The Bureau left grasping for answers.
In estimates I asked the Bureau of Meteorology how they reconciled their surface recordings with their satellite recordings against claims that they weren’t reconciled for the last five years.
I also asked if they kept two thermometers at every ground station in accordance with recommendations made back in 2011.
The idea behind two thermometers is if one breaks down or the two start to record different temperatures the Bureau has a backup.
I also asked, given the enormous size of Bureau’s budget if the CEO could give a break up of the staff that work on measuring actual observations against those that don’t.
The number of backflips in their answers would make a Romanian gymnast proud.
Environment and Communications Legislation Committee
24/02/2025
Estimates
CLIMATE CHANGE, ENERGY, THE ENVIRONMENT AND WATER PORTFOLIO
Bureau of Meteorology
Senator RENNICK: Hello, guys. How are you going?
Dr Johnson : Good, thanks.
Senator RENNICK: In regard to the solar management instruments having returned no data to stakeholders for almost five years, what is that in relation to? Is that measuring atmospheric temperature?
Dr Johnson : No.
Senator RENNICK: What’s a solar measurement instrument?
Dr Stone : Thanks for the question. If you like, the day-to-day solar measurements are done by satellite, and the ground stations are used for calibration and validation of the satellite images.
Senator RENNICK: Okay. So, when it says measurement instruments have returned no data, what does that mean? It’s returned no data to whom? You’re not recording it? What does it mean?
Dr Stone : The data that customers want is the stuff from the satellites, so that keeps coming. Then, as I said, there’s a smallish number of ground stations that you use to calibrate and validate the satellite data. Because the satellite calibration and validation is very stable, you don’t actually need a whole heap of ground measurements in order to make sure you’re getting the right data out of the satellites.
Senator RENNICK: You’ve still got me confused here. Are you telling me that you’re not reconciling the ground data to the satellite data?
Dr Stone : We do, but we don’t do it every day.
Senator RENNICK: But it’s saying here that you haven’t done it for five years. Is that not correct?
Dr Stone : It’s not quite correct, no.
Senator RENNICK: You don’t do it every day, but there’s a big difference between not doing it every day and not doing it for five years, so where’s the truth here? What is it?
Dr Stone : I’m not sure, to be frank. For the ground stations, the data is fundamentally used to calibrate and validate the satellites, and that function is effectively met.
Senator RENNICK: Okay, because that then comes back to a report that was done in 2011 about ways to improve your observation data, and it said you should have multiple thermometers at each weather station so you can check for redundancies. You have redundancy and you can calibrate between the thermometers. How many thermometers do you use on average at a weather station, just the one or two or three?
Dr Stone : I understand it is two.
Senator RENNICK: Are sure about that?
Dr Stone : No, I’m not.
Senator RENNICK: You’re not?
Dr Stone : We could ask or bring in Dr Karl Braganza.
Senator RENNICK: The reason I ask is because if this is the point: we need to know the margin of error on these figures because that same report in 2011 had a margin of error plus or minus half a degree, so if we’re talking a whole degree—
Dr Stone : We’re talking lots and lots of different things here—loads of different things. The satellite data that I referred to for measuring solar insolation across the whole country actually tells you at any point that you would like to know what the solar radiation hitting the landscape is. If you have solar stations you only get a measurement where that station is. They are there for the purpose of calibrating, if you like. The satellite just gives you a bunch of numbers and you need to turn those numbers into a solar energy figure, and that is what the ground stations are used for. Temperature data is a completely different thing.
Senator RENNICK: This article in the Saturday Paper seems to imply that spending and delivery of quality assurance on actual outcomes, measuring data, are being underdone. My question to you is: what ratio of your employees at the Bureau of Meteorology are involved in recording the weather and actually maintaining quality assurance over actual observations versus the number of people in the climate division that model and do things like homogenisation? Do you have an overview of the breakup between your staff and the money that is spent between these two? Because one is modelling and one is actual.
Dr Johnson : Maybe I can make a general statement. We have many hundreds of people in our data and digital group. Ms Brinsmead, who is online, can give you the exact numbers of people involved in maintenance.
Senator RENNICK: Is that data and digital group an actual?
Dr Johnson : Peter can answer this in a minute, but we spend more on maintenance on our observing assets in the bureau at the moment than we have at any time since I’ve been here.
Senator RENNICK: My question here is: what is the ratio between maintenance and actual upkeep of the instruments that record?
Dr Johnson : We would have to get you the numbers on that. But we have a very small team. Peter might want to elaborate here.
Dr Stone : On the measurement accuracy for temperature, to be clear, I know in the past we have discussed measurement tolerance of the equipment. That is the maximum allowable error. It is not the actual error; it is the maximum that is allowed. When we actually look at the uncertainty of temperature measurement, it is very low. We estimate that to be 0.04 of a degree—sorry, 0.084 of a degree.
Senator RENNICK: Is it because you have two thermometers at each weather station?
Dr Stone : They get calibrated at least twice a year.
Senator RENNICK: But that is the point: to calibrate, you have to match one against the other, because you have nothing to calibrate against.
Dr Stone : You don’t calibrate thermometers by having two out in the field. They could both drift, so you have reference thermometers that you use. Our technicians go out and actually calibrate that the thermometers used—
Senator RENNICK: It was the recommendation in the peer review back in 2011—
Dr Stone : That has been met.
Senator RENNICK: that you had two thermometers—one for redundancy if one breaks down. It says here that 10 per cent of planned maintenance doesn’t not take place, that 43 per cent of these incidences did not meet agency targets mean time restore, which means that, if something breaks down, you have gaps in your data series. That was why the report was pushing for two thermometers at every weather station.
Dr Stone : With that independent review from 2011, we have met all but one of the recommendations there and that is the one around digitisation.
Senator RENNICK: So if you met them all you would have two thermometers per weather station. Is that correct? How come you didn’t know that before?
Dr Stone : I’ve answered the question in a different way. I know we have met all except one of the requirements in the independent review.
Senator RENNICK: Okay. I don’t want to take up too much time. You’re the CEO, Dr Johnson. You do, I presume, monthly management accounts here, where you’re looking at your front-end expenditure on actual upkeep of collecting real data, and then you have your model data, your homogenised data, that is based off the real dataset. Your model data is only going to be as good as the real data that you collect—garbage in, garbage out, right? So what is the proportion of expenditure between actual upkeep at the Bureau of Meteorology and what I call homogenised—
Dr Johnson : We’ll take that on notice.
Senator RENNICK: But wouldn’t you have a rough idea of that in your head?
Dr Johnson : We’re talking about a $700 million business. I don’t have the exact numbers. I’d rather give you a correct answer than speculate here. But, as I said in my general comments, we have a very large team—an excellent team, I might add, of dedicated people—
Senator RENNICK: I’m sure they are.
Dr Johnson : who sustain our assets at a very high level of performance and a very small team that deals with the issues that you’re referring to. I don’t know the exact ratio, but we’ll take it on notice and we’ll get it for you.
Senator RENNICK: How many employees are there all-up at the bureau?
Dr Johnson : Right now, we have 1,722 in our ASL. I think that number is at December.
Senator RENNICK: And what’s your annual budget?
Dr Johnson : It depends on how you count it, but it’s between $700 million and $800 million, depending on which team you use.
Senator RENNICK: Okay. And most of that ROBUST project is on the actual collection—
Dr Johnson : ROBUST has closed. It closed last 30 June. ROBUST no longer exists.
Senator RENNICK: Okay, but the purpose of that was to upgrade actual—
Dr Johnson : No. The purpose of ROBUST was to look at the security, stability and resilience of the bureau’s IT and OT, so—
Senator RENNICK: So that was back-end expenditure?
Dr Johnson : Sorry, I’m not sure what you mean by ‘back-end expenditure’.
Senator RENNICK: Your front end is recording all the data, and then you’re storing it in IT centres—
Dr Johnson : The bureau has a whole range of information and observing technologies. ROBUST was designed to be a comprehensive injection to make those systems secure, stable and resilient, and it funded a whole range of activities in the bureau, both operating expenditure and capital expenditure. That program has finished, and—going back to the questions from Senator Davey and Senator Pocock—in the 2021 budget we were appropriated funds to sustain the benefits of the ROBUST program. The government of the day took the view that the reason ROBUST was needed in the first place was many, many years of underinvestment in the bureau. The investment in sustainment funding was to, hopefully, avoid being in the same situation that we found ourselves in in late 2015 and early 2016, where the bureau’s systems had been compromised by a state actor and we’d had—I wasn’t there at the time—apparently, a number of major system outages. That’s what ROBUST was for and what the sustainment money is for.
Senator RENNICK: No worries, okay.
