August 21, 2020 | Published by |

While we find ourselves in challenging times as a result of the COVID-19 pandemic, we have still found moments to celebrate and have been delighted by the enormous range of extraordinary research that has being published over the past four months.

Without doubt the most impressive paper published during the past few months was the international analysis of climate sensitivity led by Steven Sherwood and published in Reviews of Geophysics. This is a ground-breaking piece of work that has narrowed the range of equilibrium climate sensitivity for a doubling of carbon dioxide compared to pre-industrial times from 1.5°C–4.5°C – where it has stood for 40 years – down to a robust estimate of 2.6°C-3.9°C. This is fundamental work that will undoubtedly be a major part of the next IPCC AR6 WG1 report.

A major component of this work was the examination of paleoclimate and instrumental records, which allowed the researchers to look back at past responses to interpret future changes. A similar approach of looking backwards to prepare for the future was the basis of a study with Associate Investigators Pandora Hope (BoM) and Jo Brown. Here our researchers used documentary sources and paleo records to see how modern infrastructure would have coped with past pre-instrumental floods. It found that pre-instrumental flooding in some areas of Australia, particularly in the 18th Century, were far higher than those in the 20th and 21st Century, suggesting we may not be able to cope with future floods, especially once the influence of climate change was factored in.

While past conditions can give us indications of a future under a changing climate our best predictors of the future and even some aspects of the present can still be found in climate models. A team of international researchers investigated whether increases in heavy precipitation from 1951-2015 could be ascribed to human caused climate change. Using CMIP6 models they found it was difficult in the Southern Hemisphere to disentangle natural variation from greenhouse gas influences because of the high variability. By contrast the observed increases in extreme precipitation over the global land, Northern Hemisphere extratropics, western and eastern Eurasia, and global ‘dry’ and ‘wet’ regions, were largely explained by the influence of greenhouse gases. 

Model studies like this play an important role in future climate projections but, as we all know, they are not perfect and a big part of our research program’s work is improving them, particularly their representation of future rainfall, which remains a vexing issue.

Increasing the resolution of climate models has been put forward as a way of better capturing the patchy nature of rainfall. A paper by Margot Bador and colleagues found that simulated precipitation extremes were more intense with increasing model resolution but to get a systematic improvement of rainfall representation required physics tuning. Adding to the difficulty of modelling precipitation, real-world observations were so variable that it made it difficult to quantitively evaluate model performance.

Figure 2 (above): Videos of typical behaviour in the UM (left) and WRF (right). Shown is CAPE in black, surface potential temperature in yellow/blue, precipitation in purple, reflectivity in white.

Research from an international team led by Martin Jucker also revealed the choice of convection resolving models when looking at locally forced convection had an impact on results. Focusing on the Met Office’s UM model and WRF, the team found each model produced what it liked best, independent of what happened in the real world. The UM produced very strong, small thunderstorms early in the day while WRF produced an impressive squall line in the evening. The researchers speculated this could be an historical artefact of model development and proposed observational campaigns that could address ways to correct these flaws.

Another study by CLEX researchers and colleagues found in tropical regions cold pools beneath thunderstorms can organise thunderstorm clusters. When cold pools are weak, thunderstorms align parallel to the prevailing wind and they can also have a direct impact on anvil clouds, which affects their ability to reflect solar radiation. These effects will need to be considered in climate models.

Beyond localised rainfall, modelling high impact tropical cyclones and how they will change in the future is of growing importance. Pavan Harika Raavi and Kevin Walsh examined two parameters used for tracking tropical cyclones – the CSIRO tracking scheme and the Okuba-Weiss zeta parameter – to determine relative performance in simulating tropical cyclone frequency characteristics. The found the Okuba-Weiss zeta parameter corresponded more closely to observations and was able to differentiate between monsoon lows and tropical cyclones.

Ocean temperatures are one of the key characteristics that help in the development of tropical cyclones. And that, according to Margot Bador and colleagues, creates a large area of uncertainty in modelling how tropical cyclones will change in the future. They found correcting for the reliability of future sea surface temperatures has a profound effect on cyclogenesis, reducing it by up to 55%. The researchers found that this uncertainty in the future patterns of sea surface temperatures could strongly hamper the reliability of projections of South Pacific tropical cyclones. In addition, the researchers found that this strong reduction in tropical cyclone activity was caused by stronger vertical wind shear in response to a South Pacific Convergence Zone equatorward shift.

But it’s not all bad news for climate models. CLEX researchers and colleagues compared the performance over Australia of the new CMIP6 models to the previous CMIP5 group. The results showed improvements in the latest generation, including better reproduction of land and marine heatwaves and sea-level rise as well as improved relationships between Australia’s climate drivers and rainfall. However, some models continue to show increased sensitivity greenhouse gas emissions and increased 21st Century warming.

But if our models are not perfect, fascinating research by Ewan Short on direct edits by forecasters to numerical weather prediction models suggests humans have their shortcomings too. These edits are often made to better resolve land-sea breeze, and boundary layer mixing. CLEX researchers compared edited and unedited forecast data with weather station observations. The results were nuanced but broadly showed that when winds were considered at individual stations or averaged over small spatial scales like that of an individual city, the human edited forecast dataset generally exhibited larger errors than unedited model data. However, the human edited forecast can occasionally produce lower errors than the blended, ensemble average forecast, because ensemble averaging overly smooths the daily varying component of the wind field.

Real-world observations not only allow us to adjust the behaviour of climate models they also continue to give us insights into our climate and weather that can help us prepare for extreme events.

One of the best places to observe cyclones is the Mediterranean, which not only has frequent storm events but has a dense network of observational instruments. This network enabled CLEX researchers to examine cyclone like sub-structures that can form within massive storms by focusing on an event that hit the Mediterranean in October 2012. The results showed how the upper atmosphere contributed to the development of storm and was then amplified again by the terrain below. Together this interaction produced a stable, long-lasting sub-cyclone that by itself caused significant damage. Understanding how each of these characteristics contributed to the development of this storm within a storm can help us to recognise the features that create and sustain these sub-cyclones. With this knowledge, we hope to improve our ability to forecast them and give residents in their path more time to prepare for their impact.

Another massive storm this year, Storm Gloria, that struck the Iberian Peninsula in 2020, also gave CLEX researchers insight into which storm generated waves caused the most damage to infrastructure. Winds contributed the most to the storm surge – around 70% along the entire coastline; atmospheric pressure was generally negligible, while wave setup itself accounted for up to 40-50% of the storm surge in some areas. This is useful information for forecasters that will improve warnings of damaging surf similar to those that have recently hit the NSW coastline and washed houses into the ocean.

While the Mediterranean is well observed, instrumentation on other regions of the world are sparse. In these regions other processes are needed to produce the observations that improves the quality of our science. Satellite data is often used to estimate rainfall, but when these estimates are compared with data in areas where rain gauges exist, there is still a significant margin of error.

To reduce this error, a CLEX researcher and international colleagues developed a hybrid approach to estimate recent rainfall that combines satellite-based rainfall estimates with satellite-based soil moisture estimates. When this approach was tested against independent rain gauge measurements it showed notable improvements, in a range of metrics when compared to other existing satellite-derived estimates.

Another way of getting useful observations is through radar networks. CLEX researchers developed the Radar Organisation Metric (ROME) that can assess the degree of convective organisation in the tropics. ROME’s statistical properties suggest it is able to distinguish between the degree of convective organisation, and it also captures different regimes of the monsoon in Northern Australia. This adds new capabilities that other metrics lack.

CLEX researchers have played a pivotal role in a Special Issue currently underway in Environmental Research Letters Focus on Extreme Precipitation Observations and Process Understanding’. This collection of papers was put together as part of a joint initiative by the World Climate Research Programme (WCRP) Grand Challenge on Extremes, GEWEX and the International Precipitation Working Group (IPWG). It follows on from the development of a global database of precipitation – Frequent Rainfall On GridS (FROGS) – in which CLEX researchers also played a key role. This database contains global gridded daily-precipitation products from in situ, satellite and reanalysis products in a common format, enabling researchers to intercompare observed rainfall extremes easily and begin to understand some of the large uncertainties that exist in our global products.

Postdoc Nina Ridder recently gained a key position in the climate science community when she was elected on to the executive committee of the Young Earth System Scientists (YESS). The committee helps maintain an overview of YESS and guides the community’s activities.

Meanwhile, Associate Investigator Yi Huang has received some funding from the Joyce Lambert Antarctic Research Fund to study Clouds, Precipitation, and Boundary Layer Characteristics in Subantarctic Mesoscale Cyclones.

And as conferences were cancelled around the world, we found way to continue sharing knowledge online. Margot Bador presented a Zoom Session on evaluating and improving precipitation in climate models as part of the EGU Online activities. We also provided three lectures by Hugh Morrison which can be viewed via this link, to access the zoom recordings. If you are outside UNSW, contact CLEX for password details.

We were also very fortunate to welcome a collaborator who arrived in Australia before the COVID restrictions took hold. Dr Peter Knippertz from Karlsruhe Institute of Technology (KIT) has been working with Michael Reeder and Christian Jakob on the dynamics of synoptic-scale propagating weather systems in the tropics, more specifically their interactions with moist convection and their predictability, until September

Amid all this research and scientific engagement, close to home we have had plenty of cause to celebrate.

CI Julie Arblaster was made a full professor at Monash University. This makes her the only female Professor with a continuing academic position at the Monash School of Earth, Atmosphere and Environment. Julie has also recently joined the Coupled Modelling Intercomparison Project (CMIP) panel. This panel oversees the design and coordination of climate model experiments that feed directly into the Intergovernmental Panel on Climate Change’s assessment process and is at a critical stage of planning for the CMIP7 round of coordinated experiments.

Todd Lane was acknowledged in the Dean’s Circular at The University of Melbourne for his exceptional leadership of the Weather & Climate Group. To quote the Dean: Todd Lane (Earth Sciences), was nominated twice for keeping the Weather & Climate Group connected and stable, displaying a warm, relaxed and empathetic leadership style that gives everyone a boost, and finding time and energy to support his group from the very start.

Over the past four months, we have watched some of our students move on to the next phase of their career. Pavaan Harika Raavi delivered her PhD completion seminar focused on the relationship between climate and the mechanisms of tropical cyclone formation and Masters student Genevieve Tolhurst submitted her thesis Synoptic and mesoscale drivers of Victorian precipitation extremes. And while we’re acknowledging our impressive students, it was great to see PhD student Kim Reid get some good coverage in the Sydney Morning Herald and other Fairfax/Channel 9 outlets in the article ‘Atmospheric river’ 2000km long wallops Western Australia, heads east. This was a great example of opportunistic media with Kim and her supervisors recognising that a newsworthy event was occurring that involved CLEX research.