 So I'm Andy Pittman, I'm the director of the ARC Centre of Excellence for Climate Systems Science based at the University of New South Wales and my personal research is focused on terrestrial processes and their representation in climate models. So the basic underpinning laws that climate models are built from includes the basic Newton's laws of motion, conservation energy, conservation of mass, basic physical principles that physicists discovered a very long time ago. They're the basic underpinnings of the core of a climate model. Then on top of that you build parameterizations or representations of processes which need to be resolved at scales that we can't explicitly model in the climate models. Things like clouds, things like convection in the atmosphere, things like eddies in the ocean, things like land surface processes and a lot of the challenging in modeling the detail and the regional patterns of climate come from how we represent those parameterizations. We don't so much tune parameters in climate models. What we do is take observations, learn from those observations how processes work and then we build mathematical representations of those basic processes into the climate models in the form of computer code. All a climate model is is a million lines of computer code running on a really really big computer system and that computer code is a suite of algorithms built from observational data where that exists or basic physical principles where we lack the observations to understand the detail process. But when you run a climate model and you get data from it you always reintroduce observations at that stage to evaluate the model to check the model. We always analyze the model results to find weaknesses and there are always things that work better in climate models or worse in climate models and we use observations routinely to tease out what does and doesn't work well in the models and of course focus on improving those things that don't work well. The things that climate models struggle to capture well would include some extreme events. They struggle with the location of the storm tracks. They struggle with the detail of cloud fields. They struggle with some major challenges we don't really represent at all the processes which might trigger a abrupt climate change. So methane release or permafrost melt although some people are beginning to work on permafrost. So I think it all depends upon your perspective if you ask what do the climate models struggle to represent in terms of the simulation of whether it will warm for a doubling of CO2 my answer would be nothing at all because they do that really well. If you ask what the climate model struggle to predict at the scale of a region and its response to doubling of CO2 in terms of rainfall lots of things. They don't get the details of the clouds the convection the rainfall processes the detailed synoptics blocking a whole range of things because the spatial resolution that we use for climate models is probably too coarse to capture a lot of those kinds of key phenomena and so one of our grand challenges in climate models is to dramatically improve the spatial detail that climate models use and that's really a computational problem. We just need bigger supercomputers to to really resolve the detail of those things. Look if you go to observational data sets of the global patterns of anything pressure rainfall temperature you won't be able to tell which is from a model and which is from the observations. The climate models large scale simulations of the climate are outstanding a truly phenomenal achievement over the last 20 or 30 or 40 years. So at the large scales they are outstanding there's probably no parts of a climate model of a modern climate model that still reflects what was done in the 70s almost everything's been rebuilt or rewritten I think. The resolutions increased from something around 700 by 500 kilometres to 100 by 100 kilometres. The detail in the vertical has increased dramatically, oceans have been properly and fully coupled, the land surface has been completely revised to incorporate a whole suite of processes, sea ice models have improved dramatically, cloud parameterizations have improved, we've resolved most of the water vapour feedback problem. It's like asking what's the relationship between a Formula 1 Grand Prix car in 2014 compared to 1970 and the answer is there probably isn't a single widget that's shared. Because climate models are built from the basic physics they should be applicable to any environment so you should be able to take a climate model run for the earth and use it for a different planet and people do do that. Similarly because the basic physical processes and Newton's laws of motion don't change and conservation of energy and mass don't change with time in any context that we understand it anyway. The climate model that's built for today ought to be able to do the past, paleo climates for instance and the future and indeed they can. They're not by any stretch of the imagination perfect if we try to simulate the last glacial maximum or 30 million years ago but they do capture an awful lot about long term past climates and indeed there've been discoveries around things that happened in the long-distance past which have been hypothesized using climate models and then discovered from observational data. So if the climate models do a pretty decent job of simulating the long-term past climate change and can simulate the 20th century extraordinary well we have considerable confidence that they can simulate the near future extremely well and by near future I mean out to 2040, 2050. As you move forward into the future if mechanisms begin to develop that are outside of the observational record the observations we've used to build the model in uncertainty increases. So for instance we don't represent the processes that might trigger a abrupt climate change we suspect there are things that might evolve in the climate system that aren't in the observational record and whether or not we will capture those nasty surprises time will tell. So paleo climatic modeling is extremely useful because it tests the models for very large changes in the Earth's climate in the past. The problem is they're not necessarily good analogies with the next 50 to 100 years. Paleo climate mostly was overtly driven and driven by movements of constants. Under time scales of 50 to 100 years we're not expecting massive changes in solar luminosity if there were we have bigger problems than global warming and we're not expecting Africa to suddenly move a thousand kilometers west. So the triggers and the drivers of paleo climatic change are profoundly different to the drivers of human induced climate change and so it's useful to test models but it's insufficient it's useful but insufficient necessary but insufficient. There's a whole variety of reasons why we're confident in in the skill of climate models for the problems that they were designed for. First of all they're built upon physical principles and those physical principles are known unless of course that Newton was stupid which I don't think he was. So we have basic fundamental theory not sourced from climate science but sourced from basic chemistry basic physics basic biology that and applied mathematics that says the core of climate models is sand. Secondly of course they use routinely in other applications like weather forecasting and so we have effectively can evaluate a lot of our science routinely and weather forecasting is becoming increasingly accurate irrespective of what some of your listeners might think. If they actually write a diary of a five-day forecast and then check off how those five-day forecasts evolve they'll find that they are shockingly accurate nowadays. So that's the second test. Thirdly we routinely test our models against observations over the last century and earlier and they do extremely well in that in that respect and finally we can test our models against perturbations so we can for instance simulate a volcanic eruption for example and check that the climate models respond appropriately to to what a volcano does to the atmosphere. So there are multiple lines of evidence and they all point to the climate models being reliable for what they were designed. So on longer time scales so not five years but for periods where the initial conditions aren't critical so on time scales of decades and longer they're very very good and for problems that are at larger spatial scales so continental scales and above they are I think extremely reliable up to 30 to 50 years into the future. As you move further into the future I become less confident so I think they are a good and indicative guide for what the climate might be like in 2100 but I wouldn't be absolutely sure and I wouldn't be sure not because I think they're overestimating the response but I think they lack a suite of mechanisms that might amplify global warming and so I suspect if anything they are underestimating the scale of change out to the end of the century but this science isn't clear on that so I'm being a bit speculative. Secondly spatial resolutions at say the scale of Queensland or New South Wales the detail isn't in the models because of the computational cost so as you move down to finer and finer scales as more and more mechanisms are interwoven to give you the regional pattern of change I think the climate models become things that we're less confident about. Finally the things that are not based upon well understood physical principles so changes in agricultural yield require a whole variety of things to be done well ranging from soil moisture and rainfall and temperature and lots and lots of all lots of things that interact. When you need to get dozens of things right finer spatial scales out of a climate model to give you an impact then you should treat what the climate model tells you about that impact very carefully. There's a lot of myths out there on what we think or how we build our understanding of what will happen in the future. There are many lines of evidence that are used to understand how climate might change in the future and if you could take climate models away we would still be just as worried about the future climate. Climate models merely inform and embellish and add color and flavor to the future of the climate the projections of future climate but we would be just as worried based on theory and data. Climate models are one strand of evidence for future climate change but by no means do they underpin our concerns what underpins our concerns is physics. So there's a lot of research out there that gives us insight on on climate sensitivity. It ranges from some fairly sophisticated analysis of paleoclimates through to some rather simplistic approaches to understanding how much the climate may change. Climate sensitivities how much the planet will warm in equilibrium to a doubling of CO2. The range in the last IPCC report was 1.5 to 4.5 degrees. I think the 1.5 will turn out to have been overly cautious or optimistic from the science community. I think the right number is in the high 2s or low 3s and I base that upon multiple lines of evidence. The evidence to suggest it's below 1.5 is is very arguable I think. People need to understand that there's a multiple line of approach to the peer reviewed literature. So somebody can write a paper and get it through the peer review process that's the first check but the much tougher check is time and the test of time. So people will publish papers that suggest climate sensitivity six degrees or 1.3 degrees and both of those are big claims needing strong evidence and critically independent reproduction. And what we tend to find is the big claims that climate sensitivity is very low or very high turn out over time when checked and reproduced and rerun and reanalyzed don't stand independent scrutiny and that's the proper process in science. There's nothing wrong with that if you come up with the really strong piece of evidence that the climate sensitivity is six degrees and you can get that published good for you. But it's then a requirement that the community do their best to destroy that. And if they can't destroy it then that's really scary in the case of six degrees and really good news in the case of one degree. But so far as I know every single time somebody has come up with values very low or very high subsequent analysis has demonstrated that they're wrong. But I do like the scientific method because a lot of people don't understand how it works they assume that if you've published a paper the papers right. Obviously everything I've published is right including the things that I have contradicted in subsequent papers. Science evolves and the bigger the claim the bigger the attack on the claimant and it's right that that's true. It's good that that happens. One of the really emerging challenges in climate science is the question of extremes. Weather forecasts are built to capture extreme weather or to predict extreme weather and it's really really hard. It's particularly hard when an extreme is a combination of multiple events occurring simultaneously. And climate models cannot model or simulate the detail of extreme events. So for instance a tropical cyclone. I don't think climate models tell us very much about what will happen to tropical cyclones because they're just not built at the spatial resolution that enables us to capture the detail. And in and also things like blocking where a high pressure system forms and holds the climate in place for a significant period of time. Commonly leads to extreme conditions particularly heat waves and drought. And climate models just don't have the finesse the spatial detail to capture the mechanisms which drive those sorts of things. So there has been a huge growth in analysis of climate models and extremes but most of it is the statistics of extremes. So what would be the probability of particular events occurring in the future? Not what would a specific event look like in 2030 or 2050. So when you're looking at the statistics of extremes I think climate models have some skill in showing you the sign of the changes but I'm not overly confident they yet have the skill to give you the detail around how extremes will change. The problem is that whichever extreme you look at seems to get worse in the future in part because of the intensification the hydrological cycle and the just the background warming that we see from CO2. So I think we have a good understanding that a lot of extremes will get worse but of course some extremes will get much rarer like extreme cold at least in Australia. It's not necessarily true that extreme cold will get less common in higher latitudes where you see an increase in the strength of the storm tranks and more variability for instance bringing Arctic air down over parts of North America and Europe. Global warming doesn't mean everywhere has to warm. The relationship between global warming and the hydrological cycle is something we're beginning to understand much better. Basically you can start at a whole variety of points in the hydrological cycle because it's a cycle. If you start with evaporation if there is more energy there is more energy to drive evaporation. So if you increase CO2 and you increase downwelling infrared radiation there's more energy in what we call net radiation and it's the net radiation that's used to drive processes like the turbulent energy exchange with the land the latent heat flux or the evaporation and the sensible heat flux which directly warms the lower atmosphere. So if you have more energy you drive the evaporative processes faster and you pump more energy at the base of the atmosphere in the sensible heating that drives deeper convection more moisture being driven up higher into the atmosphere that tends to increase the probability of rainfall and more intense rainfall and you can see how a cycle begins to form but basically it's an acceleration or an intensification of the evaporative processes. So we suspect that areas already susceptible to drought will end up in drought for longer periods because you tend to see a situation where the evaporative process driven by the energy sucks the water or drives the water out of the landscape a little bit quicker, not massively quicker. We're not thinking this process would drive a one month drought to be a 20 year drought but if you have more energy and there's water available that water evaporates a little bit faster. If you evaporate the water a little bit faster you dry the landscape out a bit quicker and that tends to intensify drought but the major driver of droughts across the landscape is the larger-scale synoptic patterns and the modes of variability like ENSO and El Nino and El Nino cycles. So the key regional drivers of drought are the large-scale modes of variability and what I was talking about in terms of latent heat and sensible heat drying out the landscape is if you like the detailed brushstrokes on a larger canvas. So you have to look at the system in its full coupled glory to understand the difference between a drought that's occurring becoming slightly more intense because of evaporation as distinct from the large-scale patterns driving more frequent droughts and those are two quite different mechanisms. So if you're in an area that's already susceptible to drought it may well be in some years enough to drive so much deficits which are problematic for agriculture or for water resources in general beyond a threshold such that they're problematic. Also if you have a general drying when it does rain that water doesn't necessarily make a substantial difference for agriculture or for water resources. So a much drier landscape where everything's in deficit if there is rainfall that landscape will suck up that rainfall and hold it which is a problem for for dams and water supply as we have seen in Perth for instance. So a small change in the dryness and a small change in the rainfall commonly translates into a big change in the runoff into dams for example and there's good understanding of the mechanisms behind that. When I was an undergraduate I was offered a choice of either going to Canada to work on land surface processes or France to work on cloud processes my French is appalling so I chose Canada. Okay so my personal area of research is terrestrial processes and there are a huge number of opportunities for bright students to contribute in in this area. There's a whole variety of interesting questions it depends where you sit. One is how do we parameterize or represent urban landscapes in climate models. Urban environments appear to amplify some extremes. Those aren't represented in the climate models because they're too fine a spatial scale. We're now moving down to their scales that we need to resolve those landscapes properly and how to represent urban landscapes in a climate model so we get the right moderation or amplification of a global warming signal is very important. Another one is how do we represent agricultural systems in the climate models they act quite differently to natural landscapes for instance they're harvested and they have a quite a different phenology in terms of the way that they interact with the atmosphere and how to do that is a real problem. Next is something around soil carbon soils store a vast amount of carbon and there's good evidence to suggest they'll store less carbon in the future because microbial activity will break down that carbon more far more quickly and recycle it back to the atmosphere. That's a six million dollar or six billion dollar question because if the soils lose their carbon as a consequence of global warming it's a massive positive feedback. If the soils hold on to their carbon as well as they currently do or even increase their ability to store carbon that's a negative feedback on global warming and while we're pretty sure now we have a positive feedback we're not sure of the scale of that feedback it's extremely messy and complex to represent those processes and it's something we have to do much more skillfully and the last is the whole issue of human modification in the system things like irrigation damming of rivers how those flow into the oceans because that impacts on the ocean circulation that whole issue of how humans interact with the energy carbon and water cycles is a massive challenge that we're only beginning to chip away at. Most of us who work in the field think that using soils to store carbon if we did it really well and really actively and on very large scales might enable a country to attain its current levels of soil carbon. So by default we'll lose soil carbon we're losing it by erosion because massive amounts of soil are eroded annually across the Australian landscape because filing practices and land clearance and a whole range of other things. So the default is we lose soil carbon over the next 30 to 50 years at a significant level. If we really actively work on that system we may be able to hold the current soil carbon in place. The notion of using soils as a carbon sink to offset global warming is optimistic. So traditionally I would have a conversation with them. I've started depends who they are but I've started saying look everyone's entitled their own opinions but not their own facts and if you happen to want a conversation around why the science is now clear and robust I'm happy to have that conversation with you but if you've already decided that we're a bunch of global conspirators who've managed to orchestrate a complete conspiracy that has convinced most of the world's governments of a problem that doesn't exist I probably can't do much for you. Climate scientists are trained in maths, physics, computer science, biology, chemistry and as most people know people trained in those fields have ordinary communication skills typically. Our definition of success is our ability to write code or do a chemistry experiment or whatever it is. And I've recently spent some time assessing graduate students what's called the Monash Foundation and a number of the students who come in to be interviewed for scholarships are from law and they have spent lots of time mooting and debating and their ability to communicate, to succinctly summarize their interests, their desires, their vision is phenomenal because they're practiced. So one of the things we're trying to do is get our graduate students into a much more personalized program where they do talk more frequently. We're trying to build opportunities for them to be trained a bit more in science communication because I think it's critical and the answer to the question is practice and understand that the first time and the 15th time, the 50th time you'll probably do it not very well but in the end you'll develop some skills. I also think we have to read more broadly. I do not believe that we can all go and do a second degree in science communication or in psychology. Climate denial is increasingly understood by the psychology community and there are some fairly accessible books ranging from things like Naomi Oreski's book on motions of doubt which gives you a really good insight on some of the problems but there's also books in psychology and there's even great posters which show you the psychological trick that someone trying to deny climate change is using and why it's falsifiable. It's worth knowing about those things so you can see the techniques being used against you but it does require a little bit of a broadening of the training that we all have and we have to be very careful with that because what we really need is really committed bright students doing climate science not trying to be everything to everyone so it's a balance but but at least engage in some of the the basic background to why people have these scientifically invalid views is a useful thing to do. I think I'd simply say that if we're emitting 10 billion tons of carbon dioxide in the atmosphere per year and you don't think that can have an impact you need to look around you and look to see what happens in a city like Beijing when a fraction of that level of pollution is put into the atmosphere people can't breathe in summer so we can look around us and if you open your eyes you will see changes going on around you that are clearly demonstrable evidence of climate change and the consequences and if you choose not to look there isn't probably very much I can do for you.