Six Point Media Overview and Q&A

3 point overview

  1. Yes, satellites show basin-scale drying in the Cambrian Limestone Aquifer between 2014 and 2022 — no surprise, those were dry years that included some of the driest on record (2019,2020). But the study stops in 2022 and leaves out the most recent years, which show a rebound in total water storage, taking us back to where we were before the dry run. It’s like watching a few minutes of a the movie, finishing before the twist and saying you know who did it.

  2. The authors claim in media releases and interviews that groundwater pumping is a significant driver of this change and NT Government water regulations need an overhaul, but they have no evidence for this. They have used GRACE satellite data, which measures total water storage over a huge area, to make claims about local groundwater extraction without any evidence of the actual use, which is too small for the satellites to detect.

  3. Think about it this way: the GRACE satellite measurements are like a monthly bank balance for the whole basin; GRACE data doesn’t itemise transactions, let alone work out who spent the money, where or on what.

3 extra detail points

  1. They quote big-sounding losses — six to eight cubic kilometres a year in terrestrial water storage and nearly four cubic kilometers in ground water storage — but spread over the aquifer’s total area that’s less than depth of your thumbnail, around a centimetre of water lost in a year. Even so, these loses are about 270 times larger than actual metered extractions and estimates for stock and domestic. In fact, total extraction is about 10 to 25 times smaller than the uncertainty in the GRACE measurements the authors published for their own study. Basically, they claim they can hear a whisper behind the stage at a rock concert.

  2. The GRACE satellites are great for tracking drought over large areas, for looking at groundwater at a basin-scale, or for tracking ice sheet loss over Greenland and Antartica, but they are the wrong tool for attributing Cambrian Limestone Aquifer drying to pumping at the rates of use we see in the Northern Territory. During the Millennium Drought down south, the same satellites saw huge water deficits from the drought in the Murray-Darling Basin and still didn’t isolate an irrigation signal. If satellites didn’t see groundwater pumping in the Murray-Darling, how will they see it up here?

  3. The study’s work is over a short time period. Twenty years is not long given the variabilty in climates across the north. We would expect to see natural declines and rises in aquifer volumes over 20 years. Pulling out a trend that starts when the aquifer is high and ending it when the aquifer is low, which is what the study does, tells us nothing about the impacts of extraction, which are too small to see over natural variability over such a big area – from Katherine into western Queensland.

Concluding message

  1. The authors of this study have told us what we already knew: 2014-2022 was a dry period. They think because government granted licences to extract water around the same time, licence issue, not actual use, drove changes we already knew were happening because of natural variations in climate. The claims don’t match the facts, but I guess it’s a good story for people who have a negative view of water management in the Territory.

Q&A — “So is the aquifer drying, and is pumping to blame?”

  • Drying trend 2014 - 2022: yes.
  • Blame: not shown; wrong tools; GRACE measures total storage over a huge area; it doesn’t label causes.
  • Scale check: losses 6–8 km³/yr; method uncertainty ~0.3–0.75 km³/yr; use ~0.03 km³/yr (30 GL). The pumping signal is 10x smaller than the noise.
  • What’s terrestrial water storage vs groundwater storage? The GRACE satellites can only measure changes in Terrestrial Water Storage that is the total amount of water stored on and under the land — that includes rivers, lakes, soil moisture, snow, ice, and groundwater combined. Groundwater storage is just one part of that total — the water held in underground aquifers. To get groundwater alone, scientists subtract estimates of the other parts, like soil moisture and surface water, from the total GRACE provides. This increases the uncertainty in the measurements.
  • Bridge: Nothing I’m saying rules out local effects of pumping near big users — I’m just confirming that satellites can’t detect them at basin scale.

Q&A - “Don’t the authors also claim surface water has gone ‘permanently’?”

  • We’ve seen that in some of the media reporting. The authors surface water analysis is based on a dataset that ends in 2021 and they cherry-pick years to compare to maximise the “loss”. There are datasets that extend to 2025 and given the overall rebound in total water storage, no one can say that surface water has gone “permanently.” It’s like showing a single photo of a river in the dry season and claiming the river is always dry — it ignores big changes that occur in our systems over time.

Q&A - “Don’t the authors also claim the aquifer is taking longer to recover?”

  • Yes. This is based in their analysis of the dry period between 2011 and 2022. The GRACE data from 2022 to now show how quickly terrestrial water storage recovers. It’s likely we’d see the same pattern in the aquifer if we had more data.

Q&A “You say they stopped short. What changes if you include 2022–2025?”

  • The picture changes: post-2022, total water storage rebounds.
  • “Accelerating drying” wasn’t tested, and the omitted years cut against it.
  • Analogy: they’ve ended the movie three minutes before the plot twist and claim they know who did it.
  • Surface water: their maps are single-year snapshots with a global threshold — good for context, not proof of “permanent loss.”

Q&A “What evidence would you accept before blaming extraction?”

  • Use the full satellite record to 2025, then add:

    1. metered extraction by sub-region (actual use, not allocations),
    2. climate controls (rain, PET, ENSO/IOD) with sensible lags,
    3. local corroboration (groundwater levels near production bores, baseflow in springs/streams),
    4. a detectability test — can a ~0.03 km³/yr signal rise above the noise of ~0.3–0.75 km³/yr uncertainty? It’s like trying to hear a whisper coming from behind the stage at a rock concert.

  • Bottom line: until that’s done, saying “extraction is a significant driver” exceeds what the data can show; at present volumes any effect is below detection. Claims that the Cambrian Limestone Aquifer in the Northern Territory is drying out and that groundwater pumping is a significant driver of this change are based on no proper evidence. The scales are all wrong. It’s just not possible for the current level of extraction to provide any sign of impact at a basin scale amid the noise of the high climate variabilty we have in the NT.