Melbourne’s green spaces are often described through parks and trees. This report looks at something smaller: garden beds in streets and parks. Using one public City of Melbourne dataset, it explores where these beds are, how their condition differs, and how planting type, irrigation and size relate to their condition.
The assessments were carried out in 2023–2024. These charts describe the garden beds recorded in that assessment, not every green space in Victoria or their condition today.
The charts below analyse 4,040 distinct garden bed records after removing repeated asset IDs.
We start by comparing the number of recorded beds in different neighbourhoods. Colour separates beds in streets from those in parks. Use the legend to hide or show either group.
Answer: Docklands has the most garden beds recorded in this inventory, followed by Parkville and Kensington. Parks contain most of the recorded beds in several neighbourhoods, but the mix of park and street beds is not the same everywhere.
Explanation: Each horizontal bar represents one Melbourne neighbourhood. A longer bar means more garden beds were recorded there. The coloured sections separate beds located in parks from those along streets. For example, Parkville has a large park-bed section, while street beds make up a more noticeable proportion in Southbank and the CBD Hoddle Grid. You can hover over a coloured section to inspect its recorded count. This chart compares the number of beds, not the total garden area or the amount of greenery available to residents. A small bed and a large bed are each counted once.
Here, each bar represents a known bed type. The segments show the percentage of beds in each condition category, so types can be compared even when their total numbers differ.
Answer: Yes. The recorded condition differs between garden bed types. Annual beds have a more noticeable share rated Excellent, while perennial and mulch beds include substantial shares rated Good and Poor.
Explanation: Each bar represents a type of garden bed, and every bar totals 100%. The colours show how its beds are distributed across the five condition ratings, from Total failure (1) to Excellent (5). This allows us to compare types even if one has many more beds than another. The bulb meadow bar is entirely Very good in the displayed data, but that does not show how many bulb meadow beds were assessed; a small group can produce an extreme percentage. These ratings describe assessed garden-bed condition, not biodiversity or the number of plants.
The dataset classifies beds by vegetation origin. This chart compares the four recorded origins across parks and streets, excluding unspecified origins.
Answer: The planting mix differs between parks and streets. Exotic vegetation forms a large share in both, but parks have a higher proportion classified as Indigenous, whereas streets have a higher proportion classified as Native.
Explanation: The two bars represent garden beds in parks and streets. Each bar totals 100%, with its coloured sections showing the shares recorded as Exotic, Indigenous, Native or Mixed. Comparing the sections shows differences in planting origin rather than differences in the total number of beds. Categories with unspecified origins have been excluded. In the dataset, Indigenous and Native are separate categories, so they should not be combined or treated as synonyms. A bed’s origin classification does not tell us how many individual plants it contains.
This chart compares assessed condition scores for beds with different irrigation records, split by park or street location.
Answer: There is no clear difference in the typical recorded condition between garden beds with irrigation and those without it. The two groups show very similar score ranges.
Explanation: The x-axis separates beds recorded as having irrigation (Yes) from those without irrigation (No). The y-axis is their condition score from 1 to 5. In each boxplot, the box covers the middle half of the recorded scores and the line inside marks the median. The whiskers and dots show the wider spread of scores. Both groups have strongly overlapping boxes, mainly around scores 2 and 3. The displayed comparisons are largely based on street-bed records. This is an observed comparison: it does not prove that irrigation improves condition, has no effect, or causes worse condition.
Finally, each dot represents one assessed garden bed. Its horizontal position is its area, while its vertical position is the condition rating. The area axis is logarithmic to make small and large beds easier to compare.
Answer: No consistent relationship is visible between garden-bed size and condition. Beds of many different sizes receive scores of 2, 3 and 4.
Explanation: Each dot represents one assessed garden bed. Its position from left to right shows its area in square metres; its height shows its condition score from 1 to 5. Colours separate park beds from street beds. The area axis uses a logarithmic scale, so every step from 1 to 10, 10 to 100 or 100 to 1,000 is a tenfold increase in area. The dots have been spread slightly vertically so overlapping observations are easier to see; the condition ratings themselves are whole numbers. Because low and high ratings appear among both smaller and larger beds, the chart does not show that size by itself determines condition.
Taken together, the five graphs show that Melbourne’s garden beds vary by neighbourhood, by bed type and by planting origin. Docklands, Parkville and Kensington have the most recorded beds, while the park-versus-street balance changes across neighbourhoods. Condition ratings differ by bed type, and the balance of native, indigenous and exotic planting differs between parks and streets. The irrigation and size charts show why it is important not to judge condition using only one feature. These are useful patterns for future investigation, but this dataset does not establish what caused them.
The data was collected by Royal Botanic Gardens Victoria for the City of Melbourne during 2023–2024. Only the City of Melbourne municipality is included. Multiple records for the same garden bed were reduced to one asset ID to avoid counting it several times. Missing or unspecified categories are excluded where they cannot answer the chart question. Condition scores outside 1–5 are excluded from condition comparisons. The condition score is an ordinal rating from 1 (total failure) to 5 (excellent/new). This is a historical snapshot, not a live report on current conditions.
City of Melbourne. (2025). Renewals for Nature: City of Melbourne garden bed inventory 2024 [Data set]. City of Melbourne Open Data Portal. https://data.melbourne.vic.gov.au/explore/dataset/renewals-for-nature-city-of-melbourne-garden-bed-inventory-2024/
I acknowledge the use of ChatGPT (OpenAI, 2026) solely for the refinement of written work, including grammar correction, clarity, and readability. The tool was used to improve the presentation of the written content without changing its original meaning or intent.