Acoustic survey, 30 wetlands, 4,300 hours

Ecology research poster example

An ecology poster is usually read for two things: what was found where, and whether the sampling was good enough to believe it. This board answers both for an acoustic survey of six frog species at 30 wetlands across two wet seasons, with the species counts written as one ordered results block, the grazing comparison in a figure and the seasonal means in a comparison bar block.

Create a poster with OneCraftA0 Portrait, printed at 841 by 1189 millimetres

The whole board

The poster at full size, exactly as it prints. Every number, citation and caption on it was written for this example, so the layout is being judged on real content.

MU
Six frog species across 30 wetlands: what 4,300 hours of acoustic recording found
N. Arrowsmith ¹, D. Kuruvilla ¹, H. Torrens ²
1 School of Environmental Science, Marrawong University · 2 Regional Wetlands Trust, Marrawong
ABSTRACT
Frog surveys are usually done by ear on a handful of warm nights, which finds the loud species and misses the short callers. This survey replaced the night visit with an acoustic logger left in place for a whole wet season. Thirty wetlands across one catchment were recorded over two consecutive wet seasons, 4,300 hours in total, and every detection was first classified by an automated call recogniser and then checked by a person. Six species were detected. The marsh froglet called at all 30 wetlands and the basin tree frog at 3. Fifteen of the 30 wetlands are grazed to the water edge and 15 are fenced. Three of the six species were never recorded at a grazed wetland, and mean species richness per season at grazed wetlands was 2.0 against 4.1 at fenced wetlands. The pattern held in both seasons and did not depend on wetland area or on how long each logger ran.
INTRODUCTION
A frog calls when the weather suits it, not when a surveyor is free. That mismatch is the oldest problem in amphibian monitoring: a species that calls for eleven nights a year will be absent from almost every survey that visits three times, and its absence will be read as a decline rather than as a missed window. Acoustic loggers change the sampling problem into a storage problem. A recorder left on a post for a season hears every chorus, and the cost moves from field nights to the hours of listening afterwards, which is where an automated recogniser earns its place. This survey was designed to answer two questions that a visit based survey cannot answer well.
METHODS
Sites. Thirty wetlands in one catchment, 15 grazed to the water edge and 15 fenced for at least five years, matched in pairs on area, depth class and distance to the nearest permanent water.
Recording. One logger per wetland, mounted 1.5 m above the water line, recording 4 minutes in every 20 from dusk to dawn across two wet seasons, giving 4,300 hours of audio in total.
Detection. An automated recogniser trained on 1,200 reference calls flagged candidate detections, then a person checked every flagged minute and a random 5% of unflagged minutes.
Measure. A species counts as detected at a wetland when at least three verified calls were heard on two or more separate nights, which rules out a single pass through bird or a machine noise.
Analysis. Occupancy compared between grazed and fenced pairs with a paired test over the 15 pairs, and species richness compared per wetland per season rather than pooled across both seasons.
DISCUSSION
Three of six species absent from every grazed wetland is a stronger signal than a richness difference, because it does not depend on how a marginal detection is counted. The likely cause is structure rather than water: the three missing species call from sedge and reed at the water edge, which is the first thing stock remove. Loggers hear calling, not breeding, and a fenced wetland that fills later gets a longer effective survey. Baxendale, P. (2024) Acoustic monitoring of anurans in grazed landscapes. Wetland Ecology Reports, 16(2), 91 to 114. Idris, M. and Torrens, H. (2025) Recogniser error in wet season frog audio. Journal of Bioacoustics, 9(4), 233 to 251. Arrowsmith, N. (2026) Edge structure and calling habitat. Marrawong Ecology Working Paper 2026-02.
Season 1
Season 2
GrazedFenced02468
Figure 1. Mean species per wetland by grazing status and wet season.
Source: 30 wetlands, two wet seasons, verified detections.
SPECIES PER WETLAND
Fenced, season 1
4.0
Grazed, season 1
2.0
Fenced, season 2
4.2
Grazed, season 2
2.0
What was heard
PRIMARY OUTCOME
Six species, ordered by the number of the 30 wetlands each was verified at: marsh froglet 30, reed piper 24, gully bell frog 19, sedge chirper 11, claypan toadlet 7 and basin tree frog 3. The first two were heard in both seasons at every wetland they occupied. The last three were heard only at fenced wetlands, and none of them was ever verified at a grazed one across the whole 4,300 hours. Mean richness per season was 2.0 at grazed wetlands and 4.1 at fenced, and the order of the six species was the same in both seasons.
Six species verified across 4,300 recorded hours at 30 wetlands
Marsh froglet at all 30 wetlands, basin tree frog at 3
Three of six species never recorded at a grazed wetland
Mean richness 2.0 grazed against 4.1 fenced, both seasons

Block by block

What each block on the board is for, in the order a reader walks it.

Title band and three authors
The title names the six species, the 30 wetlands and the 4,300 hours, so the effort is visible before a reader steps closer. Three authors sit over two affiliations, and the recording effort is named in the figure source lines.
Abstract
The whole survey in one block: why a logger replaces a night visit, the 30 wetlands and two seasons, the six species, and the finding that three of them were never heard at a grazed wetland. The per season richness figures of 2.0 and 4.1 appear here first.
Why record instead of listen
The introduction block explains the sampling problem an acoustic survey solves: a species that calls on eleven nights a year is absent from almost every three visit survey. It also names the cost the method moves, from field nights to hours of listening.
Methods as a labelled protocol
Five labelled entries covering the paired site selection, the recording schedule, the automated recogniser and the human check, the detection rule, and how occupancy was compared between the 15 matched pairs.
Figure 1, richness by grazing status
Mean species per wetland for grazed and fenced sites in each of the two seasons, four bars in total. Two categories and two series keep the axis labels clear of the legend, and the source line names the 30 wetlands and two wet seasons behind the bars.
Species per wetland, the comparison bars
Four labelled bars repeat the figure with each value printed beside its bar: fenced 4.0 and grazed 2.0 in season 1, fenced 4.2 and grazed 2.0 in season 2. A reader who cannot judge bar heights from a metre away still reads the numbers.
What was heard
The results block lists all six species with the number of wetlands each was verified at, from marsh froglet at 30 to basin tree frog at 3, then states that the bottom three were heard only at fenced wetlands. The counts are written as one ordered sentence rather than as cards.
Discussion, limits and references
Why structure rather than water is the likely cause, two limits stated plainly, and three numbered references. This layout has no reference block, so the citations sit at the end of the discussion block instead of being dropped.

What makes this board work

The sampling effort is on the board

Thirty loggers, four minutes in every twenty from dusk to dawn, two wet seasons and 4,300 hours appear in the methods block. An ecology reviewer discounts a species list until the effort behind it is stated, because absence from a short survey is not absence.

A detection rule, not a judgement

A species counts as present at a wetland only after three verified calls on two separate nights. That rule is printed in the methods block, so the 30, 24, 19, 11, 7 and 3 in the results block mean the same thing for every species.

The grazing result is a count, not a model

Three of six species were never recorded at any of the 15 grazed wetlands across the whole survey. That statement survives any argument about how a marginal detection is counted, which is why it leads the conclusions rather than the richness difference.

Questions people ask

What goes on an ecology research poster?

The question, the sites and how they were chosen, the sampling effort in hours or visits, the detection or measurement rule, the results by species or site, and the limits. Ecology reviewers read the effort first, because a species list without it cannot be compared to anything else.

How do you show species counts on a poster?

This board writes them as one ordered sentence in the results block, species by species from 30 wetlands down to 3. If you want them drawn, a horizontal bar chart holds species names without rotating them, and it is one of ten chart kinds a figure block offers. Keep the species in the same order in the text, any chart and the conclusions.

Should absence be reported as absence?

Only with the effort attached. This board says three species were never recorded at a grazed wetland across 4,300 hours and two wet seasons, which is a much stronger statement than saying they were not detected. Print the rule that decides presence, then the counts mean one thing.

How large does an ecology conference poster print?

This board is A0 portrait, which is 841 by 1189 millimetres on the wall, built on a 1656 by 2340 pixel canvas. The export runs at twice the canvas, 3312 by 4680 pixels, which is roughly 100 dots per inch at full size. There is no bleed, no crop marks and no CMYK output.

Can I put a site photograph on the board?

Not on this layout. Seven figure blocks across the forty poster designs are photo blocks that cannot be filled, and this design is not one of them: its single figure is a chart. If a photograph is essential, choose a layout with an image block and check that it can be filled before you write the copy.

How many species should a card grid hold?

As many as the layout has cards for. This design carries one figure, a four row comparison bar block and four checked points rather than six species cards, so the species and their counts were written into the results block as one ordered list. Confirm what a design holds before planning content for it.

Build your own in about a minute

The button below opens the generator with this use case already described. Change the wording to match your own, generate, then edit anything you like.

Make one like this

Other poster examples

Sports science research poster example

A sports science board is usually built around one figure, because a coach standing at it wants the gradient before the method. This poster gives the whole top half to sleep against 40 metre sprint time across 416 athlete weeks, then explains underneath what 0.12 seconds is worth on the pitch.

Agriculture research poster example

A variety trial poster is read table first, because a grower wants the ranking and the agronomist wants the spread. This A0 landscape board puts an eight row results table across the middle, with yield, protein and lodging for each variety, and keeps the method and the caveats on either side of it.

Physiotherapy research poster example

A non inferiority trial poster has to make the margin visible, because the whole claim rests on it. This board randomises 180 patients to telehealth or in clinic knee rehabilitation, states the 4 point margin in the methods, and puts the 1.2 point difference and its interval in a stat block next to the two arm means.

Want the steps in the builder? Read Add charts and diagrams, then choose the template, theme and size. For everything this generator can do, see the poster maker.

Sources

Written and checked by the OneCraft team. Last checked .