Ten people died in Australian mines in 2024, the worst year since 2015. Nobody can tell you how many have died this year

The number that does exist

Safe Work Australia's Key WHS Statistics Australia 2025, released in October 2025, is the current authoritative national picture. It reports 188 worker fatalities from traumatic injuries across all industries in 2024, a rate of 1.3 deaths per 100,000 workers. Both are slightly below the five-year averages of 191 deaths and 1.4 per 100,000.

Mining ran the other way. The report records 10 mining fatalities in 2024, which it notes is 39 per cent above the five-year average of 7.2 for the industry. Mining's fatality rate, 3.4 deaths per 100,000 workers, was the third highest of any industry, behind agriculture, forestry and fishing on 13.7 and transport, postal and warehousing on 7.4.

Rate and count answer different questions, and the distinction matters here. By count, mining is nowhere near the worst: transport recorded 54 deaths in 2024, agriculture 44 and construction 37. By rate, mining is in the worst three, because it employs comparatively few people. A reader asking "is mining dangerous" is asking about rate. A reader asking "how many people died" is asking about count. Both are below.

Read by us from Safe Work Australia's Traumatic Injury Fatalities detailed data file, Table 1.1, division B Mining. See the methodology note.

Read across that decade and 2024 stands out. Mining sat between four and seven deaths a year for eight consecutive years, then went to 10. Nothing in the published data explains why, and we are not going to invent a cause: a jump from six to ten in an industry this size can be a change in risk or it can be the arithmetic of small numbers, and one year cannot tell you which. What it does tell you is that the 2025 and 2026 figures matter, and those are the ones nobody has.

Why there is no 2026 number

We went looking for a current-year count through four separate channels. None of them produces one, and the reasons are different in each case.

The finalised national dataset stops at 2024. Safe Work Australia's Traumatic Injury Fatalities detailed data file is the authoritative series, broken down by industry, occupation, mechanism and jurisdiction. The edition published in October 2025 runs from 2015 to 2024. That is the file the table above comes from, and it is roughly a year behind by construction.

The preliminary counter exists but will not answer for mining. Safe Work Australia does publish a preliminary fatalities page for the current year, which is the fastest national figure it produces. Three things limit it. It is a dashboard embed rather than a published number, so there is nothing to cite from the page itself. Its own notes say the dashboards "employ suppression techniques for small and/or unreliable results", which is exactly the situation a single industry's handful of deaths creates. And the page states the preliminary count is "drawn from initial media reports" and is "usually less than the number of finalised fatalities", so even a visible figure would be a floor rather than a total.

State safety alerts are not a death register, and counting them would be wrong. The NSW Resources Regulator publishes safety alerts, and it is tempting to treat the series as a fatality list. It is not. The regulator's own description is that an alert is issued following "the occurrence of an event such as a fatality, dangerous occurrence or any incident that is considered to be of significance to the industry". Counting alerts would mix deaths with near misses and produce a number that looks authoritative and means nothing.

Queensland's alert database was not working when we checked. Resources Safety and Health Queensland runs what is normally the richest incident-notice database in the country. On 30 July 2026 its Safety and Health Alerts page returned a search error and zero results, and the page footer read "Last updated: 23 Oct 2024". We record that as an observation on one day, not a verdict on the agency.

What this means if you need the number

Our view, and it is a view. For a duty holder, an investor, a union or a journalist, the honest position today is that Australia's mining death toll is knowable about eighteen months after the fact and not before. That is a reasonable design for a statistical series that has to be verified, coronial findings and all, and it is a poor design for public accountability, because the period in which attention could change something is precisely the period with no published figure.

The workaround people actually use is to count media reports, which is what Safe Work Australia's own preliminary page does. That is defensible as an early indicator and it is not a substitute for a count. If you need a defensible current figure for a board paper or a submission, the defensible thing to write is that the most recent official count is 10, for 2024, and that no current-year figure is published.

We will update this page when the 2025 data lands, which on the current cadence means the next edition of the detailed data file and Key WHS Statistics, expected around October 2026.