Australia Winter 2019 Seasonal Outlook

Hello I’m the Snowy Hibbo (Zac), and welcome back to the Australian season of long range snow outlooks. Being in Melbourne, it has certainly become cooler over the past few weeks. I will start the season outlook with the typical model analysis as all the seasonal outlooks tend to feature. But of course I will then proceed to go into a deep analysis of the climate drivers that help us forecast the weather and climate. Firstly though, I wish to partake in some self-reflection.


Last year’s forecast was for 182cm, and the Spencer’s Creek snow depth which this is based upon, ended up being 224.6cm. That’s more than 40cm off, but at least I was on the right track with a snowier than average season (but obviously a lot more than I said).

My last two seasonal outlooks have been rendered relatively poor, due to the large outliers we have had, which nobody managed to fully predict. It is still a blunder on my part, and because of that, I have updated some of my methodologies behind the scenes, while understanding some of the limitations of the format I use. On that note, we shall begin…


Firstly we have the EUROSIPS, which contains a variety of European models, including the two best models in the world, ECMWF and the Met Office UKMO model. It certainly doesn’t paint a good picture for our winter, with lots of ridging in the Australian Alps region, leaving less room for snowfalls.

CFS also shows ridging over us for the winter, but focuses the stronger ridging over SW WA and towards New Zealand, which means that we would be in the break between these two major ridges, where lows could squeeze in and deliver snowfall.

CanSIPS, the Canadian model, shows neutral anomalies across the Australian Alps, indicating a neutral outlook for snowfall prospects from this model.

Worthy of note, is the NMME model, the combination of North American models like the CFS and CanSIPS. It is important to note the above CFS and CanSIPS are not represented in the below, because it is an old version from last month.

NMME shows mildly-moderately warmer temperatures over the Australian Alps, reducing our chances for snowfall.

Climate Drivers

Firstly we have our 10 year average (2009-2018) of 186cm. That is 13cm higher than the 173cm average of 2007-2016 I used last year. Two very strong snowfall years is obviously going to skew the dataset as expected with a relatively small range of depths. Regardless I will continue the analysis with those numbers in mind.

PDO: There is a weak correlation between a positive PDO and lower rainfall in our region. This decreases moisture for snowfall, which is supported bythis graph created by Gerg. We are currently in a neutral PDO phase, but ECMWF is forecasting a shift towards a positive PDO. Therefore I am going to remove two centimetres. 186-2 = 184cm

SAM: There is a fairly strong correlation between negative SAM and more snowfall for the Alps. I talked about the seasonal prediction of SAM here. El Niño Modoki (the only seasonal prediction factor of SAM, other than a weaker correlation to Eastern Pacific Nino) is at a Warm Neutral phase, so no forecasted effect on SAM(on the JAMSTEC model). Adding in climate change(that increases chance of positive SAM), I am going to cut off 3cm. 184-3 = 181cm

IOD: There is a correlation between the Australian snowfall and the IOD. There is a Positive IOD forecast for later this winter by some models, others are just warm neutral throughout our winter. This isn’t great for our snow later this season, but there is perhaps a conversation that to be had whether ENSO and the MJO dictate the IOD’s impact. I am going to remove 4cm here. 181-4=177cm

ENSO: There is a correlation between ENSO and Australian snowfall. The current model predictions show a fairly weak El Niño at the moment and into winter. This would make our snowfall season a bit weaker, but we are seeing some signs of a progression towards a more neutral ENSO winter. SOI and trade winds in the Pacific are neutralising, and the Atmospheric Angular Momentum (more here) indicator is become more neutral after a long period of strong positive cycles. I predict a neutral Atmospheric ENSO, and perhaps a weak oceanic Nino. This will be slightly bad for Australian snowfall, so I will remove 2cm here. 177-2=175cm

MJO: Australian snowfall and MJO have a very strong correlation. The CFS model shows a Phase 8 MJO. This isn’t good for Australian snowfall prospects, but at least there is a reasonably strong MJO going around on the model. The decrease in GLAAM/GWO positive cycles (see link in ENSO section for more info) means that the MJO is probably going to be around less than normal and weaker than typical. If there is some sudden boost towards an Atmospheric Nino, strong MJOs in our region could be more regular, but I don’t think it will happen. I am going to take off 5cm here, 175-4 = 171cm

Sunspots: There is a weak correlation between a higher amount of sunspots and more snow, as found by Francisco Sánchez-Bayo and Ken Green. We are at the solar minimum, so I am going to cut off 2cm. 171-2 = 169cm

Volcanoes: If a large stratospheric volcano explosion occurs, Australian snowfall substantially increases for that winter. None have occurred in the last year, so our forecast won’t be affected. 169-0 = 169cm.

SST anomalies: Based on current charts, the Bight is absolutely perfect(nice and cold). ECMWF shows warm anomalies in the Tasman, which increases moisture, but reduces cold rapidly in a double edged sword scenario. Gerg has a graph for this(Tasman SSTs). I think it is in our better interests to get this moisture from the NW of Australia through MJO or other wave activity. This is improved with higher than normal SSTs north and northwest of Australia, which we don’t have. With a weaker than normal MJO and a increasingly +IOD, I think the Tasman positive SSTs are a necessary evil to inject moisture into our systems. ECMWF also forecasts a bit of a weakening of the cold SSTs in the Bight, but it should stay relatively cold, given the support of models like CFS and CanSIPS through winter. Thanks to these factors, I will add 7cm to the final tally, 169+7 = 176cm

So my final maximum Spencer’s Creek Snow Depth for this year is 176cm. This number has came around from hours of analysis, tinkering and changing. The model outlook in the first section of the blog, is certainly in favour of a weaker snowfall season. The 176cm figure is just higher than the 10 Year figure from 2007-2016 of 173cm mentioned above. It is however below the current 10 year average of 186cm. Taking the recent outlier years into account, my forecast is for an average snow season.

There have been various calls for some sort of “when is the most snow going to fall” forecast. Based on the tropical signals and the Indian Ocean, I tend to think that the season will be relatively early. I am not going to talk months or specifics, I try to keep as wide of a span as possible with this matter.

I hope you enjoyed this analysis, and I will see you for a long range outlook before opening weekend.

Thanks again for reading this Australian snow season outlook, follow me on Twitter @longrangesnow and subscribe to my email list by clicking on the tab on the main header above.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s