Scientists fight for climate records as snow data on thin ice

There’s an adage in economics: you value what you measure.  So what would it say about how we value Australia’s alpine country if we stopped regularly measuring – say – snow?

An alarm was raised during the 2023 winter when scientists noticed a particular snow measure was becoming more … erratic.  At issue was the Spencers Creek snow course: a hitherto generally weekly and publicly available measure of snow depth near Perisher in the Snowy Mountains, which had been run by Snowy Hydro since 1954.  

Ken Green, legend of Alpine research, was among the first to notice the inconsistency.  “I hadn’t looked at the data for a couple of years and when I did I noticed they’d missed a few – in one case there was a gap of 4 weeks.” Four weeks is a long time in Australia’s short snow season.

Further analysis revealed that since 2003 the number of measurements had roughly halved, recently dropping to fortnightly or less.   

Soon, a petition with over 250 signatures was presented by the mountain-lovers’ climate advocacy group, Protect Our Winters, begging the measure’s consistency be restored. Scientists too fired off agitated letters. 

One scientist was moved to near tears, and another to exclaimed: “I’d defend long term records of anything to do with climate with my life!”

But an ambiguous response from Snowy Hydro just fuelled their concerns.

In a season of incredibly early snowmelt big questions were raised: what value do we place on rare long term climate records? In fact in an era when everything seems to be contested, why do we value data at all?

One scientist was moved to near tears, and another to exclaimed: ‘I’d defend long term records of anything to do with climate with my life!’

When the Spencers Creek snow course was established in the 1950’s, it was just one of many measuring sites designed to assess water runoff into the Snowy Hydro system. The method is straightforward; hydrographers plunge a hollow metal tube into the snow until they hit the ground, pull out a snow “core” which gives them snow depth and weight, then average the readings from 7 set positions.  Over time, most of the sites lost importance, leaving only three key locations which survive to this day: 3 Mile Dam at 1460m above sea level, Deep Creek at 1620m, and Spencers Creek at 1830m. 

Since the 1950s, the snowline – the area which remains consistently under snow for winter – has moved up.  Which means Spencers Creek is the last remaining site that best reflects coverage over the whole snow season.  Skiers and back country enthusiasts started relying on it for snow conditions, alpine scientists for their work, the Bureau of Meteorology for its State of the Climate Report.  Through an accident of history, Snowy Hydro ended up custodian of the only long term record of snow in this country.  

As Dr John Morgan, Director of the Research Centre for Applied Alpine Ecology says, “It would be like having just one temperature record of the Great Barrier Reef! It’s a unique situation – that something as important as the Australian Alps is being measured only in one place.” 

The common thread for all these scientists is what the Spencers Creek records allow them to do – benchmark measures into a long term time line.  Even Victorian alpine scientists rely on them.  Because, while there are Victorian snow measures, as Professor John Morgan explains, they are hard to access and less consistent than the publicly availably Snowy Hydro measures. “Without historical records we can’t see what’s changed in the seasons, and that helps explain everything about what’s happening to snow gums, for example” he says. 

Spencers Creek is the last remaining site that best reflects coverage over the whole snow season

Professor Neville Nicholls, eminent climate scientist and the man who exclaimed so flamboyantly he’d “give his life for long term climate records”, credits historical weather data for allowing scientists to predict El Niños.  And he’s published papers using the Spencers Creek data.  “That’s how we were able to show that while there’s been some decline in maximum snow depth, the biggest decline is in spring. Snow melts faster.  You need to know that detail not only to track global warming, but to adapt. ”

And perhaps nowhere is this more crucial than for trying to understand the future of some of Australia’s most important rivers.  

The Snowy Mountains are the major headwaters of three of Australia’s famous rivers.  All of the Snowy River, much of the Murrimbidgee, and 20-30% of the lower Murray flows from this region.  “Obviously the snowpack is a big part of how that works,” says Dr Duanne White.  “How much this changes in future will have a big influence on how much water we get out of the mountains into the rivers.”

So what’s at stake: “How about the future of the Murray Darling Basin!,” says White. “Look how important that is to towns and agriculture and ecosystems.  In the future, we’re looking at 20% -40% less runoff from the mountains into the lower Murray Darling.”

Dr White is part of the Australian Mountain Research Facility, which has recently set up 9 major recording sites – some automated – across alpine areas.  But they also use Spencers Creek records to put the findings into context.   “Without those continuous records, our future modelling of river flows will be far less reliable ,” says Dr White.  “Bottom line – it will be much harder to predict what will happen to our rivers.”

 Without historical records we can’t see what’s changed in the seasons, and that helps explain everything about what’s happening to snow gums, for example

Dr John Morgan, Director of the Research Centre for Applied Alpine Ecology

What this whole investigation has highlighted for me so far is the potential fragility of some of our irreplaceable long term records.

Take rainfall.  Back in the day, many farmers began sharing their detailed records with the Bureau of Meteorology (BoM).  They were a point of pride, the responsibility passed down through generations.   But as farms were sold or amalgamated, some of these so called “observational partners” dropped away.  

Similarly with temperature; the BoM itself rolled out a number of stations in 1910, of which only 61 survive.  “They were set up for weather forecasting,” explains Blair Trewin, leader of historical record analysis for the BOM, “it’s only since the 1990s we realised their importance for climate change.”

So significant are these rare records that have survived the vagaries of history that the World Meteorological Organisation set up a system of recognition for what it calls “Centennial Observing Stations”, stating: “Long term meteorological observations are part of the irreplaceable cultural and scientific heritage that serves the needs of current and future generations for long term, high quality climate records.”  Australia has 11 such records recognised by the WMO; 8 standard, 2 rainfall and one tidal.  

But only one long term continuous snow record.  

Even with modern satellites “Ground Truth” data – in situ observations – help scientists to be certain the satellite data matches reality.

“If they screw around with this too much we won’t be able to do this work in the coming decades,” says Nicholls. “Honestly, in an age where we are so worried about climate change and the future, the idea anyone would jeopardise one of these pure gold long term records is beyond my understanding.”

Environmental engineer, Jeremy Kinley is acting manager of water for Snowy Hydro, responsible for a hydrography and water modelling team of 12. “The feedback we’ve been receiving has been fairly consistent,” he acknowledges wryly.  “The scientific community has concerns, and the ski community are keen to get more measures for conditions on the ground.”  

Getting clarification on what changed in recent years isn’t easy, but health and safety is one key factor.  As Kinley points out, trailering two heavy skidoos up to Perisher and driving them out to the snow course with variable weather and snow conditions isn’t without hazards.  “Safety of our staff is our number one priority,” he says, “if the risk to safety is too high, we don’t go.”  

“We recognise that they’ve become the primary long term record of the Australian snowpack and I guess we have assumed custodianship of this highly valued public data set.  It is a public good: we are listening and have heard calls for more frequent measurements.  

“And so while there isn’t any operational need for us to increase the frequency of measurements, we have agreed that we will commit to more regular measurements at Spencers Creek from next year onward.”  

“We’re still working on the details,” he says, “and safety still comes first, but you can quote me on that commitment.”