass (r ≈ 0.15), so a dot far from its species' cloud flags an unusual or mis-typed record, not “fitness.”
Each dot is one animal, by hind-foot length × weight, coloured by species. Tap a dot to pin a card; open its QC history card for every capture's measurements and automatic flags, then download any of it.
Pick one species from the filter to add its size–mass fit line (drawn only where length really predicts mass: n ≥ 15, |r| ≥ 0.3) and median crosshairs that split it into body-shape quadrants.
The gold diamond marks the individual you're currently tracking.
Trap-grid home range
NEON small-mammal plots use a grid of
up to 10×10
traps spaced 10 m apart (columns A–J, rows 1–10).
Not every plot is the full 10×10:
some sites or bouts deploy fewer traps. For example, a reduced grid is used where capture rates are very high. Empty cells here can mean “no trap set” as well as “not caught.”
The
heatmap
colors each cell by how many times this animal was caught there; white dots mark the actual capture cells and the gold ✕ is its centre of activity.
The
replay
animates its captures in order: press ▶ to watch it move.
Hotspot blur
smooths the grid to show its core area.
Trap-grid home range
NEON small-mammal plots use a grid of up to 10×10 traps spaced 10 m apart (columns A–J, rows 1–10).
Not every plot is the full 10×10: some sites or bouts deploy fewer traps. For example, a reduced grid is used where capture rates are very high. Empty cells here can mean “no trap set” as well as “not caught.”
The heatmap colors each cell by how many times this animal was caught there; white dots mark the actual capture cells and the gold ✕ is its centre of activity.
The replay animates its captures in order: press ▶ to watch it move.
Hotspot blur smooths the grid to show its core area.
Where on the plot's trap grid this animal kept turning up. Hit play to replay its captures over time.
Species diversity across the site
Each circle is a NEON trapping
plot
, placed at its real coordinates.
Circle
size
= total captures there;
color
= species (same colors as the charts).
When an individual is selected, its plots get a pulsing
gold ring
.
Recapture movement
(toggle, top-right): curved
teal arcs
link grids where the same tagged animals were recaptured: thicker = more individuals made that move. It's mark-recapture ("here, then there"),
not
a tracked route; a selected individual's own moves draw in gold.
Hover a circle for its species + counts; switch basemap top-right.
Site map
Each circle is a NEON trapping plot , placed at its real coordinates.
Circle size = total captures there; color = species (same colors as the charts).
When an individual is selected, its plots get a pulsing gold ring .
Recapture movement (toggle, top-right): curved teal arcs link grids where the same tagged animals were recaptured: thicker = more individuals made that move. It's mark-recapture ("here, then there"), not a tracked route; a selected individual's own moves draw in gold.
Hover a circle for its species + counts; switch basemap top-right.
Each marker is a plot, sized by captures and colored by species. The selected individual's plots glow gold.
Capture leaderboard
Every animal NEON caught is ranked by how often it turned up in traps.
Click any row
to open that individual's dossier.
Switch
category
to re-rank by weight, career length, roaming, or weight-for-its-species (chonk).
Rarity tiers
come from total captures:
⭐ Legendary
15+ captures
💎 Epic
10–14
🔵 Rare
6–9
🟢 Uncommon
3–5
▫ Common
1–2
How the leaderboard works
Every animal NEON caught is ranked by how often it turned up in traps.
Click any row to open that individual's dossier.
Switch category to re-rank by weight, career length, roaming, or weight-for-its-species (chonk).
Rarity tiers come from total captures:
Every individual ranked. Pick a category, then click a row to open its dossier.
Search the network
Search every NEON small-mammal site at once, straight from the bundled data so it answers instantly.
Find a species
lists every site where that animal was trapped, with its within-site index and the years it was caught there.
Find sites
filters sites by how widespread a species is, or by how many species each site has recorded.
The number shown is
MNKA
(minimum number known alive): a within-site index of how many of that species were known to be alive at one time. It is not a population count, and it is not a fair ranking between sites with different trapping effort.
Search the network
Search every NEON small-mammal site at once, straight from the bundled data so it answers instantly.
Find a species lists every site where that animal was trapped, with its within-site index and the years it was caught there.
Find sites filters sites by how widespread a species is, or by how many species each site has recorded.
The number shown is MNKA (minimum number known alive): a within-site index of how many of that species were known to be alive at one time. It is not a population count, and it is not a fair ranking between sites with different trapping effort.
Look across all NEON sites at once. Find where a species turns up, or which sites are the most species-rich. Pick a result to load that site.
Each capture's weight (navy) and hind-foot length (cardinal) plotted over time.
The shaded band is the middle 50% of adult weights for this species, so you can see whether the animal runs heavy or light against grown adults (juveniles would make the band misleadingly low). The ♦ marks its heaviest capture.
An honest adult weight percentile within species : i.e. "how heavy is this animal for its kind?"
50 = a perfectly typical adult; the delta shows how far above/below typical it sits.
Why not a body-condition index? In these desert rodents foot length barely predicts mass, so a fancier index would just rank noise. The body-size map below shows the real relationship.
Every measured animal plotted by weight × hind-foot length . The faint grey dots are all other species; the colored cloud is this animal's species (by life stage).
The gold diamonds are this individual's captures: high in the cloud = a big one.
A dashed size–mass fit line is drawn only for species where length actually predicts mass (so you're never shown a fake trend.)
Every individual capture event for this animal: date, plot, trap cell, measurements, and field notes. Use the search box to filter.