Transcript of the weather forecasts and seasonal outlooks and climate change projections webinar

Heather Field:

Thank you everyone, and welcome to today's webinar, where we'll hear from Graeme Anderson on important background on weather forecasts, seasonal outlooks and climate change predictions, and what they can and can't do. Graeme Anderson will share some insights and tips for understanding the different types of forecast products, and where and when to have more confidence in what they are telling us. I'll now pass over to Graeme Anderson from Agriculture Victoria.

Graeme Anderson:

Thank you very much Heather, you doing a great job with these AgVic webinars. We've got a bunch of good speakers on the way as well and they're all recorded there so everyone can catch up if you happen to miss whose speaking. So I'm just going to kick things off. Just standby, [inaudible]. But basically we get a lot of questions and people looking at seasonal forecasts and looking at weather forecasts and climate change projections. So what I've got together here is a bit of a selection of learnings from the last 15 years of running through a lot of our climate sessions and working closely with forecasters to find out what's happening. Now Heather, I've got a screen freeze down this end. So not sure if I just might...

Heather Field:

Do you want me to bring up your PowerPoint?

Graeme Anderson:

Yeah, I'll just [inaudible 00:01:51] sharing for tick.

Heather Field:

Okay.

Graeme Anderson:

Go back into my PowerPoint mode. That again, click it off. So is that there Heather?

Heather Field:

Not yet.

Graeme Anderson:

Not yet.

Heather Field:

You want me to just click and share your screen? [crosstalk 00:02:44]. If you just make that into presentation mode.

Graeme Anderson:

All right, so you should come with a sort of journey to explore, making sense of forecasts and models, and basically some of you might already know the team in AgVic and Dale Grey produces The Break, which has regular monthly commentary on seasonal forecasts. We've got Dale Boyd doing the soil moisture updates, really useful. There is a lot of forums that we speak to around the state and there's a climate webinar series which are happening, and thanks for organising them, Heather. And also, I'll probably refer to a couple of the projects. The work we're doing with GRDC South on using seasonal forecasts across Vic, SA and Tassie. And also Forewarned is Forearmed project as well. So, a lot of discussion is around models. And I guess it's wise to be cautious about models and one of the important things is knowing what's in them and what's not.

Graeme Anderson:

So often like to use an example as models that that farmers might have in their head, if someone says, ‘Oh, that particular paddock, what sort of wheat yield might you be expecting from that.’ Now, there's a lot of data that farmers uses because they know that actually, ‘I do know roughly what sort of wheat yield we would expect out of that paddock.’ And they know that based on history, because they've seen it, they experienced it. It's built on facts of knowing what's happened there in the past, I know what the range is of rainfall. I know the well-drained bits of the paddock. I know it can change a bit with fertility, but those previous seasons have locked-in and so that model in farmer's mind is based on history and saying that yep I know we could grow four tonne a hectare.

Graeme Anderson:

But farmers also know there's possible curveballs, such as sowing issues or disease pests that pop up, you know there is weeds, frost, heat, drought or herbicide issues. All those sorts of things or of course neighbours stock get in. So yes, there's a model there about knowing what you would expect but also, what can go wrong. And I guess one of the things with all models, it doesn't matter whether it's crop models, or climate models...there's key rules are that models are built around key rules effects that underpin, and they are based on history, so they're knowns. But a key bit when you're looking at model outputs. And certainly a key bit if you're trying to get confidence is does the model know about these particular other things or potential curveballs.

Graeme Anderson:

So that's what we're going to run through for weather forecasts and seasonal forecasts. And a key bit to forecasting is, understanding where forecasts have confidence and where they don't, also understanding the key assumptions underneath them, and what they do and don't consider. And often we farmers run the business out in the weather, always looking for the right amount of rain, but sometimes knowing the uncertainty can be more important than actually knowing the amount of rain that's predicted. And last week's weather event was a good case in point. So we're going to run through some of this with forecasts. And remember that these forecasts outputs are only one input into risk or opportunity analysis. It's combined without the info to help make good decisions and really they help us map out our scenarios.

Graeme Anderson:

I do get asked this quite a bit. If you can't accurately predict the weather in three months’ time, then how can you even try and predict 30 years ahead? So that's a pretty fair question. So come with me on a journey to try and tease that out. So I'm going to start with weather forecasts. And this is just looking at weather forecast models out to eight days and the weather forecasts out to seven, eight, nine days they're called deterministic forecasts where basically the models are run and they coming back with you with a deterministic forecast. I bet on day three, we're expecting temperature to be X and rainfall to be that. But it's interesting that no one has a perfect view of the future. And here's an example and I really like this. An example from a rain event.

Graeme Anderson:

It's from Bureau of Meteorology, and sometimes for some of the big weather events and I like how they're doing this, is they're putting up a few different model versions of what might happen. And this is just looking at over a four day period, but that you can see for this particular rain event, it's got the Bureau's access weather forecast model at the top. But then it's also got that a U.S. model on the bottom left, and then the European model on the right. I'm just showing that these are all incredibly powerful, really good models, but sometimes with particular weather events, they don't all exactly agree on where bulk of that rain might fall out. And so that's important because there is no model in the world that actually gets it right all of the time. They all have particular events, one of them will be right and and the others will miss it.

Graeme Anderson:

So that's an important thing that when you're getting a forecast, just understanding that it's one of the versions of what people think will happen in the future. Here's another example. And it's got a range of different models there, you can see each lines is a different model and it's looking over a week. And it's from a U.S. forecast site. But it's showing some different global models, just looking at for 10 days ahead. And you can see there's quite a range there, depending on the site here, could be expecting anywhere from one to seventeen millimetres. And depending on which phone app farmers are using, you could be getting any one of these models and what's really interesting now is that it is terrific that we've got phone and iPad apps that we can get forecast apps and models from anywhere.

Graeme Anderson:

But it's also quite common now that probably half of the farmers I speak to, they think they're getting the Bureau of Meteorology’s forecast, but actually, the models that underpin the particular app they're using is from other parts of the world. Now there's nothing wrong with that. It's just important to know which model are you getting. So two farmers can be standing at a fence, and one expecting, 18 mil and another expecting one and they're thinking what the hell's going on. So that's a key bit, is just trying to understand which model is your forecast you looking at coming from. And also sometimes for some weather events, that will vary a bit. And then there are other times when they'll all be sort of focusing in on a particular rain event where they might say most of the models are thinking between sort of 20 and 30 millimetres, so that gives you more confidence.

Graeme Anderson:

There are a few products that I use a bit. The chat viewer that the Bureau's got is a really good one just to look at how they access model plays out for the next seven days. The water and land products been up there with the Bureau for a while and it gives rain outlooks but it does a bit of clever work and then it brings together six global models and puts them all together. So it gives you a bit of shandy of what's expected across models. And the Bom Met Eye is obviously great too. I've got a thread on Twitter of a bunch of different rainfall outlooks, so you can go and play around with some of them there. The main thing is you understand what you're viewing and some different models and apps have a better way of showing the data than others, but also just learn a bit more to understand where which model is this forecast getting it from.

Graeme Anderson:

So and for those who haven't been to Twitter. Twitter's just basically website. So if you Google, @climatedogs, you'll find that tweet there, and there's a bunch of links, you can go and test and have a look through. Well, there's a bunch of good weather forecast sites out there. It's really important to know that if there for any extreme weather events happening in Australia, the Bureau of Meteorology is to go to source because there's a lot more human forecasters that are looking and they're looking at a range of models, and they're providing you with that latest up to date info. So when there's extreme weather events forecast, don't just rely on a model from an overseas app, you really need to tune in and get the good oil from a local forecasters. So that's sort of it about the weather forecast. Next eight days.

Graeme Anderson:

The highest confidence is around the closer, the more confidence day once more accurate than day eight, but they can vary between model runs and between individual events. And also just beware curveballs such as thunderstorms, or storm lotto as farmers call it, an [extreme 00:11:29], because while a forecast can predict, say tomorrow might have thunderstorm conditions in the afternoon. The actual location of thunderstorms isn't really known until they are actually just appearing. So you might only have an hour or two notice of that. So it's really important to know, and I know especially with some of our summer thunderstorm weather we can get some quite big falls on that time. People say, ‘Well that wasn't forecast.’ Well, We can predict thunderstormy weather and the conditions for it, but actually who is underneath those thunderstorms. That's down to good old storm lotto.

Graeme Anderson:

So seasonal forecasts in the next three months are a really important tool, but also there's a big transition when we go from a weather forecast to seasonal forecasts. And climate outlooks, basically, they're still using weather forecasts, but climate models just run out for a few extra months. And there's many different sort of starting values and conditions. And for example, I might run a weather model out, give it 100 runs for the next three months and if there's 100 runs and for your particular location, 70 runs are dryer and 30 of those runs are wetter, that's where you get this above or below median sort of discussion where it might say 30 per cent chance of exceeding the median. So the important bit is of seasonal forecasts, [inaudible] it's all to do with what oceans and everything are up to but no one actually knows what's going to happen.

Graeme Anderson:

So it's no different to a forecast of the share market, or forecast of urea prices or grain prices. No one actually knows what's going to happen there, then sometimes is more predictability than others. And I'll talk about when that might be for forecast. But this is a really big change in that once you get beyond that sort of 8-Day Forecast, and we're talking, out into months ahead, we're talking about probabilities. So pretty well you've got to plan for anything could happen, but sometimes there's a stronger slant, depending on some big Ocean Drive is pushing it. So here's an example going back to July 2019. It's from the Bureau's Outlook, the three month outlook and people are probably familiar with this sort of thing where you're hearing a probability or a chance of above median rainfall for the three months ahead.

Graeme Anderson:

So you can see these sort of maps here. And you can see at that time, last year there was a dry sort of pattern that an increased or reduced chance of exceeding median rainfall. And if you look at somewhere like Bendigo you might have a 30 per cent chance of exceeding median rainfall. So that's a drier pattern. But also what does that mean? That it's important to understand the detail of what sits behind these forecasts. There's been a lot of work done, there's a project underway at the moment, called Forewarned is Forearmed - it's developing new forecast products, it's a huge collaboration across Australia - project led by MLA and you can see heap of partners that are involved in that and with funding from the Australian Government's Rural R & D for Profit programme.

Graeme Anderson:

So we're going to run through some of this work there because there's quite a bit of work working closely with the Bureau of Meteorology on... Is there ways we can improve understanding some of these forecasts. So here's a bit of an example. So for that forecast, back In July, last year, you can sort of see that on the left, the chance or the chance of exceeding median, 35 per cent chance above median is the... I'll just get my pointer up. So you can see, and that indicates how a drier look, but should you just lock in it's going to be dry? Well, here's an example of what happens about...if you imagine that the model runs been run 100 times for the next three months. And here's how they fell for Bendigo. So while 35 per cent of the models were dryer, sorry, 35 per cent of the models were wetter and 65 are drier. This is actually how they're run.

Graeme Anderson:

And what I'll show is here, climatology, you just expect 20 per cent of the times you're in decile 1-2 and 20 per cent of the times you're in decile 3-4 and 5-6 and so 20 per cent in each of those two brackets. But when they've run the model for the three month outlook in this situation, what's actually happened is 33 of the models fell on decile 1-2, for that next three months, and 23 fell in decile 5-6 and eight models fell on decile 9-10. So often people can look at that thinking ‘Well, lock it in, it's going to be dry’. But it's really important to say there's quite a few models here talking to a better. But that is the true sort of range of that model output. So it's pretty important. This is the same data but just two different ways to express that information.

Graeme Anderson:

And a common one way that we get a lot of discussion is when we do happen to see a 50 per cent chance above median, which is the same as a 50 per cent chance of below median, and a lot of people saying Well, that's good, at least it'll be average. So should you be planning for average rainfall when you get that sort of an outlook and basically, no you shouldn't be planning for average - 50 per cent chance means if they've run the models, you can have a big range of split and you can actually end up having 20 per cent of models in decile 1-10 and 20 per cent in decile 9-10. So you can actually have all options are sitting there on the table for 50-50 forecasts. So you shouldn't be planning for average, it actually means that, well, there's no clear drivers behind that three month outlook, so you should be planning for either something happening at either end.

Graeme Anderson:

So that's pretty important because it doesn't mean they're confident about average in that situation. So one of the other products which are being developed and here's a different way of presenting that sort of stuff. This is from that same period for last year and this is rather than a percentage chance of above or below median. You can see here a test product from the Forewarned Forearmed project where they're saying, ‘Well, how about we just put on a map the probability of being in decile 1- per cent rainfall for the next three months.’ And you can see here for when you go back to last year for that September outlook, you can see that, normally this white areas where there's between a 10 and 30 per cent chance of decile 1-2. So when you start seeing these colours are lighting up, it's showing that there's an increased chance of decile 1-2.

Graeme Anderson:

Which is basically, for a lot of farmers that are more interested in trying to know when...tell us when the models are indicating a stronger sense of drier or a stronger sense of wetter. So these products are coming and a key bit that sits behind...when do forecasters have more confidence in climate models, usually when there's an active climate phase that's really underway. So this is sort of a drier looking pattern. If you look at this top map here, and we look at say, September 2006, these are are sea surface temperature maps and while the oceans a lot warmer along the equator, and cooler, down south, this is just showing anomalies where ocean might be warmer than it is normally, and the blue colours where air should be might be cooler than it is normally.

Graeme Anderson:

So this is a dry looking pattern, a strong El Nino in the Pacific, and you can see cooler sea surface temperatures to the north of Australia, which is usually followed by a distinct lack of cloud and moisture that then feeds into weather pattern. So when those sort of patterns are starting to form, and you run a weather model over that, for the next three months, there's a lot less moisture to play with, and that's why more of those model forecasts will come out drier. Turn to a wetter looking pattern. So October 2010, you can see a big La Nina and the IOD was in its wet phase, really warm oceans to our north, moisture galore, cloud galore in any weather pattern. There's plenty of moisture to feed in. So if you run a weather models for three months over that sort of a sea surface temperature pattern a lot of them are going to be showing more increased rainfall.

Graeme Anderson:

So these particular conditions though, only happen every now and then. So that's part of it - why some seasonal forecasts have more confidence than other seasons. Sometimes it's a bit more neutral and anything can happen, but probably useful to pay more attention, especially when you can start to see some of these climate drivers firing up. In the GRDC South project on using seasonal forecasts, we've teamed up with them and there's a new forecast tool, local climate tool at this address. This is just taking rainfall history from Dookie, looking back last 120 years, and we've just coloured in the years that were either in an active climate phase, whether there were neutral or La Nina, El Nino, or even combos. And it's quite interesting when you look at some of those really dry years at Dookie in the past that we use when an El Nino in the Pacific combined with a dry phase of the IOD which is positive the Indian Ocean Dipole.

Graeme Anderson:

So that's when there's a lot less moisture circulation and that's often those years when the seasonal forecasts have a bit more confidence about them. And you can see, 2016 we had the Indian Ocean Dipole and it's wetter phase, the wetter year. And then those last two years have been this Indian Ocean Dipole and its dry phases. So part of understanding season forecast is working out when those models have more confidence. So not every season is the same. And basically, usually winter in springs has a bit more confidence about these major climate drivers than there is in early autumn. So the current outlook, here is the Bureau's outlook at the minute, for the next three months you can see it's a bit more hopeful looking pattern than what we've certainly seen last year. But again, here's the above a chance of exceeding median rainfall.

Graeme Anderson:

So there's a few places there where it's just sitting on the increased chance of exceeding median rainfall. But again, if I take somebody like Bendigo. And you look at that for one Forewarned Forearmed test product, you can see there's a few more sitting in the decile sevens, eights, nines and 10s. But still, there's decile ones and twos in there. So if you run the model 100 times, thirteen still ended up in decile 1-2. But anyway, we'd prefer to see them slanting towards that wetter side and if you look at somewhere where it's white 50-50 like so you can still see, there's a split there across all of them all options are still possible. And of course, we're big fans of the work Dale Grey does with Ag Victoria and The Break newsletter. So it's an alternative way of looking at getting confidence out of these some of the these outlooks, Dale every month, looks at 12 different global forecast models from around the world, including the Bureau's.

Graeme Anderson:

And he just puts them on a nice simple little table saying how many of them for Victoria, how many of them are saying, pretty well neutral. How many jumped onto the dry side. And how many of jumped onto the wetter side. And this is the current one at the minute. So most of them sitting on the fence. Interesting last year in about August, it was the first time in The Breaks history of 13 years that all of the models that were looked at jumped the fence into the drier I think. So that's an example of when you can have a bit more confidence in forecasts how it looks when you start to see quite a number of them starting to pick it up. And also when you start to see those changes predicted from one month, you see them actually happening. You see cloud patterns start to change. They're times when you can have more confidence in the outlook.

Graeme Anderson:

There's also a lot of frustration because I know farmers prefer to have an emphatic forecast where someone will come along and say, it's definitely going to be a decile 9 in September. And we love in emphatic forecasts and it doesn't matter whether it's to do with share market forecasts or fuel oil price forecasts or whatever. But just know that, no one knows how the future is going to play out, so anyone that's giving you an emphatic forecast, just beware because it could be either from a crystal ball or just file it under astrology. And now there's a lot of people get quite excited about correlations and things from the past. They're really important to ask, where's the proof for causality? Correlation is one thing to prove the physics of how that particular correlation works. And that's where we rely a lot on peer review science, which is pretty hard process to get a paper approved.

Graeme Anderson:

So there we go. There's weather forecasts and now seasonal forecasts. So less confidence in specific outcomes. So we talked about probabilities, models on differ. They skew depending on factors like sea surface temperatures, and they're more accurate when there's a really strong climate driver phase underway, and they're helpful to consider scenarios but really, the forecast is a plan anything can still happen. So, climate change projection, so where are we with that? I read the climate science report back in 2003. And I had to read it twice because, while we watch weather patterns a lot it was amazing to hear about what's been measured in terms of what's happening in the atmosphere above our heads, and the oceans around Australia, and what sort of normal variability and also which bits are happening a bit differently.

Graeme Anderson:

So that was in the early days, and we're hearing a lot about El Nino Southern Oscillation. In some years its sending more moisture to us than other years. Indian Ocean Dipole and in some years it's not sending much moisture and other years, like 2016 it's sending us plenty, but also the Southern Annular Mode, really key driver with the roaring winds around the Southern Ocean and the conveyor belt of all of the cold fronts that flick Southern Australia. So I'm in a subtropical ridge, which is really the railway lines for where the high pressures come along and decide to sit. So our wetter and dry seasons are all clued in by what these individual drivers are up to in any particular year or season. And that's where the Climate Dogs analogies came out, they were basically a little science summaries and they're still around if you want to see them for a quick one-off.

Graeme Anderson:

But also one of the things is understanding the good old fashioned variability from these drivers, but also understanding that some of these key climate drivers are changing. They actually heading into new territory that they hadn't been in the past. And that's important for us to understand, because that defines what sort of seasons we're likely to have in the future. And one of the key things, the highest confidence around a lot of the climate change work is that around the temperature increases greenhouse gas emissions rise. So all of the climate models and we've got new modelling for Victoria, they are all consistent that as greenhouse gases rise, trap more heat will be... we're looking at a warmer future. There's also a bit of a range in there about some things and I'll talk about that. And I guess we're sort of at this period in time where everyone around the globe is being asked about, well, how much heat trapping gases we're going to put in the atmosphere?

Graeme Anderson:

Because that will determine how much heating we trap. And so the next couple of decades are pretty important time in terms of what a billion humans might do. And it's interesting this particular graph down here looks at climate modelling from a few scenarios, but out to a few hundred years. So just shows the different pathways of where we end up really is what happens in the coming decades because what we lock-in traps heat for hundreds of years ahead. So one question that is always asked is, well, where's the proof of greenhouse gases? Why does something like carbon dioxide which isn't an atoxic gas, we breathe it out. Why does it get such a bad rap? And this was a story told to me by the UK chief scientist, many moons ago. But it's a great little story back in the 1850s. There was work by scientists at the time. So this is the 1850s.

Graeme Anderson:

And they had observed that we get energy from the sun during the day, that warms up the Earth's surface, and they had calculated how much energy comes in. And then the sun sets and they had calculated that well, that energy that's in the land should just radiate back out to space overnight. And if it did, it should be over 30 degrees cooler every morning. So there was a bit of a mystery they're saying something is not letting all of that energy escape every night. So experiments and this is the John Tyndall experiment, which was saying we're going to get some of those gases from our atmosphere and check which ones are trapping heat because something's up there, it's invisible and it's trapping heat. So he got the gasses of the atmosphere and started passing heat through them and he started with the two most common gasses.

Graeme Anderson:

So what's over, 98 per cent of the atmosphere is just nitrogen gas or oxygen. So, he got rid of the impurities, just got nitrogen gas and oxygen gas and past the heat radiation and it slipped straight through them. So he concluded that if our atmosphere was made up of only nitrogen and oxygen gas then we would freeze every night. So okay, there must be something else. He went back to the impurities what he called and went to pass the heat through them and they absorbed all of the heat. And basically, those impurities was carbon dioxide, methane, water vapor, nitrous oxide, the greenhouse gases. So the physics of that has been known since the 1850s and the laws of physics haven't changed since then. The simple physics of it is, nitrogen gas, which doesn't trap heat is just N2, two simple nitrogen atoms stuck together.

Graeme Anderson:

Oxygen gas, which doesn't trap heat is too simple oxygen atoms stuck together. And the greenhouse gases, well, they're all more complex molecules, CO2, H20, CH4. They've just got more complex molecules and they're trap infrared radiation, which is heat. So that's been known and the whole theory was, well, if we put more greenhouse gases up there, essentially should start to trap more heat. And a key signal for climate change is that while we're seeing increased heat trapped on land, and in our oceans, the outer stratosphere is cooling. So we know it's not variations in the sun, because if it was the outer stratosphere would be warming as well. This is an inside job, we're changing the ratio of energy in and versus what energy escapes and the analogy I often users like two cars parked on a sunny day. They're both getting the same amount of energy from the sun, both dashboards are heating up.

Graeme Anderson:

But if one of those cars has the windows wound right up, then that car is going to be hotter when you jump into it later in the day, and that's because they're both getting the same energy, but the car of the window is up, is not letting out the energy as quickly as the car with the windows down. So the greenhouse gas store is largely about us winding the windows and have humans want to wind up those windows. So we've got an opportunity. We've wind the windows up about a quarter of the way and a few people telling us we probably might want to stop it there. So we were expecting, I guess a max all of the climate models if you put all greenhouse gases in them, they warm. Now this is temperature maps across Australia. Blue is cooler than average years, yellow is warmer than average, and the dark orange is the warmest on record.

Graeme Anderson:

So you can see while there's a lot of variability across our history, one of the key things is in recent trends we are seeing more of that warming. And this is just at the start. So these are the early days. And that was expected, and that's that's sort of what's already been happening. And we can see that this is just temperatures for south-eastern Australia for summer, autumn, winter, and spring. And basically looking at spring there you can sort of see, the last 10 years has been a number between one to almost three degrees warmer. So that's almost a month, summer coming a month earlier, the simple way of putting it, this temperature increase is sort of like squeezing in an extra month of summer. We'll still have rainfall variation from year to year. But this temperature trend, is the one that's sneaking up on us.

Graeme Anderson:

So in terms of climate change projections, high confidence in some elements. So the highest confidence is increasing heat and with that, [inaudible] shift of weather patterns. This happens in both hemispheres, if you warm-up the planet, the tropics expands a bit further either side of the equator. And then it nudges everything towards the poles. And that's certainly important. And it's why for in Australia, you see increased incidence of drier patterns in southern Australia in climate change projections because basically if you shift all of that disturbance from all of the storm tracks and in the fronal activity down south, if all of that hits into moisture, a few hundred kilometres further south, then that means there'll be seasons that are drier. So models can differ, some better match recent trends for other elements.

Graeme Anderson:

And I guess one of the key things that we've got to understand about climate change modelling is that there's plenty of room for more surprises. I really key bit is 90 per cent of extra heat at the moment is getting assaulted by oceans. So there's a lot of research trying to understand well, how does that affect the Pacific Ocean and the El Nino Southern Oscillation cycle? The Indian Ocean, a lot of extra heat there and where does that extra heat sit? And there's some recent analysis, put out yesterday. Just showing that when you look at some of those changes in the ocean, there's extra heat there. Where that extra heat sits is going to be really important because if more of that heat decides to stay in the western side of the Indian Ocean, then it can dial-up more, IOD positive or dry phases years for us in Australia.

Graeme Anderson:

So these things aren't fully understood, but they're important bits in terms of our future variability will play out. And things like the Tasman Sea out of southeast Australia, one of the fastest areas of warming ocean around. So if you repeat the last hundred years of weather, but you've got different areas of heat of where these oceans are, we don't get the same result. So we've got to beware of curveballs too within the climate change projections because they're pretty big models and often don't have a lot of the detail of some of the finer scale things. And so things like frost's or extreme events, there's still a lot more to understand about them. So if I was giving gold stars for these different forecasts, the highest gold star for weather forecasts...

Graeme Anderson:

The more accurate for seasonal forecast is a gold star, especially when we hit those seasons when we've got more confidence because there's a major climate phase underway, and you can actually see the forecast's unfolding. And the high confidence in the increasing heat and basically amongst all of these models, we've got higher confidence that in the fact that coming decades will be warmer than we do in what the weather will be at your farm in 11 days’ time. And here's just some of those curveballs, we want to talk about climate change too like frost, while daytime temperatures might be increasing, we've still experienced quite a few frost's and late frost and frost risk, and that's sort of part of to do with increasing subtropical region also to that drier pattern. Also rainfall intensity, basically, this is, Seth Westra had a webinar with us last year,

Graeme Anderson:

Just looking at rainfall intensity and what's going on. And some of these are happening at very small scales, but if you look at the individual thunderstorm, and this is rainfall, and basically precipitation close to this and how close it is to the centre of the thunderstorm storm cell. And this is a thunderstorm, and this is a thunderstorm at warmer temperatures, but what you could see it's the cone of where the moisture falls out. So if you warm-up a thunderstorm, warm-up the atmosphere, a thunderstorm not only can drag in a bit more moisture, but actually that moisture can fall out in a much tighter cone. So that's pretty important for farming because that could potentially mean you know more patchiness in the distribution of rainfall as you warm-up the atmosphere. So when things are right, we can get more moisture than ever before.

Graeme Anderson:

So there's some of the things that we've got to consider. They're probably aren't really well captured in some of the climate change projections. Other stuff like extreme weather, last year in southeast Australia, January was five degrees above average for the whole month. You won't find some of those things in that climate change models until into the second half of the century. So there's certainly room and there are some elements which are appearing a bit quicker than what's been modelled. But anyway, moving on from modelling, I guess the key bit is making decisions today that [inaudible] for the future. So models don't make decisions for us. They're just a bit of an input. And there's a lot of work going on. I love some of the work Cam Nicholson with Southern Farming Systems, he's doing a project on some great stuff there,

Graeme Anderson:

looking at improving decision making, and we'll be hearing from Cam in a webinar in a month or two's time. And I really like some of his decision making matrix that he has for saying, ‘Well, the forecast is important, but there's lots of other stuff that's really important that goes into a decision.’ and here is one just about a simple work through about, oh yeah, should I sell some cattle at the next sale ? He's saying, ‘Well, the things that farmers are taking into account is the likely price, the condition of stock, the feed demand, and what feeds is available.’ And these things are relatively knowable. And then you can throw in the forecast which might have some influence. But the forecast doesn't actually decide this for you, it's only an input into these other things.

Graeme Anderson:

So. So there's some really good stuff there. Cam is going to run through that in a future webinar. And also some work we're doing with GRDC using seasonal forecast project, Peter Heyman, from SARDI in South Australia and Barry Mudge have done some great stuff around decision making and how to use different forecasts and they've got a rapid climate decision tool, which is where you can put in your assumptions for a particular decision, and it might be around this one here is about whether to add extra N or no extra N. And then how often that decision is more profitable. And part of it then is saying, ‘Well, if you overlay that. If you've got a forecast, which is really swinging the odds towards drier? How much does it really change my decision?’ And what's been really interesting when we look at some of that and work through some of that with agronomists.

Graeme Anderson:

For quite a few decisions, you can actually find that actually, these things were a good decision to do whether you had a forecast for drier or wetter or not. Sometimes is, one decision is always the best, whether it's drier or not. So it's quite interesting that we don't need forecasts in probably half the decisions that we're making in agriculture, because we've got the information we need. And it's always going to be better off whether they're having to be drier or wetter. But there are times when a forecast is handy. And that's why it's important just to skill-up and understand what's in them. There's some other good stuff and guess we've seen quite a few programs around the place and I like this one which was really focusing on what farmers can do on their farm, what's under their control, the tools and tactics to deal with sort of increasing variability.

Graeme Anderson:

Understanding a bit more forecasts is one thing, but what are the things you've got under control is much more useful. And these are all things we typically see around Australia, with farms that are adopting stuff that's helping them deal with current day variability. It's well beyond forecast, but these are things that you do that set yourself up much better for the future. We've got modern R&D and better genetics and innovations happening, lots of great on farm and regional infrastructure to help ride out increasing variability. There's great demand for feed and fibre but we've also got to meet those demands, maintain our A+ biosecurity, improve business management, that's a really key focus for dealing with greater volatility of income throughout years, farm planning and basically looking after the land and natural resource management - there has been major progress there.

Graeme Anderson:

And I guess this whole process, we're all learning, we're all people working and learning together, making sure that people are tapped into good knowledge networks and channels. So we've got confidence to grow. So I'm going to finish up there have a bit of an overview of rough and tumble run through the wild world of weather, seasonal and climate change forecasts. So just wondering, is there any questions popped up?

Heather Field:

Fantastic, thank you, Graham. Some good information and tools and graphs there for us all to ponder over and look at in more detail. Yeah, just a reminder, if you do have a question, you can write that in the chat box, which is if you hover down the bottom of your screen, you'll see a little chat icon there and you can pop in your question and I can ask that of Graham. We do have one question so far from Jemma. Where can we access some of the graphs for the different locations? So some of the pie charts, that sort of thing.

Graeme Anderson:

Okay. So the Forecast for Profit website is... That's a really good one there. I might actually share my screen here Heather, if I can. And I think I can exit the... But anyway, that Forecast for Profit website, which I've got there on the slide. So it's been developed up, the local climate too and for across all of the GRDC South and South Australia, Tassie, Victoria, bit of southern New South. There's a long-term rainfall histories there, and you can click on your nearest site, and it'll run through that basically coloured diagram which sort of shows what's happened in different phases. And whenever there's been IOD positive, how often they've been drier and how often they've been wetter. And also you can check out El Nino. So it's basically understanding what's happened in the past and how those climate drivers have affected your particular region in the past.

Graeme Anderson:

And that's a pretty important bit, because once you can see where there's correlations or where there isn't, that's a pretty good guide, and then you can then use that to then better understand well, what are they talking about in the seasonal outlook. So if you know that in your particular patch, previous times, some of the bigger dry periods of when an El Ninos teamed up with a positive Indian Ocean Dipole and then you get to June and you're starting to hear all the models and forecasts talking about the Pacific Ocean heading towards El Nino and an Indian Ocean heading towards... Well, that's part of the thing that gives you increasing confidence in a forecast. Also what we see the changes from month to month can give you confidence in a forecast too.

Graeme Anderson:

So that if you see from one month, a lot of the models are sitting on neutral, but the next month, the half of them have jumped into say, drier or wetter. And then you can actually start to see changes in sea surface temperatures and changes in cloud patterns. Then you actually, I think looks like it's Game On. And that was certainly, if we look at, say, 2016, which was a wetter year in the Indian Ocean Dipole fired up, the models even in middle or later autumn were starting to swing towards wetter than average being more likely. And then when you click over to the next month, and there's more models jumped on it, and you can start to see the cloud patterns change then... Okay, I think this is a particularly a year when the forecast might have a bit more reliability. So that's a useful way to look at it.

Graeme Anderson:

It's not the same each year, I guess some years forecasts have more skills sometimes. So part of it its understanding enough about what sits behind them.

Heather Field:

Okay. Right. Thank you. We have a question from Chris. And he wants to know, what was the research that came out yesterday regarding increasing heat on climate drivers?

Graeme Anderson:

Yeah, well, there's some research on the, I think The Conversation, that basically researchers working on the Indian Ocean Dipole and looking at, they'd coral cores from off the Indian Ocean out they're off the coast of Sumatra. So that's sort of the eastern box of the Indian Ocean Dipole. And part of the whole swings of the Indian Ocean Dipole is where does the warmest water sit? When the warmest water sits over off the coast of Sumatra, there's more cloud in that part of the world that feeds into a northwest clad bands that it's a moisture source for southeast Australia. And then if more often that heat is sitting over on the western side of or over near Africa, then that's when more of a cloud and everything shifted over there.

Graeme Anderson:

So you've got these longer-term cycles where it's seesawing from one side to the other. But if there's more heat that's building up in the Indian Ocean, the question is, well, are the odds of those IOD's changing? And we've certainly in the last 20 years, seen more IOD positives than what we'd like. And I guess that study was just looking at some coral records going back quite a few hundred years to look to see if there's anything different in the recent history compared to what's there longer term because they can pick up that sea surface temperatures in the corals, so, so I haven't read the full paper yet, but we'll be doing it. But it's a good example and there's similar things in discussions happening in the Pacific Ocean, which is 90 per cent of that extra heat that's being absorbed by the oceans, where is that heat going to sit.

Graeme Anderson:

And if it decides to sit in certain places more often then that really does start to potentially change future variability that we might see in Australia. And some of those things aren't well captured and climate change models. And it's interesting, if all things were equal, and you warmed-up the oceans evenly, and you repeated the last hundred years of variability, it just means that every now and then if the warmest oceans are immediately to Australia's north, then we've potentially got more moisture coming to us. So you sort of can potentially strengthen the physics of that variability but there's a bit more going on than that. That was interesting in 2010 and 11, the sea surface temperatures to northern Australia warmest on record. So basically, as that heat sort of pops into those different oceans every now and then when it comes back to visit us, we do have the potential to be wetter than before, but not reliably, I guess.

Graeme Anderson:

And, and part of the challenge for climate change models is that yep, the tropics might be expanding from the equator, but how far and how often there could be greater variability between wet seasons and dry seasons. But certainly for southern Australia, one of the things that pick up in climate change models is just that increasing pressure pattern and things shifting south, so where the cold fronts hit the moisture feed, if that's happening, a few hundred k's for the south and that's what some of the climate models pick up as being drier. It'll still be variable in future but it just changes the potential of variability of what we might be confronted with.

Heather Field:

Thanks Graeme. We have a question from Julie. She's interested in how we can use climate information to make longer term decisions for business transformation. When to be thinking beyond crop choices, for example, in the coming season and towards more fundamental changes in production systems to different land uses.

Graeme Anderson:

Yeah, good, great question. Because I guess at the heart of adaptation, one of the key things is, climate change will turn up one year at a time or one season at a time. So we're still have to have businesses and farms set-up to deal with that variability. So understanding all the variability I've had in the past is really good. A key thing about humans is we tend to respond more to what we've experienced ourselves. So natural variability is got some big curveballs, potentially to throw at us that we haven't seen in our own lifetime. And when it does, it can really test us. So at its heart, one thing is setting up the business for dealing with the range of variability. And I know farms are not only trying to deal with seasonal variability, but also what's profitable, because we're dealing with changing markets, changing technologies, changing competitors, and some of the things that we're growing.

Graeme Anderson:

And in the end farms are trying to utilise their landscapes and their skills to make a living. And so you're adapting to, growing what the world's wanting and is what's going to pay you properly for as well as how do we set ourselves up for this variability in increasing variability? So when you're looking at increasing warming trends and a bigger summer, certainly, we see a lot of people that, if you want to find somewhere usually in southern Australia, that's one degree warmer for the whole of the year, you normally jump in a car and drive north 80 to 100 k's. And it's interesting, there's profitable farming systems all around Australia, the climate doesn't determine whether you're profitable or not, but there's very different systems that are set up in terms of what works for a particular climate.

Graeme Anderson:

And it's also interesting that sometimes when you go to some of the farmers in lower rainfall zones, some of the systems and techniques and approaches they used to deal with really big variability actually has them quite well prepared and set up. And sometimes it sounds counterintuitive, but on sometimes in high rainfall areas, and you're expecting a dam to fill-up every year and provide you with reliable water and a dam that might overflow five times a year because it's such a high rainfall area, if you have an experience where suddenly that dam doesn't fill-up with water and you haven't got alternative water, you can be quite vulnerable to those changes. So it's interesting that there's a lot to learn and share from other farm systems about how they're setting up to handle greater variability. And still make sure that we're profitable.

Graeme Anderson:

That's the key bit to be profitable to adapt because often those transitions, does require some good thinking and also some investments and changes along the way. So I hope that's helpful. Julie, that's a quick overview, but it's a fascinating story and it's a really great discussion topic for a lot of farm discussion groups just about well, what have we got covered, and then what can we learn from other farming systems that would help set us up better in your own district?

Heather Field:

Okay. We have a question from from Ralph. Great presentation Graeme, which climate driver has the biggest impact on East Coast Lows?

Graeme Anderson:

Good question Ralph. I was getting some emails with Acacia Peppler, who works with the Bureau who does done some great research on East Coast Lows. East Coast Lows, climate dog if you say, Eastie, runs around and there's really some influence of the big drivers on it but Eastie's can pop-up anywhere anytime, East Coast Lows, and probably all things being equal that subtropical ridge when it's quite strong and the pressure patterns quite strong, it's not helpful for the forming certainly of East Coast Lows down south in the southern parts and they can squeeze a few East Coast Lows out or make them a bit weaker. The Southern Annular Mode when we have the strong westerlies coming through like what we've had in the last six months with stronger westerlies from the southern annular mode that tends to prevent some of that easterly flow which you get along the southern and the East Coast.

Graeme Anderson:

And probably East Coast Lows prefer it if there is a bit of easterly flow then that's bringing in more moisture and a bit more helpful to them setting up. But there's still a bit of an unpredictable boost East Coast Lows and they appear on the weather chart and they have a big influence on the rainfall variability of places like East Gippsland and places east of the Divide. So there's some good explainers on East Coast Lows, the Bureau's got some good videos on it and also Eastie the Climate Dog gives a good run. But I think it's a bit more of a sort of a weather phenomena that occasionally can be helped or hindered by the climate drivers, but it's largely got a mind of its own.

Heather Field:

All right. We've got two or three more questions from Louisa, for the graphic showing Australia average temperatures since 1910 to 2019. Why does it start at 1910? And do we have reliable records for Australia pre 1910?

Graeme Anderson:

Yeah, a great question. All of those maps are just from a climate tracker from the Bureau of Meteorology, there are some great products there. And you can do that for rainfall and temperatures. So, for the particular tracker, it's got temperatures since 1910. I know the Bureau's got a terrific poster they've put together now for both rainfall and for temperatures and it goes to 1900. There is temperatures stuff going back, but I guess they've just drawn a line about when the use of Stevenson screens and the confidence, statistical confidence of temperatures. And when there's enough data points across Australia to be able to produce a actual map of Australia.

Graeme Anderson:

So to do with the data quality, I guess. And it's interesting, I know there are certainly hot days in the 1800s and in the explorer times, but also there's a lot of questions around a thermometer sitting against a shed wall or on a veranda wall isn't the same as a thermometer sitting in a Stevenson screen. So a shaded Stevenson screen. So a lot of the work when they're looking at what data sets they can use really is about what's a scientifically valid use of data and how far they got back when methods have changed so much, you can't use them. There's other parts around the world too where they've got longer term thermometer records and stuff in UK, so there's lots of discussion about when and where you start.

Graeme Anderson:

But the common theme amongst all of the world's heroes of meteorology is this warming trend, and we've had satellites up since the 70s. And they're all consistent with showing this increasing warming trends and it's all sort of what's been expected, but we're at the early stages of this in terms of certainly the ability of humans to try and rein that in. Hope that helps.

Heather Field:

Okay. We are at time but if you're happy to stay on the line, to answer a couple more questions.

Graeme Anderson:

Always happy.

Heather Field:

We've got one from Kim. In your experience Graeme, do farmers generally believe that the climate is changing?

Graeme Anderson:

No, not really. I think probably reflective of wider, I think people have got all sort of views on and what the things go on with weather. Certainly a lot of farmers I meet would say that there were seasons are less reliable than they once were, there's something going a bit wrong. And a lot of farmers say, listen, ‘I haven't focused too much on climate change, I mean, really focusing on trying to set the farm up better for variability.’ And so there's a big range of where people are at on the issue, I guess, probably like a lot of different topics around the place, but not everyone needs to accept climate change perhaps to be doing things that actually contribute to being adapted to it.

Graeme Anderson:

And also tackling lower emissions products and certainly, I've had plenty of situations where I've might have talked to people that perhaps aren't convinced of climate change, but they're incredibly enthusiastic about improving water efficiency on-farm or improving their water and fodder storage or understanding more about seasonal forecasts or putting in soil moisture probes to better understand how much moisture is in their soils. And they're also incredibly well, while I might be too sure about emissions reductions, they're really quite excited about renewable energy options that can reduce their energy bill or, or things like energy efficiency measures, they can take that cuts their energy losses, saves them money, all of that sort of thing.

Graeme Anderson:

So I guess we're used to dealing with people who had a wide range of attitudes so other things going for climate. I guess our role in Ag Vic, we're certainly a science based organisation, we have a lot to do with the Bureau of Meteorology and CSIRO and reading peer review papers and our extension work is largely around sharing that with farmers, because while we might have experienced lots of weather and seasons in our past, which is what farmers know a lot about. Some of these changes that are happening high above our heads and in the oceans that's around Australia, they're not necessarily things we would see so we've got to get that information and understand what that means, and that's by the scientific community. So I think there's probably a need to improve their climate literacy in agriculture, whether it's to do with seasonal forecasts, or weather forecasts or climate change.

Graeme Anderson:

But also it's not everyone's cup of tea, and I certainly meet plenty of farmers who are saying I'm just going to focus on setting up our farm better to deal with it. And I think that's quite a legitimate spot to be, too. So there's a lot more fun working on solutions, so I can tell you that than arguing about whether climate change is real.

Heather Field:

Absolutely. We have a question from Bronwyn. I have seen maps showing the ocean around South-Eastern Australia is warming at a faster rate than other ocean areas. Is there a driver for this?

Graeme Anderson:

Yeah. There's quite a bit of work going on with that. My understanding is it's the East Australian current which comes down off the Pacific and hits the East Coast, it's just happening a bit more stronger. And you can see that warmer water down off the East Gippsland and eastern Bass Strait and off the East Coast of Tasmania and I know a lot of the fishermen see it because the fish don't really care. They just follow the thermal so the catching swordfish in different things a lot further south and what they have and it's bringing other fish species and that with it. So the stories there, and certainly some of the marine ecologists that are looking at some of those really cold tolerant, kelp forests and things like that some of them are really being impacted under the water there.

Graeme Anderson:

So the main driver is by though, like I said, 90 per cent of the heat is being absorbed by the oceans. And that heat is sort of starting to be quite obvious in some pockets. But it may not be distributed evenly, depending on what ocean currents do. So that's an example of where a bit more of the heat is being pushed down and having quite an impact in a small area. It also means that if you run forward weather patterns over that, it just puts a little question mark on how might that change future variability for down in that southeast coast, because it's substantially different to what's happened in the previous hundred years.

Heather Field:

Thanks, Graeme. And we've got a last question. I believe it is the last one, and I'm not sure if I'll write this correctly, but I'll try and see if you understand. Why doesn't the BoM verification tool called past accuracy, which is the weighted percent correct calculation include consideration of the current state of the climate drivers?

Graeme Anderson:

I'm probably not equipped to answer that one. But the Bureau and it's really good they've got the skill that with those seasonal outlooks, they have a skill map on the back there. And it's always worth looking at. And that's basically saying, how often is it right? And remember, you're talking about probabilistic forecasts too, so they're not actually predicting any particular outcome. They're just saying a range of outcomes. So for some seasons since of some areas, there can be periods where there's higher skill in the current model they're doing. So I guess they're doing that from using that forecast model. And when they run it backwards on the past seeing how many times has that particular forecast got it right?

Graeme Anderson:

I guess one of the things that changes though and while we talk a lot about seasons when there's a really strong climate driver, the climate models don't necessarily put it in those terms. It's just measuring mass from all of the different squares of the ocean and atmosphere around us and coming up with those calculations. So I guess it's using the metrics of what the models use. From a human point of view, when we're looking at when is there perhaps a bit more confidence, which is a different thing to skill, are given with seasonal forecasters, there are some seasons when you could meet them around the water cooler, and there'll be a bit more confident about how the coming months are looking a bit drier, or the coming months might be looking wetter.

Graeme Anderson:

And there's a bunch of things that feed into that, and that's part of where I think climate literacy comes in. So it's not enough about just about getting the output of a forecast. But and think, again, this is the same again whether you're better to try and buy urea or whether you're about to sell livestock or sell wheat... Any prediction of the future, it's not just knowing what the the prediction is of direction, but you're also needing to understand the commentary about, okay, if you're thinking that it's heating up, what are the fundamental indicators that are suggesting that? And if someone can explain to you that well, because this and these things are all feeding in and they traditionally have affected price, and they've additionally sent price in this direction.

Graeme Anderson:

And when you can see those things actually happening, then that actually gives you a bit more confidence that why prices might go up. So farmers do that all the time when they're buying and selling and trading and I guess forecasts of seasons is no different and like I said, no one has an emphatic...it's view of, well, there are people offering that, but there is no certainty about what does happen in the future. But there are some situations when you can see a few key things that give you a bit more confidence in the outlook. So beware anyone that's an emphatic forecaster, because I know there's a great desire for that, but it's quite misleading. And I must say, there's quite a strong history of it in Australia. And good luck and it's great when they occasionally get it right. But again, that's the same for our star signs.

Heather Field:

Thanks, Graeme. So that looks like the final question. And we have had a few participants, thanking you for your presentation and also the excellent questions that have been coming from the audience as well. So we will wrap it up there. And I will remind you though, when you do close out of today's webinar, there is a short survey. So that'd be great if you can complete that. And we do have another webinar coming up in a couple of weeks’ time on the 25th of March. And at that webinar, we'll be hearing from Robert O'Connor from Agriculture Victoria and he'll be talking about tools for better irrigation scheduling. If you're a subscriber to our webinar series, you will receive a notification for that shortly. And you'll be able to register for that webinar. Unless you've got anything else you want to add Graeme-

Graeme Anderson:

No. Thank you very much, Heather you doing a great job with these webinars. And there's plenty of good speakers down the line. So thanks for joining everybody.

Heather Field:

Excellent, thanks, Graeme and have a good afternoon, everyone.

Page last updated: 04 Oct 2021