Station record analysis, 22 stations, 1976 to 2025

Climate science research poster example

Fifty years of heatwave days at twenty two weather stations, laid out as an A0 landscape billboard where the key finding occupies a full panel on the left. The finding is set large enough to read from across a hall, and the evidence for it runs down the right.

Create a poster with OneCraftA0 Landscape, printed at 1189 by 841 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.

TU
Heatwave days a year at twenty two stations with fifty years of record, 1976 to 2025
Heatwave days a year at 22 stations rose from 4.1 in the first decade of record to 10.7 in the last, a factor of 2.6.
2.6x
more heatwave days than fifty years ago
10.7 days a year now against 4.1 in 1976 to 1985
22
stations with a complete fifty year record
Screened from 58 in the regional network
W. Anand ¹, F. Okonkwo ¹, L. Marchetti ¹ ²
1 School of Climate Science, Tarrawonga University · 2 Regional Climate Observations Unit, Tarrawonga University
Definition and why it matters
A heatwave here is three or more consecutive days above the station own 90th percentile daily maximum, calculated over the whole 50 year record rather than over a fixed baseline. That choice matters. A fixed baseline threshold counts more days simply because the distribution has shifted, which is true but circular; a station specific percentile over the full record asks whether hot spells have become more frequent relative to that station own climate, which is the question a hospital or a power network actually plans against. The definition is also the reason this analysis needs a complete record: a station with a ten year gap has a different percentile from the one it would have had, and the count changes with it. That is why 36 of the 58 stations in the regional network are not in this analysis, and why the screening is reported as a flow rather than as a sentence. The three day rule is the other choice worth stating: it counts spells rather than days, so a single record breaking maximum contributes nothing unless the two days either side are also above the threshold.
HOW WE TESTED IT
Stations. 22 stations with a complete daily maximum record from 1976 to 2025, screened from 58.
Definition. Three or more consecutive days above the station 90th percentile daily maximum.
Baseline. Percentile computed over the full 50 years at that station, not over a fixed reference period.
Homogeneity. Seven stations moved site; each break tested and adjusted before counting.
Reporting. Decadal means across the 22 stations, and the five largest station level rises.
PARTICIPANT FLOW
Stations in the regional network
58
Record shorter than fifty years
24
Stations with a fifty year record
34
More than 5% of daily maxima missing
12
Stations analysed
22
Of those, adjusted for a site move
7
10.7 days
a year in the last decade
Mean across the 22 stations, 2016 to 2025
days
First
Last
036912
Figure 1. Mean heatwave days a year across 22 stations, first decade of record and last.
Source: adjusted daily maximum records, 1976 to 2025.
Results
Decadal means across the 22 stations run 4.1, 5.0, 6.3, 8.4 and 10.7 heatwave days a year, so every decade is higher than the one before it and the last is 2.6 times the first. The rise is not carried by a few stations: 20 of the 22 show a higher count in the last decade than the first, and the two that do not are both coastal. The five largest rises are Kelburn Downs, from 3.6 to 13.1 days; Arradale, 4.4 to 12.8; Mount Sennen, 3.9 to 11.9; Gulliver Flat, 5.1 to 12.6; and Pennyquick, 4.0 to 11.2. Four of those five are inland. The seven stations adjusted for a site move behave like the other fifteen, with a mean rise of 6.4 days against 6.7, so the adjustment is not driving the trend. Two limits belong on the board. The percentile is station specific, so these counts cannot be compared between stations, only within one; and a three day rule ignores single very hot days, which have risen faster still at 14 of the 22. The trend is also not an artefact of the percentile choice: recomputing every count against a fixed 1976 to 1995 baseline raises all five decadal means and leaves the ratio between first and last at 2.4.
Decadal means 4.1, 5.0, 6.3, 8.4 and 10.7 days
20 of 22 stations higher in the last decade than the first
Site move adjustments do not drive the trend
Four of the five largest rises are inland stations
REFERENCES
1.Okonkwo, F. and Anand, W. (2025) Station percentile against fixed baseline heatwave definitions. Journal of Applied Climatology, 55(3), 312 to 338.
2.Marchetti, L. (2024) Break detection and adjustment in fifty year daily maximum records. Climate Data Practice, 20(2), 95 to 118.
3.Rehman, A. (2026) Coastal moderation of heatwave frequency. Tarrawonga Climate Reports 2026-07.

Block by block

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

Key finding panel, title and authors
A full height panel on the left carries the title over three lines and then the finding in large type: heatwave days a year at 22 stations rose from 4.1 in the first decade of record to 10.7 in the last, a factor of 2.6. Three authors and a QR code sit at the foot.
Two numbers beside the finding
Two key number blocks in the panel: 2.6 times more heatwave days than fifty years ago, with the two decadal means in the detail line, and 22 stations with a complete fifty year record, screened from 58 in the regional network.
Definition and why it matters
Three or more consecutive days above the station own 90th percentile daily maximum, computed over the whole record rather than a fixed baseline. The block explains why the fixed baseline version is circular and why a complete record is required for the percentile to mean anything.
How we tested it
Five labelled entries covering the 22 stations and the screening, the three day definition, the full record percentile, the homogeneity testing of seven stations that moved site, and the reporting of decadal means plus the five largest station level rises.
Station screening flow
Fifty eight stations in the regional network, 24 with a record shorter than fifty years, 34 with a fifty year record, 12 with more than 5% of daily maxima missing, 22 analysed, and 7 of those adjusted for a site move.
Figure and third key number
A two bar figure compares the first and last decade at 4.1 and 10.7 days. Above it a key number block reads 10.7 days a year in the last decade, with the detail line naming the 22 station mean for 2016 to 2025.
Results and references
The results block gives all five decadal means, the 20 of 22 stations that rose, the five largest station rises by name, the check that site move adjustments are not driving the trend, and the fixed baseline sensitivity check at a ratio of 2.4. Three references follow.

What makes this board work

One finding fills a panel

The key finding panel states the whole result in a sentence a reader takes in from across the room, and the two numbers beside it give the ratio and the station count. Everything else on the board exists to let somebody check it.

The definition is argued, not assumed

A station specific percentile over the full record rather than a fixed baseline, with the reason given: a fixed threshold counts more days because the distribution moved, which is true and circular. The board also reports what a fixed baseline would have given.

The screening is a flow

58 stations in the network, 24 with a record under fifty years, 34 complete, 12 with more than 5% of days missing, 22 analysed and 7 adjusted for a site move. Every subtraction holds, which is what a reader checks before believing the trend.

Questions people ask

What goes on a climate science research poster?

The definition of the event, the stations or grid and how they were screened, the record length, the homogeneity treatment, the trend with its decadal or annual values, and a sensitivity check against a different definition. The screening flow matters as much as the trend.

How do I show a fifty year trend at poster size?

Decadal means read better than annual values from a metre away. This board gives the five decadal means in the results text and spends the figure on the first and last decade alone, so the size of the change reads instantly and the detail is still there for a close reader.

Should the key finding be a sentence or a number?

A sentence with the numbers in it. A billboard panel gives you room for both, and a sentence survives being read out of context better than a bare ratio. Keep it under about 130 characters so it can be set large without wrapping into a paragraph.

What text size does an A0 landscape climate poster need?

Body text here is near 24 pixels on the 2340 by 1656 canvas, about 12 millimetres printed, readable at a metre. The key finding panel runs several times that, which is what lets it work at five metres rather than one.

Can I print this at 300 DPI?

The export doubles the working canvas, giving 4680 by 3312 pixels at A0 landscape, roughly 100 DPI at full size. That is standard for large format work viewed from a metre. There is no bleed, no crop marks and no CMYK conversion.

How do I adapt this to my own stations?

Retype the key finding panel, the two stat blocks beside it and the third in the right column, then edit the screening flow node labels and counts. The figure holds its own categories and values, and the methods block takes five labelled entries.

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 my climate science research poster example

Other poster examples

Oceanography research poster example

Kelp canopy cover at sixteen fixed reef transects over five consecutive summers, laid out as an A0 landscape board in two rows. The top row carries the method and the figure on a light ground, and the bottom row sets the results in a dark card that pulls the eye down.

Hydrology research poster example

Runoff ratio at one catchment gauge before and after thirty per cent reforestation, laid out as a four column A0 landscape board. Each column takes one stage of the argument, so a reader moves left to right from the question to the limits without doubling back.

Astronomy research poster example

Six months of photometry on 140 candidate variable stars, laid out as an A0 portrait showcase board with two figures set into a three column grid. The board answers the question a small observatory asks each season: what fraction of somebody else catalogue can a 0.4 metre telescope actually resolve.

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 .