# Research Community Seminars

### NOAA's Hazardous Weather Testbed: The Spring Forecasting Experiments - What is that all about?

Harald Richter
Bureau of Meteorology
Wednesday, 21st Aug 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: As part of NOAA's Hazardous Weather Testbed, a Spring Forecasting Experiment (SFE) has been run annually for ~20 years during the peak of the severe convective season in the United States (~May) and is currently based at the National Weather Center in Norman, Oklahoma. Through application on real-time severe weather as it unfolds, the SFE aims to test emerging concepts and technologies for the prediction of hazardous convective weather, accelerate the transfer of promising new systems into operations, inspire new initiatives on operationally relevant research, and document the sensitivities of state-of-the-art experimental convection-allowing modelling systems (CAMs). For a five-week period every spring, ~100 forecasters, researchers and system developers from around the world step through structured tasks designed to support the above aims.

Above all, the SFE highlights the usefulness of CAM ensembles for the prediction of hazardous warm season mid-latitude convection. I will present some of the experimental systems involved in the more recent SFEs (which includes the UKMO's Unified Model) and elaborate on the tasks designed to utilise these systems in operational forecasting. I will present some of the SFE findings that are easily adaptable to Bureau systems and services, including work packages already proposed as part of the Public Services Transformation.

### Improving Global Precipitation Forecasts using Satellite-derived Precipitation Data

Shunji Kotsuki
RIKEN Center for
Computational Science, Japan
Friday, 16th Aug 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: This study aims to improve precipitation forecasts from numerical weather prediction (NWP) models through effective use of satellite-derived precipitation data. Kotsuki et al. (2017, JGR-A) successfully improved the precipitation forecasts by assimilating the Japan Aerospace eXploration Agency (JAXA)'s Global Satellite Mapping of Precipitation (GSMaP) data into the Nonhydrostatic Icosahedral Atmospheric Model (NICAM) at 112-km horizontal resolution. Kotsuki et al. mitigated the non-Gaussianity of the precipitation variables by the Gaussian transform method for observed and forecasted precipitation using the previous 30-day precipitation data.

Kotsuki et al. (2018; JGR-A) extends the previous study and explores an online estimation of model parameters using ensemble data assimilation. We choose a globally-uniform parameters: the cloud-to-rain auto-conversion parameter of the Berry's scheme for large scale condensation. We perform the online-estimation of the model parameter with an ensemble transform Kalman filter by assimilating the GSMaP precipitation data. The estimated parameters improve the analyzed and forecasted mixing ratio in the lower troposphere. Therefore, the parameter estimation would be a useful technique to improve the NWP models and their forecasts.

### Interannual variability of the Australian summer monsoon and its remote influence on East Asian climate

Shion Sekizawa
University of Tokyo
Thursday, 8th Aug 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: In austral summer (boreal winter), interannual variability of convection is centred around Indonesia and northern Australia, representing significant variability in the Australian summer monsoon (AUSM). This study reveals that internal variability unforced by tropical SST anomalies is dominant in seasonal mean strength of the AUSM system. The anomalous AUSM activity forces a distinct wave-train pattern toward the extratropical North Pacific and exerts a significant impact on temperature and precipitation in East Asia.

### Trends and Challenges in TC NWP for the 2020s an update on the 2009 ECMWF Newsletter article "Record-setting performance of the ECMWF IFS in medium-range tropical cyclone track prediction

Mike Fiorino
University of
Monday, 5th Aug 2019
2:00 PM - 3:00 PM
9East seminar room

Abstract: Perhaps the greatest success story in the history of Numerical Weather Prediction (NWP) is the 90% improvement1 in tropical cyclone (TC) track forecasts from the 1980s to 2018. This dramatic improvement was made possible by the high-resolution, high-quality global modeling of the leading operational centres ECMWF and the UKMO.

The talk firsts reviews the history of TC NWP from the early days at Penn State where we made the first TC NWP forecast with operational analyses in a 'full physics' model (MM0 the predecessor to WRF) in the late 1970s to the 2000s, when global models became the clear 'go-to' aid for human forecasters. During the 2000s I was the NWP officer at both the Joint Typhoon Warning Centre (JTWC) and the National Hurricane Centre (NHC) where my main task was to bring good modelling into the human forecast process. It was during my time at NHC that I demonstrated how the deterministic ECMWF HRES forecasts were often superior to the best forecast aid (consensus) and how this significant advance was not because of a horizontal resolution increase but because of (subtle) changes in the physics (see Fiorino 2009: https://www.ecmwf.int/en/elibrary/17493-record-setting-performance-ecmwf-ifs-medium-range-tropical-cyclone-track).

More recently, the limited-area model HWRF, during its 12 years in NCEP operations (2007-2018), has failed to add value to the much lower resolution GFS host global model at the medium range (72 h mean position error). Further, the inability of HWRF to make superior short-range track forecast (12- and 24-h) vis-a-vis ECMWF, despite near 'perfect' HWRF initial position error, challenges conventional thinking on the way forward in TC NWP.

This review of TC NWP and the current performance of HWRF/GFS suggest some next steps/challenges for the 2020s: 1) vortex analysis can model forecasts be improved by a more accurate analysis of the TC vortex? e.g., by assimilation of the truly unique and special observation of the TC forecast centres - 'TC vitals'; and 2) "completing the forecast" by predicting both TC genesis and dissipation. Metrics is common problem for both vortex initialization and genesis. The talk will conclude with a TC 'forecast error' metric proposal, that is more consistent with warnings and TC impacts, and some recent work on deterministic TC genesis verification.

### Improving reconstructions of historical extreme events by rescuing undigitized weather observations with citizen scientists

Ed Hawkins
Monday, 5th Aug 2019
11:00 AM - 12:00 NOON
9East seminar room

Abstract: Our understanding of past changes in weather and climate rely on the availability of observations made over many decades. However, billions of historical weather observations are effectively lost to science as they are still only available in their original paper form in various archives around the world. The large-scale digitisation of these observations would substantially improve atmospheric circulation reconstructions back to the 1850s. Recently, volunteer citizen scientists have been assisting with the rescue of millions of these lost observations taken across western Europe over a hundred years ago. The value of these data for understanding many notable and extreme weather events will be demonstrated.

### On the Environments and Dynamics of Nocturnal Mesoscale Convective Systems

Stacey Hitchcock
University of Melbourne
Wednesday, 31st Jul 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: Mesoscale convective systems (MCSs) play a critical role in the warm season nocturnal precipitation maximum over the United States Great Plains region. However, in some MCSs, convection can redevelop over (i.e. back-build), and/or repeatedly passes over (i.e. train) a region for an extended period of time and contribute to extreme rainfall and flash flooding. Yet, despite their frequency, these events are not particularly well forecast. In the past, this has been due, in part, to a lack of understanding of the complex interactions between convective outflow and the nocturnal environment, and a lack of observations of these environments. The 2015 Plains Elevated Convection at Night (PECAN) field campaign was motivated by unanswered questions about key processes in elevated mesoscale convective systems and the difficulty in accurately forecasting them. During the campaign, 15 MCS environments were sampled by an array of instruments including radiosondes launched by fixed and mobile sounding teams. The first part of this talk will briefly explore analysis of observations collected before and after nocturnal MCS passages during PECAN. The rest of the talk with explore processes that lead to convective organisation conducive to heavy rainfall in an environment with a strong stable boundary layer, with an emphasis on the relationship between convective downdrafts and a strong stable layer in the presence of wind shear.

### Data-driven spectral analysis of climate dynamics

Joanna Slawinska
University of
Wisconsin-Milwaukee
Wednesday, 17th Jul 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: The Earth's system dynamics has an intrinsically multi-scale and nonlinear nature, which fundamentally affects the ability to perform weather forecasting and climate projections. In the face of complex dynamics, conventional data analysis techniques (such as EOF analysis) have frequently ambiguous physical interpretation. Moreover, validation of the extracted signals against observations is oftentimes limited by the time span of the observational record, which can be shorter than the timescales of interest, and also significantly altered by anthropogenic factors. Here, we introduce a new data analysis technique for complex dynamical systems and demonstrate its potential by investigating Indo-Pacific climate variability in models and observations. Through this technique, drawbacks associated with ad-hoc filtering are avoided as the extracted signals span many temporal scales without pre-processing the input data. In addition, we discuss the applicability of our approach in non-stationary signals (i.e. trends), and also describe a modification suitable for skilful extraction of coherent patterns in systems lacking spatiotemporal separability (e.g. turbulence). As an extension of this work, we present a framework for data assimilation and forecasting of non-linear time series.

### On Sub-seasonal Predictability and Dynamical Processes in the Southern Hemisphere

Gilbert Brunet
Bureau of Meteorology
Wednesday, 13th Mar 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: World leading global forecasting systems in 2019 operate with around 20-100 ensemble members and a horizontal resolution in the range 10 to 50km with of order 100 vertical levels. We can currently predict large-scale weather patterns and regime transitions out to a month or more ahead and high-impact events, such as tropical cyclones, out to two weeks ahead. Under certain conditions, even sub-seasonal to seasonal forecasts have some predictive skill. Especially the intra-seasonal variability has relatively to the seasonal variability lot more degrees of freedom and the associated dynamical and physical processes are numerous and complex.

As an example research in the last three decades on tropical modes of ocean-atmosphere interaction, like the El Nino/Southern Oscillation (ENSO) oceanic phenomenon and its complex interaction between the large scale tropical atmosphere and organised moist convection, has permitted significant advances in seasonal prediction. Similarly research in the intra-seasonal variability has point out another ocean-atmosphere interaction mode, the Madden-Julian Oscillation (MJO), as an important source of predictability. MJO modulates significantly the mid-latitude intra-seasonal variability through tropical and mid-latitude teleconnections, like the Canadian winter temperature and precipitation. Another example is the two-way link between the MJO and the North Atlantic Oscillation (NAO) modulates significantly weather regimes over Western Europe and the onset of MJO.

Pushing forecast leadtime to sub-seasonal scale will unfold a new landscape of predictability, model uncertainties and ensemble prediction issues. The richness of the sources of predictability associated with these processes could provide increase predictive skill in all seasons and regions. The associated socio-economic benefits are indeed promising to inform reliably the risk of high-impact weather, including tropical cyclones, droughts, floods, heat waves and the waxing and waning of monsoon precipitation. In this talk we will do an overview of different sources of predictability that could be exploited for increasing the skill of prediction from weeks to months with a focus on the Southern Hemisphere.

### Australian rainfall and El Nino diversity: past variability and context for recent changes

Mandy Freund
CSIRO
Wednesday, 27th Feb 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: The climate system integrates internally and externally induced variability at various time scales as a result of interactions between the ocean and atmosphere. The influence of external forcing on the climate system and with it the structural changes of climate variability, in particular on seasonal and longer time-scales, is difficult to examine due to high natural variability and short observational records. The interplay between high and low-frequency variability restricts our understanding of the full range of climate variability and our ability to contextualise changes. This talk explores and evaluates the potential to use seasonal paleoclimate information to advance our knowledge of natural climate variability and the multi-century context of recent changes in the Australasian and tropical Pacific region.

Climate modes of variability including the El Nino-Southern Oscillation influence Australian rainfall and make Australian rainfall highly variable at interannual timescales. Multi-century reconstructions of past climate variability are developed for Australian rainfall at bi-seasonal resolution. In a multi-century context, the recent drying trends in parts of southern Australia, as well as the tendency towards wetter conditions in northern Australia, are found to be unusual. The cool and warm season rainfall reconstructions allow the documentation of distinct characteristics of past major droughts in terms of their spatial extent, duration, intensity, and seasonality.

Using coral data at seasonal resolution, two El Nino index reconstructions illustrate the sequence of diversity of past eastern and central Pacific El Nino events for the last 400 years. The distinct spatio-temporal signatures of both types of El Nino are exploited, and together with a novel machine learning approach, the diversity of past El Nino events is reconstructed and compared to recent changes.

To further investigate the changes and interactions between Australian rainfall and El Nino diversity, observations and climate model simulations (CMIP5) are compared to the multi- century reconstructions. The context of recent changes provided by the reconstructions in this talk advances our knowledge of natural climate variability in the Australasian and tropical Pacific region and offers new insights into the future climate of the region.

### Understanding the triggering factor of climate vulnerability: climate change impacts in Brazil

Lincoln Alves
CCST INPE
Thursday, 21st Feb 2019
1:00 PM - 2:00 PM
9East seminar room

Abstract: Brazil has a special place in the world in relation to the availability of natural resources. On the other hand, more than 80% of the population lives in urban areas and rapid urban growth has created fragmented spaces with large spatial segregation, exacerbating social inequality and environmental degradation, which have been aggravated by climate change. During the past decades, patterns of precipitation have changed significantly, temperatures have risen by 0.5∞C, resulting in greater instances of extreme events and this has affected humans and natural systems through direct impacts from the events as well as increased forest fires, an increased risk of diseases, water crisis and agricultural production. Overall, its reveal that Brazil is especially vulnerable to climate change. In this seminar, I will present a review study of extreme events in Brazil observed during the last years, with a quantification of the tendencies in each region of Brazil. In addition, I will also present analyses of future climate change projections up to 2030 for Brazil, based on the regional climate change projections. Special attention is dedicated to AmazÙnia and Northeast Brazil. At the end, we reflect on some recommendations for mitigation and adaptation measures and practices that could be raised in the debates and discussions with stakeholders and policy makers.

### About Earth observation data in operational use

Chris Rudiger
Monash University
Wednesday, 20th Feb 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: Satellite data is widely used in operational organisations and applications, including purposes for reanalysis as well as forecasting. An important aspect for this to happen adequately, it is very important to know the opportunities, capabilities, and potential error sources/budgets of satellite products. In this talk, I will present current satellite capabilities in soil moisture remote sensing and provide a background on the error estimates through ground validation, but also talk about the differences between a range of data products. In terms of EO data assimilation, this presentation will focus on work around the assimilation of remotely sensed soil moisture and leaf area data into Meteo France's land surface model and the experience gained and pitfalls we encountered, including a way to analyse model background behaviour - and what we discovered about our model.

Biography: Chris received his B.E. degree in civil engineering from the University of Applied Sciences of Wiesbaden, Germany, in 2002, and the Ph.D. degree in environmental engineering from the University of Melbourne, Melbourne, VIC, Australia, in 2008, with a focus on the potential to assimilate streamflow data into land surface models for soil moisture prediction.

In 2006, he joined the Centre National de Recherches Meteorologiques and the French space agency (CNES), Toulouse (France), focusing on the performance analysis and data assimilation of land surface variables within the French land surface model ISBA in preparation of ESA SMOS mission. He returned to Australia in 2008 to coordinate and lead a number of Cal/Val campaigns for the Australian land validation segment of the SMOS mission in the Australian arid zone and the Murrumbidgee River catchment, and subsequently participated in the Cal/Val campaigns of SMAP. Since 2011, he has been a faculty member with the Department of Civil Engineering, Monash University, Melbourne, VIC, Australia. His current research interests include the algorithm development and subsequent assessment of remote sensing products in vegetation dynamics and landscape water content at high resolution for drought monitoring, wildfire predictions and impact assessments, and the development of EO applications in precision agriculture.

### The uneven nature of precipitation and its changes

Angeline Pendergrass
NCAR
Friday, 8th Feb 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: A few days with heavy rain contribute disproportionately to total precipitation, while many days with light drizzle contribute much less. But it is not appreciated is just how asymmetric this distribution is, and the even more asymmetric nature of trends due to anthropogenic climate change. I will introduce a metric to quantify the asymmetry, and apply it to observations and climate model projections, and reflect on the implications for our understanding of extreme precipitation change. I will also discuss related results documenting that climate models project increased variability of precipitation over most of the world in response to warming. Rather than assuming a future climate with more rain in general, society needs to take measures to deal with a handful of events with much more rain, and little change in between.

Biography: Dr. Angie Pendergrass studies the hydrologic cycle, especially precipitation, and how it responds to climate variability and change. Her work focuses at very large scales, and has addressed global energetic constraints, the distribution of precipitation, and extremes. She is a project scientist at the National Center for Atmospheric Research in Boulder, Colorado, USA.

### Why we should stop explaining science

Phil Dooley
Monday, 4th Feb 2019
3:30 PM - 4:30 PM
9East seminar room

Abstract: How can science to be more captivating than Katy Perry, the footy or Kandy Krush? Conventional approaches to science communication, based on explanations, are losing traction. But there are ways we can engage people from all walks of life and win back trust in evidence-based science. This talk will cover why science can so easily fall into ruts of boring lectures, and explore different approaches that enable science to compete with popular culture and win the support it deserves. The shift will come when those doing the communication realise how research and popular culture are fundamentally different and, with a change of mindset, connect with people on a completely different level.

Biography: Dr Phil Dooley is a science writer, entertainer, pianist and singer. He has performed in science shows and festivals around the world including Glasgow, Sydney, and London, and is bringing his show 'The Most Amazing Planet' in the Universe, an Astronomer's Ode to Earth to Melbourne's Butterfly Club February 4 to 9. By day he has written for Cosmos Magazine, New Scientist, Australian Geographic, Nature and more, and was selected for the 2018 Anthology of Best Australian Science Writing. From childhood dreams of being an astronomer, Dr Phil Dooley progressed to a PhD in laser physics. After a stint as an IT trainer, he returned to science as a communicator, where he could talk about the fun stuff without the heavy-duty equations. In his career he has taken on bored teenage school kids on physics excursions at University of Sydney; created digital content for the international fusion research organisation, JET; been schooled by a battle-hardened Reuters hack in the ANU media office and avoided starvation despite being a freelance science communicator. He runs a YouTube channel - 'Phil Up On Science', and is a qualified professional trainer, with a Certificate IV in Workplace training and assessment. More on Phil at philuponscience.com.au.

### Using convective permitting models to investigate storm events over Africa

Caroline Bain
UK Met Office
Thursday, 31st Jan 2019
11:00 AM - 12:00 NOON
9East seminar room

Abstract: The Met Office Unified Model is used operationally for both weather and climate applications. At higher resolutions within and below the convective grey zone, parameterisation of convection is switched off, allowing the explicit representation of convection. This generally results in better simulations of mesoscale convective systems and has particular benefit in tropical regions where convection is on greater geographical scales. This talk will describe how Met Office models are being used for both research and operational meteorology across Africa. Examples of case studies will be given including African Easterly Waves over West Africa as well as severe storms over Lake Victoria. Finally, a field campaign over Lake Victoria that will have taken place at the end of January 2019 (concurrent with this talk!) will be presented, with the long term implications for model and forecast skill discussed.

Biography: University of Reading - Undergraduate in Meteorology; Met Office scholarship Imperial London - finite element modelling; University of Leeds - PhD in African Easterly Waves (worked on AMMA field campaign in Mali, Niger and Ghana); University of California, Irvine - tracking the ITCZ in the Pacific; UK Met Office (since 2010) and was responsible for the model over Africa. Then worked with National Met Services across Africa for a while before now leading the Tropical Science team (known now as Regional Systems Evaluation team), focussing on regional models, mostly in Southeast Asia and Africa

### Understanding Tropical Precipitation and Atmospheric Circulation Responses to Climate Change

UK Met Office
Wednesday, 30th Jan 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: Future projections of tropical precipitation show large uncertainties at the regional scales relevant to climate change impacts. In order to narrow this uncertainty and produce robust, useable information for climate change adaptation, the mechanisms that drive regional precipitation change in the tropics must first be understood. In particular, the response of the tropical atmospheric circulation to warming is crucial. This seminar will describe results from a number of approaches that have been applied to this problem, including simplified theoretical models and a hierarchy of idealised model simulations.

Biography: PhD at Reading in satellite precip observations over Africa. Then Met Office since 2010 working in the global water cycle group.

### An introduction to Indigenous Astronomy

Krystal De Napoli
Monash University
Wednesday, 16th Jan 2019
10:00 AM - 11:00 AM
9East seminar room

Abstract: Krystal De Napoli is a Kamilaroi woman undertaking an undergraduate degree in Astrophysics at Monash University. Krystal has been delivering a series of public talks on the topic of Indigenous Astronomy, presenting the knowledge gathered from Aboriginal and Torres Strait Islander elders, students, and researchers across Australia. In 2018 she commenced a research cadetship with the CSIRO’s Data61 team, and is proceeding with her research with Muruwari man, William Stevens, into the astronomy of his traditional country. Krystal will be describing some of the ways in which Aboriginal and Torres Strait astronomers use the night sky to further their understanding of their environment. She will be explaining the use of orality in Indigenous culture as the mechanism that encodes knowledge into a transmittable format. Themes include celestial phenomena such as solar eclipses, variable stars, moon haloes, and supernova, and exploring the way that the night sky was used to predict environmental changes.

### Observational Data for Global NWP- the rolling requirements review process of the WMO.

ECMWF
Tuesday, 11th Dec 2018
9:00 AM - 10:00 AM
9East seminar room

Abstract: Ensemble forecasting, at a resolution of 5 km globally, is a realistic goal for Global NWP in 2025, to deliver reliable predictions of the likelihood of occurrence of high-impact, severe weather events up to two weeks ahead. An Earth system modelling approach is being adopted in order to better incorporate the influence of the oceans, land surfaces, snow, sea ice, lakes and dust aerosol on weather. Modelling will also focus on weather patterns in the tropics and their influence on the subsequent development of large-scale weather-regime transitions in Europe, and more generally in the mid-latitudes, as these are associated with heat waves, cold spells and periods of drought or extreme precipitation. These developments lead to significantly more demanding requirements for observations. in terms of globally distributed, high-density measurements of the atmosphere and oceans, as well as the land, snow and ice surfaces.

Biography: Deputy Director of Forecasts at the European Centre for Medium-Range Weather Forecasts (ECMWF) since July 2013. The Forecast Department is responsible for the 24/7 timely production of ECMWF's computer-based forecasts, assessment of forecast quality, software development, liaison with forecast users, user support and data licensing. It also receives, processes and quality controls global observations of the atmosphere and oceans. Erik has many years of experience in data assimilation and numerical weather prediction (NWP). For a period of eight years (2000–2008), as Head of the Data Assimilation Section at ECMWF, he was in charge of the development of the 4D-Var assimilation scheme, acquiring extensive knowledge of the observing systems (terrestrial and space based) and their impacts on NWP. He is the Chair of the World Meteorological Organization (WMO) expert team on Observing System Design and Evolution. He is also the Chair of the scientific expert team of the EUMETNET Observations Programme on the evolution of the European meteorological observing systems. Erik holds a PhD from Stockholm University in Atmospheric Sciences, Dynamic Meteorology.

### High Altitude Ice Crystals and Gas Turbine Engines

Rory Clarkson
Rolls Royce
Tursday, 15th Nov 2018
12:00 NOON - 1:00 PM
9East seminar room

Abstract: The impact that high concentrations of ice particles at cruise altitudes have on jet engines will be described, looking at the physical phenomena that lead to ice accretion and engine damage. The contrast with icing associated with supercooled water is emphasised. A brief history of the ice crystal problem is discussed and what options there are for the future.

### Report on overseas mission: European Meteorological Society Meeting 2018 and UK Met Office Visit

Derryn Griffiths
Bureau of Meteorology
Wednesday, 14th Nov 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: This talk is aimed at staff associated with the Forecast Systems and Guidance Post Processing teams, but may be of broader interest. Gary and Deryn will report on the highlights of their attendance at the recent EMS conference in Budapest, and their respective visits to the UK Met Office. Topics at the EMS conference that we will cover include - Probabilistic post-processing of ensemble NWP guidance for weather forecasting, including global rainfall post-processing by the ECMWF; Post-processing for energy supply and demand modelling including applications of machine learning, and impacts of climate change; Presentation of forecasts and data, and in particular uncertainty and risk forecasts; Reflections on our service compared to other agencies, including the roles of forecast automation, high impact weather focus and meteorologists; Communication of forecasts to maximise their value – 'the last mile'; Agencies and WMO moving to probabilistic impact forecasting (what the weather will do rather than be); Thermal heat stress modelling, sometimes related to climate change and urban design; Attribution of the increased likelihood of weather events to climate change, the best joke and, the best talk.

Topics from the UK Met Office visits will include discussion of the IMPROVER ensemble-based post-processing framework that we wish to collaborate on, and supporting projects, including opportunities for collaboration Forecast and guidance verification Related organisational directions.

### Opportunistic sensing in hydrometeorology

Remko Uijlenhoet
Wageningen University
Tuesday, 13th Nov 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: Traditionally, as hydrologists we have relied on dedicated measurement equipment to do our business (e.g. rainfall-runoff modeling). Such instruments are typically owned and operated by government agencies and regional or local authorities. Installed and maintained according to (inter)national standards, they offer accurate and reliable information about the state of and fluxes in the hydrological systems we study as scientists or manage as operational agencies. Such standard instruments are often further developments of novel measurement techniques which have their origins in the research community and have been tested during dedicated field campaigns. One drawback of the operational measurement networks available to the hydrological community today is that they often lack the required spatial and/or temporal resolution for high-resolution real-time monitoring or short-term forecasting of rapidly responding hydrological systems (e.g. urban areas). Another drawback is that dedicated networks are often costly to install and maintain, which makes it a challenge for nations in the developing world to operate them on a continuous basis, for instance. Yet, our world is nowadays full of sensors, often related to the rapid development in wireless communication networks we have witnessed in the recent past. Let us try to make use of such opportunistic sensors to do our (hydrologic) science and our (water management) operations. They may not be as accurate or reliable as the dedicated measurement equipment we are used to working with, let alone meet official international standards, but they typically come in large numbers and are accessible online. Hence, in combination with smart retrieval algorithms and statistical treatment, opportunistic sensors may provide a valuable complementary source of information regarding the state of our environment. The seminar will focus on some recent examples of the potential of opportunistic sensing techniques in hydrology and water resources, from rainfall monitoring using microwave links from cellular communication networks, via crowdsourcing urban air temperatures using smartphone battery temperatures to high-resolution urban rainfall monitoring using personal weather stations.

### Adventures in Climate and Agriculture

Jaclyn Brown
CSIRO
Friday, 2nd Nov 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: A new team - Weather and Climate Decisions Team – has been established in CSIRO Agriculture and Food. We would like to share some of our research results, details of our on-going projects, and our new business model approaches with AgTech companies. We work across a range of disciplines including grains, cotton, horticulture, sugar, GBR and rangeland grazing. Using climate forecast models in agricultural models (as opposed to statistical forecasts) is quite new and there are many exciting new areas to explore. Our projects range from traditional research resulting in journal papers, to developing on-farm apps or partnering with AgTech companies providing data feeds on a fee for service basis. Our work is strongly influenced by social science and user experience research to ensure the information we provide is truly actionable knowledge. We have also been contributing to the development of a central climate data platform within CSIRO. The platform contains purchased climate data with details of commercialisation rights reducing the number of projects storing or paying for the same data. The platform is also open for external parties to purchase API feeds. We are keen to strengthen the collaborations we already have with the Bureau and look for even more ways!

### Cross-equatorial Hadley circulation and monsoon dynamics in different climates:analogies and differences between mid-Holocene and rcp8.5

Roberta D’Agostino
Max Planck Institute for Meteorology, Hamburg
Monday, 17th Oct 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: Precipitation and circulation patterns of Northern Hemisphere monsoons are investigated in the Coupled Model Intercomparison Projects phase 5 mid-Holocene and rcp8.5 experiments. While these two experiments show similar warming of the Northern Hemisphere and enhanced inter-hemispheric thermal contrast in boreal summer, they result in different monsoon responses. Changes in the spatial extent of the Northern Hemisphere monsoons in rcp8.5 do not exceed the simulated shift in mid-Holocene and only the Indian monsoon is stronger in the rcp8.5 than the mid-Holocene. Investigating similarities and differences in the anomalous moisture budget in these two climates relative to a control climate (piControl) and linking them to differences in net energy input and to the strength of the cross-equatorial Hadley circulation can help to elucidate mechanisms of the monsoon response to different climate forcing. Under mid-Holocene orbital forcing, Northern Hemisphere monsoons widening and wetting are due to thestrengthening of the dynamic component of the moisture budget (e.g. strengthening of the cross-equatorial monsoonal circulation) resulting from the anomalous net energy input over land. On the other hand, monsoons are affected to a different degree by thermodynamic, dynamic and transient eddies components of the budget in enhanced greenhouse gases experiment rcp8.5, with strong regional monsoon behaviors due to local changes in the net energy input. However, the weakening of the cross-equatorial monsoonal circulation in rcp8.5 relative to mid-Holocene represents a limit for the projected expansion and intensification of the monsoonal rainfall with global warming. This process-oriented study improves the confidence on future monsoon projections because it identifies future constraints on monsoon dynamics by comparing and contrasting it with past climate changes. It additionally shows how, despite similar changes in interhemispheric thermal contrast, the mid-Holocene does not represent an analogue for future warming induced by increased greenhouse gases.

### Building capacity for understanding severe weather through citizen science and weather radar data

Joshua Soderholm
Bureau of Meteorology
Wednesday, 22nd Oct 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: Continual improvement and growth of meteorological observations and the rise of citizen scientists on mobile platforms provides an unparalleled potential for capturing the fine-scale complexities of severe weather. In particular, weather radar and public reporting are the cornerstone of severe thunderstorm understanding in both science and industry internationally. However, accessibility and effectiveness of these datasets in Australia remains limited, despite the growing demand and benefits. This seminar introduces two new initiatives from the ARC Centre of Climate Extremes: (1) WeatheX, A cross-platform mobile app for citizen reporting Australian severe weather and (2) the Open Radar Dataset hosted as a NCI national research data collection. The WeatheX app provides a modern, streamlined interface for collecting citizen reports and photos of hail, wind damage, flooding and tornadoes to complement the limited coverage provided by the Storm Spotters and WOW services. The Open Radar Dataset represents a focused effort to recover, rebuild and merge multiple volumetric radar sources into a single consistent archive containing over 700 radar years. Ongoing efforts include satellite calibration of reflectivity, gridding and retrievals (rainfall, hail, cloud tops, precipitation regime and mesocyclone)​. Applications for this dataset will be demonstrated, including long-term climatologies.

### Overview of SSEC/CIMSS Research and Development in support of NOAA and Empowering JPSS/GOES Global Users

Allen Huang
CIMSS/SSEC
Thursday, 18th Oct 2018
2:00 PM - 3:00 PM
9East seminar room

Abstract: SSEC/CIMSS of the University of Wisconsin located at Madison is the home to the father of satellite meteorology and one of NOAA’s cooperative institutes. Since early 1970’s SSEC/CIMSS is actively conducting research and development in optimal usage of meteorological satellite data and innovating sensor technology especially in the development of ground, aircraft, and space -based infrared hyperspectral spectral sounders. This seminar will focus on satellite data collection, algorithm development, processing software, infrastructure enhancement, and research innovation. Specifically, it’ll cover topics such as:

• Community Satellite Processing Package (CSPP) LEO/GEO,
• NOAA’s Direct Broadcast Network (DBNet),
• GeoNetCast America,
• Real Earth visualization system,
• Unified Model Installation,
• GPU high-performance modeling,
• OSE/OSSE for JPSS and,
• Plan for AI and Big Data

Biography:Dr Allen Huang is a scientist at the Cooperative Institute for Meteorological Satellite Studies and Space Science and Engineering Center (CIMSS/SSEC) of University of Wisconsin-Madison, leading remote sensing research in the areas of atmospheric sounding retrieval, information content analysis, satellite and aircraft high-spectral resolution sounding instrument data processing, data compression, instrument design and performance analysis, cloud-clearing, cloud property characterization, synergistic imaging, and sounding data processing and algorithm development.

### New Frontiers in Earth Observation: Overcoming Spatiotemporal Divides with UAVS and Cubesats

Matthew McCabe
King Abdullah University
Wednesday, 26th Sep 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: The last few years has witnessed a paradigm shift in how we undertake earth observation. Traditional remote sensing has always required a compromise between spatial and temporal resolution. That is, one could obtain high-resolution imagery occasionally; or frequent imagery that is spatially coarse. These spatiotemporal constraints are increasingly being removed via the emergence of new earth observation platforms, such as those being afforded by unmanned aerial vehicles (UAVs) and constellations of small space-based CubeSat systems. Here we will review some recent developments in these technologies across the broad area of environmental monitoring, but with some focused attention on applications in precision agriculture, hydrology and plant phenotyping. Some of the outstanding challenges and opportunities that these technologies present will also be discussed.

### Towards understanding model uncertainties in aerosol forcing

Stephanie Fiedler
Max Planck Institute for Meteorology, Hamburg
Wednesday, 12th Sep 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: The radiative forcing of anthropogenic aerosol is important for understanding climate change. Climate models, however, show a persistent spread in aerosol forcing, despite decades of research on aerosols and their effect on climate. The reasons for the uncertainty are abundant, but their relative importance is typically hidden by the model complexity. Reasons include difficulties in accurately simulating properties affecting the atmospheric radiation transfer (e.g., Crüger et al., 2018) and processes that control the aerosol burden (e.g., Fiedler et al., 2016). Moreover, observational climatologies for aerosol disagree (e.g., Bellouin et al., in prep.) and aerosol effects cannot be disentangled from the meteorological development in observations (e.g., Stevens and Feingold, 2009).In this presentation, we present results of inter-comparison studies for characterising the model diversity in aerosol. We than look at how process-based assessments and deliberate reductions of model complexity aid developing a better understanding of model uncertainties. First, we focus on process-based evaluations of weather and climate models in simulating dust storms. The meteorological drivers, importance for forming ice clouds, as well as anthropogenic changes of dust aerosol are currently poorly understood. Presented are the first climatological assessments of different meteorological processes for dust-storm formation and touch on dynamical processes that are misrepresented or completely missing in models with parameterised moist convection (e.g., Fiedler et al., 2013, 2015, Heinold et al., 2013). Using this knowledge for process-level model evaluations proofs useful in understanding differences in dust/aerosol emission from re-analyses and the CMIP5 configuration of HadGEM2-ES (Fiedler et al., 2016, Huneeus et al., 2016).

Secondly, we investigate the impact on forcing from uncertainties that remain, when we pretend to know the spatio-temporal distribution of aerosols. For doing so, an ensemble of atmosphere-only experiments of a selection of new CMIP6 model configurations (Fiedler et al., 2017, submitted) are presented. All five models use the same prescribed aerosol optical properties and an associated effect on clouds from the new observationally informed simple-plumes parameterisation MACv2-SP (Fiedler et al., 2017, Stevens et al., 2017). The highlights of the model inter-comparison are a (1) strong impact of model internal variability on estimating the effective radiative forcing (ERF), (2) a small change in ERF from a substantial spatial shift of anthropogenic emissions from Europe and the U.S. to East Asia, and (3) persistently large model diversity in clouds on the macro- and micro-scale. MACv2-SP captures the aerosol development from the pre-industrial (Stevens et al., 2017) to the end of the 21st century, based on CMIP6 emission scenarios (Fiedler et al., in prep.). In the future, MACv2-SP will be used for CMIP6 experiments, e.g., in the radiative forcing model inter-comparison project (RFMIP, Pincus et al., 2016), and for other scientific endeavours, e.g., seasonal to decadal climate predictions.

### Climate Data Records for SST: Improving Accuracy, Uncertainty and Quality Estimates

Jonathan Mittaz
Tuesday, 4th Sep 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: Climate Data Record (CDR) for Sea Surface Temperature (SST) needs to satisfy quite stringent limits on accuracy and stability, and also requires a long enough time series to enable the study of climate change. As such, creating SST CDRs is challenging, not least because the sensors which have the longest continuous time coverage, such as the AVHRR, were not originally designed to have high accuracy. In this talk, I will first describe our approach to improving Level 1 data (radiances and brightness temperatures) which uses the techniques of metrology to provide accuracy improvements and which include traceable uncertainties. I will also briefly discuss the extension of metrological techniques to SST data itself. I will then discuss one of the major European initiatives to provide a set of SST CDRs under the ESA Climate Change Initiative program which provides SST CDR data from Level 2 (satellite swath data) to Level 4 (global, daily average data). I will then briefly discuss other projects related to CDR quality which are currently underway to provide improved quality information to CDR users.

### The Local Ensemble Tangent Linear Model - Accelerated with order Reduction

Craig Bishop
University of Melbourne
Thursday, 6th Sep 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: A leading Data Assimilation (DA) technique in meteorology is 4DVAR which relies on the Tangent Linear Model (TLM) of the non-linear model and its adjoint. The difficulty of building and maintaining traditional TLMs and adjoints of fully coupled ocean-wave-atmosphere-etc models is somewhat daunting. On the other hand, coupled model ensemble forecasts are readily available. Here, we show how a previously described ensemble-based method for generating TLMs can be accelerated via a rank reduction method that uses a principle component reduction of a climatology of Local TLMs. The resulting Local Ensemble TLM Reduced using a Climatology of Local TLMs (LETLM-RCL) features a low rank projection of some local influence region containing all the variables that could possibly influence the time evolution of some target variable(s) near the center of the region. We prove that high accuracy is guaranteed provided that (i) the ensemble perturbations are governed by linear dynamics, and (ii) the number of ensemble members exceeds the number of retained eigenvectors required to explain a high fraction of the climatological variance of LETLMs. The LETLM-RCL approach is faster than the LETLM approach for two distinct reasons. First, key matrix inversions are speeded by the cube of the ratio of the number of variables in the local influence volume divided by the number of retained eigenvectors. Second, the required ensemble size is reduced to just a few members larger than the maximum number of eigenvectors retained for any model grid point. The approach is illustrated with the aid of a simple coupled model. In the case examined, an LETLM-AR is obtained whose accuracy is very close to the original LETLM but it only requires 14 ensemble members instead of 28 ensemble members and the cost of the matrix inversions is reduced by 1-2 orders of magnitude depending on the model level.

### From qualitative to quantitative hail nowcasting

Aurora Bell
Bureau of Meteorology
Wednesday, 29th Aug 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: The science is not yet ready to deliver the quantitative hail size estimate from radar that are required for a qualitative warning service. The current state of the art of radar based hail detection is semiquantitative with substatial error bias. The Bureau is committed to develop a nowcasting service that would have an immediate positive impact on end-users needs. Therefore, the choice is to deliver semi-quantitative products using proven algorithms immediately, and work with the end-users so that they can make optimum use of these products.

The paper will present the Bureau's road map of enhancing hail nowcasting. Major mile-stones are:

• Improve the single-pol hail service. The Bureau developed gridded products of probability of hail of specific size based on reflectivity (single-pol) and integrated them in Rainfields3
• Adopt the NCAR Particle ID algorithm for Dual-Pol data and integrate it in Rainfields3
• Tune the two types of algorithms (dual-pol and single-pol) for consistency using case-studies and workshops with end-users
• Develop training and evaluate the changes.

The "6th element"; is to motivate the science community to improve the current state of the art, which will require the following:

• Improvement of the understanding of how to deliver an impact-based hail warning service,
• Developing of radar-based algorithms to estimate and nowcast hail size,
• Developing in-situ observations of hail-size.

### Non-linearity and the assimilation of bounded variables

Craig Bishop
University of Melbourne
Wednesday, 15th Aug 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: Unbiased observations of bounded variables can be shown to exhibit observation error standard deviations that depend on the value of the unknown true state. In particular, the observation error variance of such observations must tend to zero as the unknown true state tends to zero. Furthermore, observed bounded variables such as wind-speed, clouds, precipitation, fire and ice are highly non-linear functions of mass, momentum, and moisture variables. To simultaneously address both of these issues, we extend the previously developed Gamma, Inverse-Gamma and Gaussian (GIGG) variation on the EnKF to better account for non-linearity. Specifically, we augment the linear regression used by EnKFs to update model state variables from observed variables with a local, low-dimensional, non-linear variational step. In a synthetic Tropical Cyclone wind energy assimilation problem, the approach is shown to profoundly reduce the analysis errors associated with Tropical Cyclone wind fields. The talk will also summarize some other recent improvements to the GIGG filter. In addition, I will comment on some of the challenges that I ran into when trying to incorporate some of the features of the non-linear GIGG filter into global variational data assimilation frameworks such as 4DVar.

### Understanding connections of tropical Pacific-Arctic is necessary to explain more frequent occurrence of Central Pacific El Nino

Sang-Wook Yeh
Hanyang University
Wednesday, 8th Aug 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: El Nino, which is the most dominant mode of climate variability on Earth, has different faces in terms of its amplitude, frequency and spatial pattern. In particular, the diversity of El Nino’s spatial structure and amplitude significantly changes the impacts of weather and climate variability including droughts, floods, cold and heat waves worldwide through atmospheric teleconnections.. However, there is still a debate about the physical processes leading to the El Nino diversity. In this study, we examine the mechanism associated with the occurrence of Central Pacific (CP) El Nino using an ocean general circulation model. It is found that the role of weather noise plays a role and we hypothesize that the extra-tropical atmospheric circulation may modulate the statistics of weather noise in the tropics. We found that the North Pacific Oscillation, which is the second dominant mode of atmospheric circulation in the North Pacific, is associated with more frequent occurrence of Central Pacific El Nino. Furthermore, the North Pacific Oscillation is also associated with the variability of sea-ice concentration in Arctic. Therefore, we argue that understanding connections of tropical Pacific-Arctic via the atmospheric circulation in the North Pacific is necessary to explain El Nino diversity. This implies that the sea-ice reduction in Arctic may influence the El Nino properties including its spatial pattern in a future climate.

### Improving Wind Forecasts based on Verification and Discussions with Forecasters

Deryn Griffiths
Bureau of Meteorology
Wednesday, 27th Jun 2018
10:00 AM - 11:00 AM
9East seminar room

Abstract: Wind forecasts are available from:

• Direct NWP,
• Operational Consensus Forecasts (OCF) which are a bias corrected consensus of various NWP, and as,
• Official forecasts issued by meteorologists in our forecasting centres.

Forecasters base the Official wind forecasts on NWP and OCF but adjust them to allow for known or perceived short-comings, as well as to tailor a public message.

I will present some verification of the Official and OCF Wind Speed forecasts at automatic weather stations. At mountainous sites, the Official forecasts showed substantially smaller errors than OCF. Following discussions with forecasters, we identified a part of their technique which was amenable to systemic application and ran a hindcast for 12 months. I will describe the systematic modification undertaken and show that the results provide errors which are similar to, or smaller than, the errors of the Official forecasts.

We have made improvements to tools available in forecasting centres which will allow forecasters to provide Official forecasts with less effort than previously, but maintaining the current standard. I will touch on the aspects involved in tailoring a public message which mean that the verification presented is not the whole story. Nonetheless, we expect some forecasts will improve thanks to the information available from the verification and the associated improvements in GFE tools.

### Differentiating regional climates at 1.5°C and 2°C of global warming

Andrew King
University of Melbourne
Wednesday, 20th Jun 2018
10:00 AM - 11:00 AM

Abstract:The Paris Agreement aims to keep global warming below 2°C above pre-industrial levels with a preferential 1.5°C target. Given the great focus on these targets it is crucial that we understand how and why local climates will differ between these targets. I will discuss the merits and disadvantages of different methods for Paris-target analyses. I will then present the use of a CMIP5-based time-sampling method and illustrate its utility in answering the questions:

• Do local and regional temperatures simply scale from 1.5°C to 2°C or do non-linearities prevent a simple scaling approach from being reasonable?
• Who would experience the largest local climate changes from 1.5°C to 2°C of global warming?

I will conclude by briefly discussing my new 3-year project working with the Bureau of Meteorology on the seasonal prediction of rainfall extremes in Australia.

### BARRA: A regional atmospheric reanalysis for Australia

Doerte Jakob, Chun-Hsu Su, Nathan Eizenberg
Peter Steinle
Bureau of Meteorology
Wednesday, 16th May 2018
10:00 AM - 11:00 AM

Abstract: The Bureau of Meteorology is currently undertaking a high-resolution atmospheric reanalysis for Australia. The plans were outlined in September 2017, and this seminar provides an overview of the results obtained so far.

BARRA is 12-km regional reanalysis (BARRA-R) over Australia and the surrounding region, including New Zealand and southeast Asia. It also consists of multiple convective-permitting (1.5-km) dynamical downscaling analyses (BARRA-xx) over smaller sub-domains, nested within BARRA-R, to provide additional fine-scale detail. They are developed based on the UK Met Office's regional reanalysis and Bureau's numerical weather prediction systems, Australian Community Climate Earth-System Simulator (ACCESS)-R and ACCESS-C.

In 2019, BARRA will provide over 25 years of reconstructions of past atmospheric conditions, with plans to extend the reanalysis back in time to 1990. The Reanalysis will represent a suite of high-resolution gridded climate datasets and as such will prove valuable for a large range of applications with the first 6-year time slice (2010-2015) available for academic use and internal use.

About 100 parameters at the surface level (such as temperature, precipitation, relative humidity, wind, evaporation, soil moisture), the pressure and model levels (such as temperature, relative humidity, wind) at hourly time step are available through the National Computing Infrastructure (NCI).

In this presentation, we will report on the status of the project, report an initial evaluation of key variables, and demonstrate how to access the reanalysis dataset on NCI. We are keen to collaborate to undertake more detailed evaluations. This knowledge will guide the appropriate use of reanalysis data, will provide guidance in developing and post-processing the reanalysis products and guide future development.

External: http://www.bom.gov.au/research/projects/reanalysis/

Email: helpdesk.reanalysis@bom.gov.au

### Time Series Analysis in Operational Remote Sensing: Algorithm MAIAC and its Applications

Alexei Lyapustin
NASA Goddard SFC
Monday, 9th Apr 2018
10:00 AM - 11:00 AM

Abstract: The Multi-Angle Implementation of Atmospheric Correction (MAIAC) is a new algorithm combining the time series analysis and contextual (spatial) processing for cloud detection and retrieval of atmospheric aerosol and surface bidirectional reflectance properties. MAIAC takes advantage of different time/space variability of land surface reflectance and aerosols to separate their respective contributions in the measurements. MAIAC is currently an operational MODIS algorithm providing a suite of atmospheric and surface products at 1km resolution on global Sinusoidal grid. The products include cloud mask, column water vapor, aerosol optical depth and type, surface BRDF and spectral reflectance (BRF) over land and water leaving radiance over water. When snow is detected, we derive sub-pixel snow fraction and snow grain size. MAIAC is being adapted to VIIRS and geostationary sensors (AHI and GOES-R). Due to robust cloud and snow detection and high 1km resolution of aerosol product with low urban bias, MAIAC products are widely used the air quality research and land studies. I will give an overview of MAIAC processing and discuss some applications.

### Monitoring the changing atmosphere with satellite data

Caroline Poulsen
RAL Space UK
Wednesday, 4th Apr 2018
10:00 AM - 11:00 AM

Abstract: Human activity is changing the composition of our atmosphere. Aerosols can block the suns energy leading to a net cooling effect, aerosols can also interact with clouds providing the seeds for cloud formation where the effect on climate is much more uncertain. Methane and tropospheric ozone are both a potent greenhouse gases. The ESA's (European Space Agency) Climate Change Initiative (CCI) has enable satellite records of atmospheric composition to be created, going back nearly three decades. These satellite records will help improve our understanding of how our atmosphere's composition is changing and how these changes are affecting our climate. The presentation will give a short introduction to the ESA Climate Change Initiative programme and then focus on 4 satellite climate records, aerosol, clouds, methane and ozone which have been produced with the involvement of the Remote Sensing Group in RAL (Rutherford Appleton Laboratory) Space in the UK. The presentation will look at the motivation for creating the climate record, the algorithms, validation, early results and trends and some recent scientific analysis that has been performed with the data sets. The Remote Sensing group employs innovative optimal estimation algorithms and applies them to predominantly research and metrological passive satellites to create climate records. The group is also interested in using satellite data in Near Real Time to address air quality applications.

### Radar Observations and Simulations of the Level of Maximum Detrainment

Gretchen Mullendorf
University of North Dakota
Tuesday, 3rd Apr 2018
2:00 PM - 3:00 PM

Abstract: Parcel theory is the basis for many of convective indices (e.g., CAPE, LNB) that are used extensively throughout the community, ranging from short-term severe weather forecasting to parameterized convection in climate models. While it is widely known that parcel theory is only an approximation, and therefore the indices and algorithms that derive from parcel theory are also just approximations, much more research is needed so that we can better link our theory to observations. One example of this need is the relationship between the level of neutral buoyancy (LNB) and the level of maximum detrainment (LMD; the height at which storms are observed to become neutrally buoyant). Prior modeling studies have demonstrated how departures from LNB differ with storm morphology. Subsequently, several radar-based methodologies have been built to test the model findings. Additionally, preliminary findings from a large-scale radar analysis project quantify the difference between tropopause height (used as an estimate for LNB) and LMD for different seasons and different regions over the U.S. Initial steps towards building new convective entrainment models using LMD information will be discussed.

### Mean state dependence of ENSO atmospheric feedbacks and ENSO dynamics in climate models

Tobias Bayr
GEOMAR, Germany
Thursday, 15th Mar 2018
1:00 PM - 2:00 PM

Abstract: El Niño/Southern Oscillation (ENSO) is a coupled ocean-atmosphere phenomena in the tropical Pacific and is the most dominant climate variability on interannual time scales. Due to its large socio-economic impact by causing severe extreme weather events such as floods and droughts, it is very important to accurately predict how ENSO will change under global warming. Despite improvements have been made in simulating ENSO over the last decades, realistic representation of ENSO and its projection under global warming remains highly uncertain and varies strongly amongst climate models of the 5th phase of the Coupled Model Intercomparison Project (CMIP5). The ENSO atmospheric feedbacks (AF), namely the positive (amplifying) zonal wind feedback and the negative (damping) heat flux feedback, have been identified as major contributors to the diversity of simulated ENSO, as they are strongly underestimated in many state-of-the-art climate models. This is caused by a sea surface temperature (SST) cold bias in the western Pacific (a common problem in climate models) that shifts the Walker Circulation to the west (up to 30° in longitude) into a La Niña-like mean state and weakens ENSO AF. As the position of the Walker Circulation determines the strength of both types of ENSO AF, this explains the compensating effect between the two AF in climate models. Further, a too westward position of the Walker Circulation in climate models changes ocean-atmosphere coupling in the tropical Pacific from a predominantly wind-driven to a more shortwave-driven mode and causes different ENSO dynamics than observed. As a result these climate models simulate an ENSO variability that looks statistically not too different from observations, but arises due to very different dynamics. In contrast, climate models with a reduced cold bias show enhanced ENSO AF, which leads to a substantial improvement of important ENSO properties such as the phase locking of ENSO to the seasonal cycle, the asymmetry between El Niño and La Niña and ENSO teleconnections to the North Pacific.

### Climate Change Impacts on South-Eastern Asia Monsoonal Seasons

Bertrand Timbal
NEA, Singapore
Thursday, 8th Mar 2018
2:00 PM - 3:00 PM

Abstract: South-East Asia (SEA) is affected by two monsoonal seasons: South-West (SW) from May to September and North-East (NE) from November to March. This seasonality corresponds to the march of the Inter-Tropical Convergence Zone (ITCZ) across the Western Maritime Continent (WMC). Therefore, the investigation of future projections of changes in the monsoonal systems across SEA is a complex undertaking: two separate seasons, contrasted regions of high and low rainfall and interactions between several large-scale processes (the march of the ITCZ) and synoptic scale events. When zooming down to the consequences of the changes in monsoonal system for land locations across the WMC, the degree of complexity is further increased due to the poor resolution of the complex land-sea mask and orography within the region by climate models with resolution in the vicinity of 100-200 km. In this presentation, we will review what can be learnt from a careful assessment of CMIP5 climate stimulations and a further dynamical downscaling of a selection of GCMS from this database. Climate models tend to have a reasonable representation of key climatological features of the NE monsoon circulation, superior to the SW monsoon circulation. Sub-seasonal variability is also reasonably well represented: e.g. GCMs are able to capture cold surges. However, there are large and varied biases in the intensity and spatial distribution of rainfall over Southeast Asia during the monsoon seasons.

### The representation of convection in the Unified Model at 2km to 100m gridlength

Kirsty Hanley
UK Met Office
Wednesday, 7th Mar 2018
10:00 AM - 11:00 AM

Abstract: Convective storms, and their associated hazards (flash floods, hail, lightning and severe winds), are a crucially important forecasting problem. Such events can have wide-ranging impacts on livelihoods and infrastructure, so the timing and location of convective storms, as well as their evolution, are important to forecast accurately. Over the last decade, many operational forecast centres, including the Met Office, have moved towards running numerical weather prediction models at order 1km gridlength for short-range weather forecasting. In these models convection is represented explicitly rather than by a convection parameterisation. However, these gridlengths are unable to fully resolve the individual convective elements leading to convection still being under resolved.

This talk will be split into two halves, focussing on convection in kilometre-scale and sub-kilometre scale versions of the Unified Model (UM). Firstly we present 2.2km UM ensemble simulations of one of the cases from the 2017 NOAA Hazardous Weather Testbed. All of the 2.2km simulations were found to have less CIN than the observed soundings. The member which produced the best simulation of the supercells which developed over Oklahoma on this day had more CIN than the other members which allowed convection to initiate later and CAPE to build up. However, comparisons with surface station data found that this member was too warm and too dry. This suggests there are compensating errors in the factors controlling initiation in kilometre-scale models which comparisons with observations can help ellucidate. In the second half of the talk we focus on convection in sub-kilometre versions of the UM. While some aspects of convection are improved by going to higher resolution (such as initiation time), there are still problems – in particular O(100m) models tend to produce showers which are too narrow. This may be a sign that the explicit turbulence is not taking over from the subgrid turbulence scheme appropriately.

### Towards a more physically consistent stochastic representation of model uncertainty in the IFS

Sarah-Jane Lock
ECMWF
Tuesday, 6th Mar 2018
2:00 PM - 3:00 PM

Abstract: Since 1998, the ECMWF ensemble forecasts have included some stochastic representation of the model uncertainties associated with atmospheric processes. The focus has been on model uncertainties that arise from the parametrisations of atmospheric physics processes, which simulate sub-grid or unresolved processes. Such uncertainties have been represented via the Stochastically Perturbed Parametrisation Tendencies (SPPT) scheme, which contributes significantly improved forecast skill in the IFS at all forecast ranges.

In this talk, we will describe recent developments to improve the physical consistency of the SPPT scheme. The model response to these SPPT perturbations will be used as a baseline for exploring model sensitivities to a new and alternative approach – the Stochastically Perturbed Parametrisations (SPP) scheme. Where SPPT acts by perturbing the total tendencies from the suite of physics schemes, SPP introduces random perturbations to quantities within individual physics schemes, thereby taking the representation of model uncertainty closer to its known sources. The talk will discuss our progress with this new SPP approach and will include an outline of our plans for additional stochastic representations of model uncertainties.

### Very high resolution NWP modelling: fog and mixed phase clouds

Ian Boutle
UK Met Office
Tuesday, 6th Mar 2018
11:00 AM - 12:00 NOON

Abstract: This seminar will be presented in two parts.

In the first part, we will discuss current implementations of sub-km scale Unified Model configurations around the world. We begin with the London model, a 333m NWP model run operationally over the south-east UK, primarily aimed at improving fog forecasts at key airports. We will discuss the behaviour of the model and the mechanisms by which the enhanced resolution improves forecasts of near surface wind-speeds, temperatures, cloud cover and fog. We will then discuss the relocation of this model to San Francisco, as part of the US aviation weather testbed program in the summer of 2017, and the use of the model for forecasts of winter fog in Delhi.

In the second part, we will discuss aircraft observations and model simulations of a cold-air outbreak, demonstrating how secondary ice production in mixed phase clouds is crucial in determining the breakup of stratoform cloud into convective cloud. We will show how the model struggles with this transition, typically having too little supercooled liquid in the stratiform region and a too-fast onset of convection. We will discuss attempts to improve the model simulation, and the importance of this work for the Southern Ocean and its typical warm-bias found in climate models.

### A stochastic framework for modeling the population dynamics of convective clouds

Samson Hagos
FCSD/ASGC
Monday, 5th Mar 2018
10:AM - 11:00 AM

Abstract: A stochastic prognostic framework for modeling the population dynamics of convective clouds and representing them in climate models is proposed. The approach used follows the non-equilibrium statistical mechanical approach through a master equation. The aim is to represent the evolution of the number of convective cells of a specific size and their associated cloud-base mass flux, given a large-scale forcing. In this framework, referred to as STOchastic framework for Modeling Population dynamics of convective clouds (STOMP), the evolution of convective cell size is predicted from three key characteristics: (i) the probability of growth, (ii) the probability of decay, and (iii) the cloud-base mass flux. STOMP models are constructed and evaluated against CPOL radar observations at Darwin and convection permitting model (CPM) simulations.

Multiple models are constructed under various assumptions regarding these three key parameters and the realisms of these models are evaluated. It is shown that in a model where convective plumes prefer to aggregate spatially and mass flux is a non-linear function of convective cell area, mass flux manifests a recharge-discharge behavior under steady forcing. Such a model also produces observed behavior of convective cell populations and CPM simulated mass flux variability under diurnally varying forcing. Besides its use in developing understanding of convection processes and the controls on convective cell size distributions, this modeling framework is also designed to be capable of providing alternative, non-equilibrium, closure formulations for spectral mass flux parameterizations.

### Interannual Variability of the MJO teleconnections to the North Atlantic

Steve Woolnough
Monday, 19th Feb 2018
2:00 PM - 3:00 PM

Abstract: The Madden-Julian Oscillation (MJO)is the leading mode of tropical intraseasonal variability and the diabatic heating anomalies and associated divergent circulation can act as a source of extra-tropical Rossby Waves which give the MJO a global influence. These teleconnections provide an important source of sub-seasonal predictability for the extra-tropics. One important teleconnection is between the MJO and the North Atlantic including the North Atlantic Oscillation. This seminar will discuss the dependence of this teleconnection to the North Atlantic on the interannually varying basis state associated with ENSO, and the implications for sub-seasonal prediction and the impact of basic state errors in model representations of this teleconnection.

### Efforts to underpin regional climate prediction and services in Southern South America

Marisol Osman
Centro de Investigaciones
del Mar y la Atmósfera
Thursday, 15th Feb 2018
3:00 PM - 4:00 PM

Abstract: WMO has established a regional climate centre for southern South America leaded by the National Weather Services of Argentina and Brazil. Currently, several projects are being implemented to underpin climate services in the region in an interdisciplinary and intersectoral framework. In the first part of the talk, the scientific background for the development of climate monitoring and prediction tools in southern Southern South America being done in the context of CLIMAR and CLIMAX projects, will be presented. In particular, I will show the main outcomes of the predictability studies on seasonal timescales, included prediction skill assessment, obtained for surface air temperature and precipitation from an ensemble of climate models. Southeastern South America and the Central Andes region show the highest levels of skill and predictability, related to the remote influence of tropical ocean variability. On subseasonal timescales, I will describe the influence of MJO over the region, and in particular over the activity of the leading pattern of rainfall variability. This influence is exerted not only by the MJO passage over the tropical region but also through energy propagation along extratropical arc-shaped wave trains. During the rainy season -October to April- the phases of the MJO can be clearly associated to the wet or dry phase of the leading pattern. Then, I’ll introduce the techniques we are implementing to develop monitoring and prediction tools for Southern South America on subseasonal and seasonal timescales. Finally, I will present the efforts done to involve different societal actors in the development climate prediction tools with focus on the agricultural sector. In particular, I’ll introduce the co-design and co-production approach we are implementing with small farmers living in Southeastern South America and the collaborations established with the Argentine National Weather Service and Regional Climate Service for Southern South America to improve the social appropriation of the new climate products.

### Deep Learning algorithms for Climate Predictions

Yoo-Geun Ham
Department of Oceanography
Chonnam National University
Wednesday, 14th Feb 2018
10:00 AM - 11:00 AM

Abstract: It is quite recent that the deep learning algorithms have been widely applied to fields including computer vision, speech recognition, natural language processing, audio recognition, machine translation, etc. As one of first attempts to apply the deep learning algorithms for climate predictions, the ENSO prediction using convolutional neural network (CNN) will be presented in this talk. First, the performance of the CNN model is tested in the perfect model framework, and the forecast experiments confirm the CNN guarantees the skillful ENSO forecast than that using the simple linear model (i.e. multiple linear regression (MLR) model). The correlation skill of the Nino3.4 index using CNN model is 0.69 for 8-month lead forecast, which shows systematically superior than that using the MLR model (i.e. less than 0.4). It is also found that the CNN model generally shows better performance in predicting El Niño whose time-evolution is quite different from that in the typical El Niño.

The CNN model is also applied to predict real ENSO cases during 1981-2010. Due to the limitation in the number of samples in the observation, we utilized long-term integrations of the multiple numbers of AOGCM participated in CMIP5 to train the CNN model. The comparison of the forecast skills of the CNN models with the dynamical models participated in multi-model prediction projects (i.e. NMME), the CNN model exhibits comparable forecast skills with the NMME for 8-month lead forecast. However, it is found that the CNN model has ability to predict real ENSO events up to 17-month lead forecasts to some extent. The possible interpretation for this long-term prediction using the CNN model will be partly given in this talk.

### Tropical ocean decadal variability and connections to Antarctic and Arctic sea ice

Gerald A. Meehl
NCAR
Tuesday, 13th Feb 2018
1:00 PM - 2:00 PM

Abstract: Though there were earlier claims that “global warming stopped” after that late-1990s, it is now better understood that the slowdown in the rate of global surface temperature increase in the early 2000s was symptomatic of contributions to decadal climate variability associated with the Interdecadal Pacific Oscillation (IPO) superimposed on a long-term warming trend from increasing greenhouse gases. However, the IPO has influences on other parts of the climate system in addition to global surface temperatures. For the Antarctic, convective heating anomalies in the tropical Pacific from the negative phase of the IPO drove atmospheric circulation anomalies and a preponderance of northward surface winds around Antarctica that contributed to the increasing Antarctic sea-ice extent from 2000-2014, with secondary contributions from the tropical Atlantic and SPCZ regions. For the Arctic, decreasing observed sea ice extent trends accelerated after about 2000 when the IPO transitioned from positive to negative. For the cold season (NDJF) there was a connection to Arctic region circulation and sea ice anomalies associated with negative IPO and negative convective heating anomalies in the tropical Pacific. For the warm season (JJAS) there was a stronger Arctic connection to positive convective heating anomalies in tropical Atlantic observed in association with positive SST trends in that basin. Recent decreases of Antarctic sea ice extent, starting in SON 2016, show a connection to positive convective heating anomalies in the equatorial eastern Indian Ocean and a possible transition of the IPO from negative to positive.

### Statistical-dynamical typhoon intensity predictions in the western North Pacific using track pattern clustering and ocean coupling predictors

Il-Ju Moon
TRC, Jeju National University
Monday, 12th Feb 2018
10:00 AM - 11:00 AM

Abstract: A statistical-dynamical model for predicting tropical cyclone (TC) intensity has been developed using a track-pattern clustering (TPC) method and ocean-coupled potential predictors. Based on the fuzzy c-means clustering method, TC tracks during 2004-2012 in the western North Pacific were categorized into five clusters and their unique characteristics were investigated. The predictive model uses multiple linear regressions, where the predictand or the dependent variable is the change in maximum wind speed relative to initial time. To consider TC-ocean coupling effects due to TC induced vertical mixing and resultant surface cooling, we also developed new potential predictors for maximum potential intensity (MPI) and intensification potential (POT) using depth-averaged temperature (DAT) instead of sea surface temperature (SST). Altogether, we used six static, 11 synoptic, and three DAT-based potential predictors. Results from a series of experiments for the training period of 2004 - 2012 using TPC and DAT-based predictors showed remarkably improved TC intensity predictions. The model was tested on predictions of TC intensity for 2013 and 2014, which are not used in the training samples. Relative to the non-clustering approach, the TPC and DAT-based predictors reduced prediction errors about 12-25% between 24-h and 96-h lead time. The present model is also compared with four operational dynamical forecast models. At short leads (up to 24 hours) the present model has the smallest mean absolute errors. After a 24-hour lead time, the present model still shows a good skill comparable to the current, best operational model.

### The development and validation of a high resolution soil moisture analyses system

Vinod Kumar,
and Paul Fox-Hughes
Bureau of Meteorology
Tuesday, 6th Feb 2018
10:00 AM - 11:00 AM

Abstract: We discuss the development and evaluation of the prototype JULES based Australian Soil Moisture Information (JASMIN; Dharssi and Vinodkumar, 2017). JULES is the land surface model used in Bureau's ACCESS numerical weather prediction and seasonal forecasting systems. JASMIN has been developed primarily to provide accurate soil moisture estimates at high resolution for fire and land management. JASMIN is capable of producing hourly analyses of soil moisture and other land surface fields at a 5 km resolution. The JASMIN soil moisture product is evaluated against ground based networks and existing methods used in operational fire danger ratings. Among the results, the median Pearson’s correlation for surface soil moisture across the observation networks is found to be between 0.78 and 0.85. JASMIN is generally found to have better skill than the Keetch Byram Drought Index and Mount’s Soil Dryness Index models used operationally in Australia.

To facilitate the potential use of JASMIN in the current operational fire danger rating system, a number of rescaling approaches are applied and validated, including: minimum-maximum matching, mean-variance matching, and cumulative distribution function (CDF) matching. Validation of the rescaled products is performed using ground based soil moisture observations and MODIS fire radiative power data. Based on this evaluation, we argue that the minimum-maximum approach is the most appropriate calibration method for rescaling JASMIN output to suit current fire danger rating systems.

The rescaled land surface model product may be a valuable resource for fire agencies as it enables them to readily utilize these outputs in their existing fire prediction models. While JASMIN outputs can be adapted to suit the current operational fire danger rating systems in Australia, its potential application is greatest in the National Fire Danger Rating System (NFDRS), currently being prototyped across Australia. The NFDRS has a modular design, with new and updated modules expected to replace older or less effective ones. Individual fuels models can be updated to accept JASMIN data as inputs, and replaced in the NFDRS as they become available without requiring an overhaul of the whole system. The ability of JASMIN to estimate accurate soil moisture at several layers is expected to be advantageous in the NFDRS.

Reference: Dharssi, I., and Vinodkumar, 2017, JASMIN: A prototype high resolution soil moisture analysis system for Australia. Bureau Research Report No. 026. http://www.bom.gov.au/research/publications/researchreports/BRR-026.pdf.

### Observations of supercooled liquid clouds over the S. Ocean: From Processes to Cloud Feedback

Andrew Gettelman
NCAR
Friday, 2nd Feb 2018
10:30 AM - 11:30 AM

Abstract: Why does a global modeler go to Hobart to look at clouds? The Southern Ocean Clouds Radiation Aerosol Transport Experimental Study (SOCRATES) is measuring from Hobart to the south with a combination of aircraft, ship and station data from Hobart in January and February 2018. One of the major goals of SOCRATES is to understand S. Ocean supercooled liquid and mixed phase clouds. These clouds are important for climate, and may also change as the structure of the atmosphere over the S. Ocean evolves due to climate change. Global models have traditionally struggled to represent these clouds. This talk with show some preliminary SOCRATES data from flights over the S. Ocean, and show how we will link it to global model simulations of cloud microphysical processes, mean climate, and even cloud feedbacks over the S. Ocean. Preliminary implications for how we can modify global model cloud microphysics will be discussed.

### Mechanisms of Pacific decadal variability: past and future

Emanuele Di Lorenzo
Georgia Tech.
Thursday, 1st Feb 2018
10:00 AM - 11:00 AM

Abstract: The climate variability of the Pacific basin is often characterized in terms of statistical modes of seas surface temperature (SST), sea level pressure (SLP) and sea surface height (SSH) variability. These modes include the El Niño Southern Oscillation (ENSO), the Pacific Decadal Oscillation (PDO), the North Pacific Gyre Oscillation (NPGO), the South Pacific Decadal Oscillation (SPDO) and the Inter-decadal Pacific Oscillation (IPO). In the low-frequency limit (e.g. timescales > 8 years), the Pacific modes tend to converge into a common spatial pattern of ENSO-like decadal variability. Using observations and modeling data we show that the dynamics of growth, peak and decay of the pattern of Pacific decadal variance can be understood by examining the interaction between two dominant set of coupled ocean/atmosphere “dynamical” modes that act along the zonal (e.g. equator) and meridional planes (e.g. extra-tropics to tropics). The Meridional Modes (MM) are energized by stochastic atmospheric variability in the extra-tropics in both the northern and southern hemisphere, their growth is associated with the wind-evaporation-SST (WES) thermodynamic feedback, which allows them to propagate towards the equator. In the tropical Pacific, the MM SST anomalies are amplified further by dynamic ocean/atmosphere feedbacks along the equatorial plane with the growth of the Zonal Modes (ZM), which ultimately shape the different expressions of the ENSO pattern and its global teleconnections. Using this simple mechanistic interaction of ZM and MM between tropics and extra-tropics, we develop a unified framework for understanding decadal variability in the tropical Pacific and its connection to both the northern and southern hemisphere climate modes (e.g. PDO, NPGO, SPDO, etc.). This framework provides a mechanistic basis for understanding past climate variability and for exploring how anthropogenic forcing may amplify Pacific decadal variability and impact marine ecosystems.

### MONSOONS AND TROPICAL RAINFALL IN MODELS AND OBSERVATIONS

Gill Martin
UK Met Office
Tuesday, 30th Jan 2018
10:00 AM - 11:00 AM

Abstract: Despite considerable efforts worldwide to improve model simulations of monsoon systems, significant biases still remain in climatological seasonal mean rainfall distribution and circulation strength, timing of onset, and inter-annual and intra-seasonal variations. Generally, models often fail to reproduce the observed spatial and temporal distributions of tropical precipitation. The need for improved understanding of how a warming climate may change precipitation variability and extremes has focused model developers' attention on the inability of convection parameterizations to represent the observed range of deep convective processes. Under particular scrutiny are the consequences of poorly simulated sub-daily, grid-point precipitation variability on rainfall distributions at longer (e.g., daily, seasonal, decadal) timescales and larger spatial scales.

The Met Office benefits from a unified modelling strategy where essentially the same modelling system is used across timescales ranging from days to centuries. This allows errors in the monsoon simulation to be examined as they develop, and the relationship between problems with short-timescale variability and longer-term systematic biases to be established. Various areas of study are being pursued at the Met Office, in conjunction with our many collaborators in the UK and worldwide. These include analysis of Asian and Australian monsoons in observations and in simulations with various global configurations and resolutions of the MetUM and various other CMIP5 models, the representation of synoptic-scale systems such as monsoon depressions, and the temporal and spatial intermittency of sub-daily precipitation. We are also exploring the use of a variety of complementary diagnostic techniques applied to the problem of understanding sources of model systematic error in monsoon simulations. These include (i) a nudging/relaxation technique to explore remote forcing/teleconnections on monsoon circulations, (ii) the use of initial tendencies from short-range NWP simulations, including a PV tracer technique, to explore the local interactions between parametrised heating/momentum forcing and monsoon circulations, (iii) studies of the role of model resolution and coupled atmosphere-ocean-land interactions at weather and climate timescales, and (iv) use of a perturbed parameter ensemble (PPE) to explore the role of parametrization uncertainties. Examples of some of this research are given in this talk.

### Atmospheric transport modeling activities at Norwegian Institute for Aire Research

Ignacio Pisso
NILU
Wednesday, 13th Dec 2017
10:00 AM - 11:00 AM

Abstract: In addition to longstanding monitoring and composition databases, the Norwegian Institute for Air Research (NILU) hosts a section of research in atmospheric transport modelling. I will present an overview of recent developments of the Lagrangian particle dispersion model FLEXPART and its applications including global and local scales. Of particular interest are the applications of particle dispersion in backward mode to emission fluxes assessment using composition measurements (e.g. for greenhouse gases). These activities are also supported by the Nordic Center of Excellence on eScience for Climate Change (eSTICC) that NILU coordinates. eSTICC also includes activities in data management, high performance computing and model parametrizations that will be briefly presented.

### Atmospheric Circulation Reconstructions over the Earth (ACRE) – the first decade

Rob Allan
UK Met Office
Tuesday, 12th Dec 2017
11:00 AM - 12:00 NOON

Abstract: This presentation will be an overview of two interlinked international climate initiatives that I lead; the International Atmospheric Circulation Reconstructions over the Earth (ACRE) Initiative (http://www.met-acre.net/) and the EU Copernicus Climate Change Service (C3S) Data Rescue Service (DRS). ACRE undertakes and facilitates historical global surface terrestrial and marine weather data recovery, imaging and digitisation, feeding these data into the international repositories responsible for such material, seeing that they provide the best quality and quantity of surface weather observations for assimilation into all dynamical global 4D weather reconstructions or reanalyses, and ensuring that reanalyses outputs are freely available and feed seamlessly into the climate science, climate applications and services, impacts, risks and extremes communities. Under ACRE’s broad international focus, it has worked to develop various regional data rescue foci in Chile, the Pacific, China, Canada, Meso-America, the Indian Ocean (with the World Meteorological Organisation [WMO] /ACRE/Global Framework for Climate Services [GFCS] INdian Ocean DAta REscue (INDARE) initiative, SE Asia, South Africa, the Arctic, Argentina, Australia and Antarctica, all of which are at various stages of development (http://www.met-acre.net/chapters.htm).

The ACRE initiative is also expanding its strong citizen science program linked to Old Weather (https://www.oldweather.org/) and Weather Detective (http://www.weatherdetective.net.au/), plus developing an integrated cross-disciplinary focus (climate science melding with social sciences and humanities) on historical reanalyses and weather reconstructions, ensuring that the global historical weather observations and reanalyses outputs are analysed and assessed in a longer historical context, incorporating climate knowledge based on historical narrative accounts of the impacts and responses associated with past climate variability and extreme weather events, and thus adding higher-value to existing climate services approaches (http://onlinelibrary.wiley.com/doi/10.1002/wcc.379/pdf).

The new 4-year EU Copernicus-funded C3S DRS brings together the Met Office and 13 other subcontracted partners to provide international leadership in the fields of terrestrial and marine data rescue. Building upon existing WMO and international data rescue activities and standards and interlinked closely with ACRE, the Service will construct a managed, integrated, state-of-the-art repository (portal and registry) of information about past, current and planned data rescue projects, plus visualisations of outputs and products, whilst piloting the use of new tools, techniques and approaches in data digitisation. The C3S DRS has funding support for data rescue activities and the testing of new data rescue tools and procedures in three new/evolving ACRE regional data rescue regions in the Southern Hemisphere centring on Argentina, South Africa and the higher latitude Pacific sector of the Antarctic continent, which interlink with the new ACRE Australia regional focus.

### A climatology of Australian elevated thunderstorms derived from sounding and reanalysis data

Ross Bunn
Bureau of Meteorology
Thursday, 7th Dec 2017
3:00 PM - 4:00 PM

Abstract: Thunderstorms that source air from above a stable near-surface boundary layer divorced from the effects of diabatic heating are known as elevated thunderstorms (hereafter referred to as ES), as opposed to surface-based thunderstorms (SS). These storms can produce heavy rain and snow (Moore et. al. 2003, 1998), large hail and damaging winds (Goss et al., 2006), as well as lightning which can affect aerodromes and cause bushfires. However, forecasting the development and cessation of these storms is a challenge for Australian forecasters where sparse upper-air observational networks hinder effective monitoring of the pre-storm environment. Whilst the majority of U.S. ES occur overnight and in winter (Colman 1990a), no Australian climatologies or case studies of ES are available in the literature and the seasonal and diurnal occurrence of Australian ES remain unknown.

An effective inflow and cloud-to-ground lightning based discriminator was developed and applied to ERA Interim reanalyses to classify thunderstorms as either ES or SS. A seven year Australian ES and SS climatology has been developed in this work, showing for the first time, seasonal and diurnal characteristics in ES and SS occurrence. ES are primarily an austral spring and summer phenomena, occurring most often overnight, in the Great Australian Bight and Southern Ocean, as well as affect- ing southern Western Australia, South Australia and Victoria. ES exhibit seasonal meridional shifts which may be related to meridional shifts in anticyclone and upper jet seasonal occurrence. Thunderstorms in subtropical and tropical regions are exclusively SS, owing in part to the inability of deep surface stable layers to penetrate into those regions.

Wolfgang Fisterle
PMOD, Davos
Wednesday, 6th Dec 2017
3:00 PM - 4:00 PM

Abstract: An overview of the metrological challenges in the field of solar radiometry will be given and how PMOD/WRC tackles these challenges. Our latest developments include radiometers for space and ground-based applications, at ambient and cryogenic temperatures. Ongoing research projects aim at improving the optical and thermal properties of the sensor, and to establish full traceability of solar irradiance to SI base units. I will present the latest results from our research and give an outlook on the future of the PMO6-cc series of absolute cavity radiometers.​

### Jive - Interactive Verification Environment

Ioanna Ioannou
Bureau of Meteorology
Wednesday, 6th Dec 2017
10:00 AM - 11:00 AM

Abstract: Jive provides an easy way to extract, explore, and verify forecast and observation data. Users can interact with Jive through Jupyter notebooks in the browser, or use it as a Python library in their scripts. Jive is designed to be flexible, usable, automatable, and trustworthy. The project was developed to facilitate the large scale verification needed to make decisions around forecast automation, but is equally adept at performing case-study scale analysis. The Forecast Systems Team provides web-browser access to Jive for anyone in the Bureau, with forecasts and guidance from the Graphical Forecast Editor (GFE), including our official public weather forecasts, and curated observations from Automatic Weather Stations. Jive is flexible enough that its rich library of verification metrics, statistical and stratification functions can also be applied to external data.

### CAM Ensembles: Post-Processing, Verification, and Design

Israel Jirak
NOAA
Tuesday, 5th Dec 2017
10:00 AM - 11:00 AM

Abstract: The limited predictability on the convective scale has prompted the development and testing of CAM ensemble systems to quantify uncertainty and provide a range of possible convective storm solutions. In extracting useful information from CAMs, specialized hourly maximum storm-attribute fields are produced to diagnose simulated storm intensity and potential severity (e.g., updraft helicity). In addition, specialized post-processing techniques are applied to the storm-attribute fields to summarize information from CAM ensembles, including the use of temporal and spatial neighborhood approaches. These neighborhood approaches provide meaningful probabilistic information when ensemble members have similar, but not identical, placement and timing of storms, which is also useful in the verification process. Furthermore, performance characteristics of experimental CAM ensembles have been documented for several years in the annual NOAA Hazardous Weather Testbed (HWT) Spring Forecasting Experiments (SFEs). While major advances have been made in processing, visualizing, and verifying these large and complex datasets, progress toward identifying optimal CAM ensemble configurations has been inhibited because the different systems have been independently designed by the diverse collaborators. Beginning in 2016, a much more coordinated effort was established for the SFE with regard to CAM ensemble design. This was achieved by working with collaborators on a common set of model specifications (e.g., model version, grid-spacing, domain size, physics, etc.) so that the simulations contributed by each collaborator could be combined to form one large, carefully designed ensemble known as the Community Leveraged Unified Ensemble (CLUE). Some of the CLUE findings and general themes regarding CAM ensemble design that have been observed over the past several years in the HWT SFE will be discussed.

### Guidance post-processing, impacts and verification at the Met Office

Ken Mylne
UK Met Office
Monday, 4th Dec 2017
10:00 AM - 11:00 AM

Abstract: Dr Mylne will present an overview of Met Office work and plans in these areas, followed by discussion.​ The Met Office 'IMPROVER' systems aims to provide seamless calibrated probabilistic guidance from convection-permitting and coarser-scale numerical weather prediction ensembles, from nowcasting to medium-range timescales and beyond. Verification is applied at each stage of the processing. New maintainable and extensible platforms are being built to support the handling of massive data volumes and product generation. The Bureau and the Met Office are considering collaboration in these areas.

Biography: Ken Mylne is the Head of Post-processing, Verification and Impacts at the Met Office, and is also a prominent contributor to WMO ensemble strategies.

### Investigating SST diurnal variation over the tropical warm pool region

Haifeng Zhang
University of Melbourne
Tuesday, 21st Nov 2017
10:00 AM - 11:00 AM

Abstract: Sea surface temperature (SST) diurnal variation (DV) signals over the tropical warm pool region (TWP, defined as 90°E-170°E, 25°S-15°N) are investigated in this study. First, the general DV features and DV cycles are investigated using the v3 (version 3) geostationary MTSAT-1R (Multi-functional Transport Satellites-1R) satellite SST data set. This data set was produced by the Australian Bureau of Meteorology (the Bureau) over the TWP spatial domain and for four months from January to April 2010. Together with surface wind speed and solar shortwave insolation (SSI) outputs from the Australian Community Climate and Earth-System Simulator-Regional (ACCESS-R) numerical prediction model, statistical analyses of DV signals are conducted. Good correlation is found between DV events and low wind and high SSI conditions. The dominant role of wind speed in SST DV events over SSI is also revealed. Second, we investigate the seasonal patterns of DV using the Bureau's five-year (2010-2014), v2 AVHRR (Advanced Very High Resolution Radiometer) SST data. A double-peak seasonal pattern of SST DV is observed: the strongest DVs are found in February/March and October/November, and the weakest in June/July. The more important role of the late morning and early afternoon (07-14 local time) winds in determining the DV amplitude is also shown. Then we look into the physical phenomena behind the double-peak/trough DV/wind seasonal pattern over the Australian North-Western Shelf (NWS, defined as 105°E-125°E, 25°S-10°S). It is found that this seasonal DV/wind pattern over the NWS is largely due to the westerly Australian summer monsoon, which reduces the easterly trade winds during the monsoonal period. As a result, wind speed over the region reaches the minimum at the beginning and end of the monsoonal period with the weakest wind mixing, favouring the SST DV development. Finally, a group of DV models, including one empirical model, two physical models, and one air-sea coupled model with DV schemes implemented, are estimated against v3 MTSAT-1R data set and inter-compared with each other. The sensitivity of v3 MTSAT-1R data to true SST changes is tested beforehand. Results indicate that all models are able to capture the general DV patterns but with differing accuracies and features.

### Avoided heat-related mortality under targeted Paris Agreement scenarios

Dann Mitchell
University of Oxford
Wednesday, 15th Nov 2017
10:00 AM - 11:00 AM

Abstract: The impact of extreme heat events on human health is amongst the most concerning consequences of climate change. Even small changes in globally averaged temperature can lead to amplified extremes and localised impacts on health. Recently, under the Paris Agreement on Climate Change, there has been a call for research into impacts associated with a 1.5C or 2C globally-averaged surface temperature anomaly. Here, for the first time, we use climate model experiments designed specifically for the Paris Agreement to assess the human health impacts associated with extreme heat. Our analysis shows that in high-population regions, e.g. Central Africa, India and Europe, an additional 10-20 days of heat events can occur on average every year. Modeling the most extreme historical heat-mortality event on record as if it occurred under future climate scenarios shows that for key European cities, stabilising climate at 1.5C would decrease temperature-related mortality by 15-25% per summer compared with stabilisation at 2C, assuming no adaptation and constant vulnerability. Given the robustness of our results, we maintain that for heat-related mortality there is a scientific argument for stabilising climate at 1.5C over 2C.

### Key Processes in Modeling the Madden-Julian Oscillation

Xianan Jiang
University of California
Thursday, 9th Nov 2017
2:00 PM - 3:00 PM

Abstract: Despite its significant impacts on global weather extremes, representation of the Madden-Julian oscillation (MJO) remains a grand challenge for present-day climate models. In this presentation, based on a moisture-mode framework for the MJO, key model processes in regulating MJO amplitude and propagation are presented by analyzing multi-model simulations from the recent MJO Task Force / GEWEX GASS MJO model comparison project. The efficiency of precipitation generation per unit column water vapor, or the convective time-scale, is found to be closely linked to model MJO amplitude. On the other hand, moistening and drying due to horizontal advection of lower-tropospheric seasonal mean moisture are critical in driving the eastward propagation of the MJO. Faithful model simulation of the low-level (900-650hPa) mean moisture pattern over the Indo-Pacific region, particularly near the Maritime Continent area, is found to be essential for realistic representation of the MJO eastward propagation in climate models as confirmed by multi-model simulations. Specific analyses related to model biases in simulating the MJO in ACCESS model will also be discussed in the context of these findings.

### Event Attribution: Past, Present and Future

Fraser Lott
UK Met Office
Thursday, 9th Nov 2017
12:00 NOON - 1:00 PM

Abstract: Attribution of extreme events has come a long way in a short time. At the turn of the 21st century we still believed that climate change couldn't be seen in individual events. Now, large collaborations like Eucleia and its successor, Eupheme, are bringing us ever closer to attribution on an operational basis. What ideas do we have for the next steps in event attribution? Dr Fraser Lott did a Plasma Physics PhD at Imperial College London and a postdoc at CEA Cadarache in France, both applying infrared imaging to potential fusion reactor exhausts. He then moved to the Met Office and briefly worked on developing an infrared satellite-based climate dataset before moving into detection and attribution of climate change. He has looked at optimal detection of radiosonde temperature and dry spell attribution in East Africa, then spent a few years examining reliability in attribution, with a quick break to examine it in solar flares instead! Having exhausted his interest in that, I'm now considering where land use fits in with the attribution problem.

### Understanding of boundary-layer processes and improving their representation in models using Large-Eddy Simulations.

Fleur Couvreux
Meteo-France, CNRM
Wednesday, 1st Nov 2017
10:00 AM - 11:00 AM

Abstract: In this presentation, I will present Large-Eddy simulations of shallow clouds or clear boundary layer where most of the processes are resolved. After a brief description of such simulation and their main advantages and drawbacks, I will describe in more detail how they can be used both for the understanding of boundary-layer processes and for the improvement of their representation in models. I will in particular detail a simulation of initiation of deep convection observed during the AMMA field campaign that occurs over West Africa. With such a simulation, we show the different processes leading to initiation of deep convection in this very dry environment not favourable to convection. I will also show how these simulations can be used to derive parametrization-oriented diagnostics that have been used to improve boundary-layer parametrizations.

### ACCESS City Ensemble (ACCESS-CE3): Status, First Results and Plans

Shaun Cooper
Bureau of Meteorology
Wednesday, 25th Oct 2017
10:00 AM - 11:00 AM

Abstract: Many Meteorological agencies are running their own Numerical Weather Prediction (NWP) ensembles. NWP ensemble seek to sample the possible future state of the atmosphere by acknowledging the sources of uncertainty in weather forecasting and attempting to account for these. High-resolution NWP ensembles are able to provide information on high impact weather events, and their uncertainties, on the timescales of days. ACCESS City Ensemble (ACCESS-CE3 or CE3) is currently under active development and significant progress has been made with this system. This system leverages information from the Global Ensemble (ACCESS-GE3) and the City deterministic system (ACCESS-C3) to generate a 2.2 km, 12 member ensemble over the 6 ACCESS City domains. ACCESS-CE3 is a convection permitting model based on the UK Met Office's high-resolution ensemble, MOGREPS-UK. This presentation is a status report on the system that will discuss how the ensemble is created and present a first look at some of the initial cases that have been run, including the East Coast Low from June 18-19, 2016. Comparisons of radar observations and forecast rainfall will be presented along with some products that can be generated from the ensemble output. A brief summary of the work still required prior to running trials and readying the system for operations will also be discussed.

### Almost everything you ever wanted to know about ensemble verification metrics

Debbie Hudson
Bureau of Meteorology
Wednesday, 20th Sep 2017
10:00 AM - 11:00 AM

Abstract: What characterises a "good" ensemble forecast? No single metric can adequately describe forecast performance. There are a number of key attributes of forecast quality of ensembles, including reliability, resolution, discrimination, sharpness, accuracy and skill - all of which can be represented by various graphs and scoring metrics. The choice of verification metrics will depend on how the ensemble is interpreted. The most common interpretation and application of ensembles is as probabilistic forecasts derived from the ensemble distribution (e.g., probabilities above a threshold) and there are a number of standard probabilistic skill metrics that are appropriate. However, the ensemble can also be viewed as a set of deterministic forecasts making up the forecast distribution and specific metrics exist that are designed to evaluate the properties of the full distribution. The chosen metric will also be more or less appropriate depending on the purpose of the ensemble verification, whether it be for end-users, comparing forecast systems or trying to diagnose errors.

This presentation will review the attributes of forecast quality and the typical diagnostic plots and scoring metrics used, such as attributes diagrams, the relative operating characteristic (ROC), the Brier score and skill score (BSS), the ranked probability score (RPS), the continuous ranked probability score (CRPS), rank histograms and spread-skill graphs. New approaches and the key challenges facing the verification of ensembles will also be discussed. Examples will be taken from research across a range of forecast timescales, spanning NWP to seasonal.

### Sea Surface Temperature Forecasts for evaluating Coral Bleaching Risk in the Great Barrier Reef

Grant Smith
Bureau of Meteorology
Wednesday, 13th Sep 2017
10:00 AM - 11:00 AM

Abstract: Great Barrier Reef (GBR) Marine Park managers rely on sea surface temperature (SST) seasonal forecasts to better inform their management of coral bleaching events. The new global coupled ocean-atmosphere seasonal prediction system ACCESS-S (with land surface and sea ice components) has higher spatial resolution (approximately 25 km ocean grid in the Australasian region) than the previous Australian seasonal model POAMA2 (approximately 100 km to 200 km ocean grid) and will be run more frequently (daily vs every three days). This model will provide greater detail within the Great Barrier Reef Marine Park, with the benefit of having individual forecasts for different management zones. The accuracy of the SST forecasts depends on the location, the forecast lead time, and the time of year the forecast is initialised. The skill of the SST forecasts were assessed against satellite observations for GBR region for the hindcast period 1990 to 2012. Operational SST anomaly forecasts will be available as both spatial maps and regional indices for the GBR, developed in consultation with Great Barrier Reef Marine Park stakeholders.

### The joy of investigating Australia's weather and climate: thirty five years research in the Bureau of Meteorology

John McBride
Bureau of Meteorology
(Retired)
Wednesday, 6th Sep 2017
12:30 PM - 1:30 PM

Abstract: This is a talk about the process of doing research on weather and climate. Though I have been retired for some years, occasionally I am invited to carry out scientific visits. During a visit to the Chinese Academy of Meteorological Sciences earlier this year, I was asked to give a talk aimed at young scientists. I was asked to draw on examples from my research, show what I learned that was new or exciting, and how I went about it. Examples include: Research on Tropical Cyclogenesis (McBride and Zehr, 1981), The Organization of Tropical Convection (McBride, Tellus, 1983), The Effect of ENSO on Australian Rainfall (McBride-Nicholls, 1983), Predictability on Seasonal Timescales (Haylock and McBride, 2001), A New Approach to Verification (Ebert and McBride, 2000), Tropical Cyclones and Climate Change (Knutson, McBride, et al, 2010), Response of Sea Surface Temperature to Tropical Cyclones (Dare and McBride, 2011), and my series of monthly talks on the major weather event of the previous month (NMOC-NCC chart discussion, 2008-2011). I was also asked to end the seminar with some dot points of advice to young scientists. This is that talk.

### How novel compute technology transforms life science research - From Hadoop/Spark to server-less

Denis Bauer
CSIRO
Monday, 28th Aug 2017
3:30 PM - 4:30 PM

Abstract: Synopsis: The talk showcases how genomic research has leapfrogged to the forefront of BigData and Cloud solutions. We outline how Apache Spark helps identify genomic association on population-scale whole genome sequencing data, as well as how the accuracy of genome editing approaches can be improved with massively parallel server-less cloud functions.

Biography: Dr. Denis Bauer is the team leader of the transformational bioinformatics team in CSIRO's ehealth program. She has a PhD in Bioinformatics from the University of Queensland and held Post-doctoral appointments in biological machine learning at the Institute for Molecular Bioscience and in genetics at the Queensland Brain institute. Her expertise is in computational genome engineering and BigData compute systems. She is involved in national and international initiatives tasked to include genomic information into medical practice, funded with $200M. She has 30 peer-reviewed publications (14 as first or senior author) with six in journals of IF>8 (e.g. Nat Genet.) and H-index 11. To date she has attracted more than$6.5Million in funding as Chief investigator.

### So hot right now: why the mad scramble for our climate data from the financial and corporate sectors?

Sarah Barker
Minter Ellison
Thursday, 24th Aug 2017
2:00 PM - 3:00 PM

Abstract: Sarah will describe recent developments in financial markets and corporate reporting frameworks, shareholder activism and litigation exposures, and consider how these translate into significant opportunities for climate science and data. Sarah is recognised as one of the world's foremost authorities on the issue of climate change risk in investment and corporate governance, with her work in this area sought by institutions from the Bank of England and United Nations PRI to the OECD and European Union. In addition to her role at Minter Ellison, Sarah sits as a director of the Responsible Investment Association of Australasia and Emergency Services and State Super, and the Technical Working Group of the Climate Disclosure Standards Board. She is also part of the faculty at the Cambridge Institute of Sustainability Leadership, and is a member of the Sustainable Finance Programme at the Smith School of Enterprise and the Environment at the University of Oxford.

### Rejuvenating IDL Applications with New Technologies

Weidong Jiang
Harris Geospatial Solutions
Wednesday, 9th Aug 2017
10:00 AM - 11:00 AM

Abstract: IDL and its applications have been playing important roles in many meteorological organizations. For example, EUMETSAT Polar System (EPS), Air Force Weather Agency Satellite Imagery Display and Analysis System (SIDAS), UK Met Office 4D Weather Cube, German Federal Offices for Water Management MeteroViewer, DLR Institute of Atmospheric Physics WxFusion, and Australian Bureau of Meteorology Tsunami Decision Support Tool (DST). As new technologies such as cloud computing, thin client and Software as a Service (SaaS) come into mainstream, some might think IDL-based applications are outdated and should be phased out when considering system upgrades and/or new procurement. We will talk about such concerns and how we can embrace the challenges with our new enterprise offering: IDL Service Engine (ISE), ENVI Service Engine (ESE), and Geospatial Service Framework (GSF).

### Lidar Technology and its applications for Environmental Pollution in China

Zhenyi Chen
CAS
Wednesday, 2nd Aug 2017
10:00 AM - 11:00 AM

Abstract: China is now faced with some of the world's most severe and complex environmental problems due to the rapid economic development in recent years. Therefore it is significant to promote research on the atmosphere. Lidar is one of the backbones of the research field that deals with the profiling of the atmosphere. It provides high spatial and temporal resolution of the measurements, the possibility of observing the atmosphere at ambient conditions, and the potential of covering the height range from the ground to more than 100 km altitude. These and some more aspects of lidar are presented in this lecture giving an overview on the state of the art of the basic lidar techniques (especially Mie lidar, Raman lidar and Ozone lidar) used in the investigation of the Earth's atmosphere. Furthermore several lidar observing cases are discussed to illustrate the aerosol/ozone profiles, boundary layer evolution and evaluate aerosol impacts on climate changes. Finally some recent publications are presented and the prospects discussed.

### A new quantitative smoke forecasting system for Victoria

Martin Cope
CSIRO
Wednesday, 26th Jul 2017
10:00 AM - 11:00 AM

Abstract: Following the devastation of Victorian communities in February 2009 by the Black Saturday fires, a Royal Commission into the tragedy concluded that heavy fuel loads present in fire risk zones were an important contributing factor, and recommended substantial increases in preventative fuel reduction burning of state-owned land. Although modified to a risk based approach in 2016, fuel management through prescribed burning remains the predominant tool in Victoria for mitigating bush fire risk.

Prescribed burning in Victoria is primarily undertaken in autumn and early winter when atmospheric conditions are stable- thus light winds, moderate temperatures and humidity, preferably extending over several days. Unfortunately, such conditions are also conducive to the build-up of air pollution, from the ambient fires, and also from domestic wood combustion and other anthropogenic sources. Because there is a strong mandate to reduce fuel loads, and generally only a narrow window of opportunity each year to undertake prescribed burning, land managers are reliant on smoke transport forecasting tools to help manage the air pollution exposure risk associated with a program of burns.

AQFx is a multi-tiered smoke forecasting framework which was developed for the Department of Environment, Land, Water and Planning by CSIRO, the Bureau of Meteorology and four university partners. In this presentation we summarise key aspects of the AQFx development including the field work and analysis undertaken to better characterise the fuel layers and combustion characteristics of Victorian forest fuels; the three-tiered approach used to provide progressive layers of intelligence over a 10-day outlook; a demonstration of AQFx operation; and some considerations for the on-going development and application of the system as it nears operational status.

### Expanding the remote sensing repertoire of HF radar systems

Stuart Anderson
Monday, 17th Jul 2017
2:00 PM - 3:00 PM

Abstract: We are presently witnessing a resurgence of interest and investment in over-the-horizon radars operating in the HF band, unlike any such development since the 1970's. Of the order of 500 such radars have been deployed around the world, including about 150 now operational in Asia. While the great majority of these systems are low-power commercial radars, designed originally for current mapping and wave measurement, several countries have chosen to develop new, highly capable military-grade HF radars and integrate them into national (and trans-national) netted systems for airspace and ocean surveillance. At the same time, emerging scientific and technological developments in physics, mathematics, materials science, signal processing and artificial intelligence are all contributing to an expanded range of remote sensing capabilities, including some reliant on unorthodox radar configurations which were hitherto not realisable. In this talk I shall discuss some of the latest advances in HF radar oceanography and meteorology, focussing on those which have direct relevance to Australia in general and to the responsibilities of the Bureau of Meteorology in particular.

### An attribution study of southeast Australian wildfire risk

Mitchell Black
Bureau of Meteorology
Wednesday, 12 Jul 2017
10:00 AM - 11:00 AM

Abstract: In the aftermath of the recent bushfires in Australia the scientific community was faced with the challenge of quantifying the event's link to different causal factors, including human-induced climate change. While there are a number of recorded attribution studies for temperature and precipitation-related events, no such study exists for fire weather. This study investigates how the likelihood of extreme fire weather across southeast Australia has been changed due to the competing influences of human-induced climate change and modes of inter-annual climate variability. The analysis benefits from the use of the recently launched weather@home Australia-New Zealand distributed computing citizen science project to generate very large ensembles of regional climate model simulations over Australia. The likelihood of extreme fire weather is examined for different phases of the El Niño Southern Oscillation under present climate conditions and climate conditions with no human influences.

### Tropical Expansion in the Southern Hemisphere

Daren Waugh
UNSW
Wednesday, 28th Jun 2017
1:00 PM - 2:00 PM

Abstract: Observational evidence indicates that the tropics are expanding, but a wide range of expansion rates have been reported, global climate models underestimate the observed trend, and there is no consensus on the cause of this expansion. Here we examine trends in southern edge of the tropics during austral summer. It is shown that the weight of the evidence clearly points to stratospheric ozone depletion as the dominant driver of the tropical summertime expansion over the period in which an ozone hole was formed (1979 to late 1990s), although sea surface temperature and internal atmospheric variability have also contributed (especially since late 1990s). The large role of internal atmospheric variability may explain the underestimate in the ensemble mean trend from climate models. We also show that the wide range of expansion rates is due, in part, to the great variety of metrics used to define the tropical width, with many measuring different aspects of the tropical circulation.

### Challenges and Opportunities of continental scale modelling: Experiences in modelling the European subcontinent for water resources, water quality, forecasting and projections.

Chantal Donnelly
Bureau of Meteorology
Wednesday, 28th Jun 2017
10:00 AM - 11:00 AM

Abstract: The Hydrological Predictions for the Environment, HYPE model (Lindstrom et al. 2010), was introduced in 2007 as a semi-conceptual/process based distributed hydrological and nutrient model to address the needs of the European water framework directive in Sweden. The model was initially applied to simulate flow and nutrients across all of Sweden but quickly proved useful at continental scales. It is currently applied for Europe (E-HYPE), the Arctic (Arctic-HYPE), the Indian subcontinent (India-HYPE), for Sweden (S-HYPE) and the La Plata and Niger rivers, in many cases using open data sets. In 2017, the first GLOBAL-HYPE runs were made. Working with the same model code in diverse regions around the world has led to insights hydrologic functioning which have fed back into the model setups, code development and calibration. The E-HYPE model lead much of this development in international, continental scale modelling. This presentation will showcase some of the research and products resulting from the 8 years since E-HYPE was conceived including:

• Key factors for improving large-scale hydrological model performance
• Calibration and validation of continental scale hydrological models
• The impacts of 1.5, 2 and 3C global warming on European hydrology
• Overcoming challenges of continental scale forecasting in an international context
• Riverine and nutrient fluxes to the Baltic Sea – scenarios for nutrient management and governance in a changing world

Biography:Chantal Donnelly recently started as the unit manager for the Water Resources Modelling Unit in the Water Division. She previously spent 8 years at the Swedish Meteorological and Hydrological Institute leading the development of the E-HYPE pan-European hydrological model and developing methods and services for future climate impact assessments on hydrology. She did a PhD on Coastal Hydraulics and Sediment transport at Lund University in Sweden and a Bachelor of Engineering (Civil) at University of Queensland.

### Using observations to improve airborne volcanic ash dispersion forecasts

Linden Ashcroft
Bureau of Meteorology
Wednesday, 21st Jun 2017
10:00 AM - 11:00 AM

Abstract: The current high-quality datasets of temperature and rainfall available for Australia generally only begin at the start of the 20th century. This limits our ability to understand the full range of Australia's natural climate variability. Historical instrumental weather observations taken before the formation of the Bureau of Meteorology in 1908 have the potential to improve our understanding of climate variability in Australia. Many 19th century weather records exist in libraries and archives across the country, particularly in the highly populated southeast region, but have not been assessed for their scientific potential. Using these weather observations taken by convicts, farmers, astronomers and explorers, I will explore climate variability in southeastern Australia back to European settlement in 1788. I will discuss the benefits and limitations of 150-year-old data, and describe techniques used to quality control and homogenise the rescued historical record. With this extended climate record, I will identify some relatively unknown features of southeastern Australia's pre-1910 climate. I will then examine changes in the relationships between southeastern Australian rainfall and large-scale circulation features (such as El Niño-Southern Oscillation) over the past 220 years.

### Using observations to improve airborne volcanic ash dispersion forecasts

Meelis Zidikheri
Bureau of Meteorology
Wednesday, 14th Jun 2017
10:00 AM - 11:00 AM

Abstract: The Bureau of Meteorology monitors and issues forecasts of volcanic ash during eruption events over a region that includes Australia, Indonesia, Papua New Guinea, and the southern Philippines. This information is mainly used by the aviation industry for operational risk management purposes. Satellite based remote sensing techniques are the primary means of identifying the locations of ash during eruption events. Forecasts of ash locations are obtained from dispersion models which employ prognostic meteorological fields such as wind speed and direction to predict the future locations of ash. In current operational practice, the dispersion models are initialized by a columnar source of ash, with uniform mass distribution, and with a maximum height determined by best estimates of the ash cloud height, often relying on cloud infrared brightness temperature and trajectory analysis of model wind fields at different heights.

Both the remote sensing algorithms and the dispersion model contain significant uncertainties. Remote sensing of ash in tropical regions is especially challenging due to the difficulty of demarcating ash from the ice and water in convective clouds which results in frequent missed or false detections of ash. Methods for retrieving ash properties such as ash mass load have recently been developed which have the potential of enhancing the information contained in forecasts but these are not immune to the general problems facing remote sensing of ash in tropical regions. Dispersion models also contain uncertainties. Some of these uncertainties arise from uncertainties in the meteorological fields. Other uncertainties arise from uncertainties in the ash source characteristics such the height of the ash column, the distribution of mass within the ash column, and the particle size distribution.

In this presentation we focus on improvements to the both the representation of meteorological fields and ash source by making better use of available observations. The method presented herein is based on an exhaustive sampling of parameter space for the optimal source parameter values, with pattern correlations between modeled and observed ash fields as the utility function. The method is applicable to problems where only a binary valued ash detection field is available - which is frequently the case - and to problems where a retrieved ash mass load field is available. We further show how the method can be used in intermediate problems where only qualitative estimates of the ash mass load field - in the form of high, moderate and low loading values - are available. Finally, we address the problem of verifying and calibrating ensemble based ash forecasts with both meteorological field and source term variations included, the former being based on the ACCESS-GE ensemble model. Several cases studies are used to demonstrate the efficacy of this approach in improving forecasts of airborne volcanic ash.

### High-resolution ensemble prediction of the east coast low of April 2015

Dragana Rajak
Bureau of Meteorology
Wednesday, 7th Jun 2017
10:00 AM - 11:00 AM

Abstract: During autumn and winter months the eastern coast of Australia is periodically affected by rapidly developing and intense extratropical low-pressure systems that are known as East Coast Lows (ECLs). ECLs bring damaging winds and heavy rainfall with flooding that can last for several days. Due to their rapid development, many forecasting issues arise; such as: predicting which part of the coast will be impacted, and the intensity and location of maximum winds and rainfall. The use of ensembles can help overcome these forecast challenges and provide insight into how these systems form.

The event that is studied here occurred during 20-23 April 2015, with the worst impact on 21 April. It was a devastating event for Dungog and Maitland area, with at least four deaths reported and widespread damage. This event was simulated using a 24-member ensemble of ACCESS nested models (global, 4.0 km and 1.3 km). The smallest grid spacing (1.3 km) sufficed to capture the dynamics of the event. Simulations show that forecast rainfall (averaged over the 24 ensemble members) is in good agreement with observed rainfall and the ensemble identifies Dungog as the area at moderate risk of extreme rainfall.

Here, ensemble sensitivity is investigated, to understand how different dynamic features of the flow are related to the predictability of the event. A subset of ensemble members shows very little rain in the Dungog area; instead it is located further south or east, indicating that a large part of the coast is at risk of significant rain. These differences are also discussed here.

### Thermodynamics of Pyrocumulus Formation

Kevin Tory
Bureau of Meteorology
Tuesday, 6th Jun 2017
10:00 AM - 11:00 AM

Abstract: In favourable atmospheric conditions, large hot fires can produce pyrocumulus cloud in the form of deep convective columns resembling conventional thunderstorms (pyroCb), which may be accompanied by strong inflow, dangerous downbursts and lightning strikes. These in turn may enhance fire spread rates and fire intensity, cause sudden changes in fire spread direction, and the lightning may ignite additional fires. Dangerous pyroCb conditions are not well understood and are very difficult to forecast. In this seminar we will illustrate the turbulent nature of fire plumes with the aid of PyroCb videos, schematics of plume dynamics based on laboratory experiments, and large eddy model (LEM) simulations. The turbulence entrains significant amounts of environment air that rapidly dilutes the plume buoyancy. In most fires the plume loses buoyancy before it rises sufficiently for adiabatic cooling to induce condensation and cloud formation. PyroCb formation is likely to be influenced by a range of factors (e.g., fire size and intensity, environment temperature, moisture, wind and stability). In this study we focus on the plume thermodynamics, and in particular how plume dilution from entrainment of environmental air affects the height of condensation. The results help correct some common misconceptions, and provide an additional explanation for why pyroCb often form with the arrival of a cooler wind change (e.g., sea breeze or cold front).

### Secondary eyewall formation in tropical cyclones

Jeffrey D. Kepert
Bureau of Meteorology
Wednesday, 24th May 2017
10:00 AM - 11:00 AM

Abstract: Roughly half of all intense tropical cyclones experience at least one eyewall replacement cycle, in which a new eyewall forms concentrically around the original one. This secondary eyewall develops its own wind maximum, and both the secondary eye and the wind maximum typically intensify and contract, whilst the original eyewall and wind maximum weaken and eventually dissipate. While the evolution of a storm with concentric eyewalls is reasonably well understood, the mechanism or mechanisms by which the outer eyewall forms remain elusive. Understanding secondary eyewall formation is an important problem, for the subsequent eyewall replacement cycle can significantly affect the intensity of the storm, and the formation process and replacement cycle are usually associated with a major expansion of the outer wind field. Both these factors significantly affect the cyclone's impact.

We investigate a high resolution simulation of an eyewall replacement cycle. Boundary layer convergence due to friction substantially influences the evolution of the convection, and we present evidence for a positive feedback involving convection, vorticity and frictional convergence that governs the subsequent evolution of the system. In this feedback, frictional convergence strengthens the convection, stretching of vortex tubes in the buoyant updrafts increases the vorticity, and the vorticity structure of the storm determines the strength and location of the frictional updraft.

Changes in the structure and intensity of tropical cyclones makes it more difficult to manage their impact, especially if these changes occur in the last day or two before landfall. Our improved knowledge of these processes will lead to better forecasts and mitigation.

### Towards an improved land dryness estimate for fire prediction

Vinod Kumar
Bureau of Meteorology
Wednesday, 17th May 2017
10:00 AM - 11:00 AM

Abstract: In operational fire danger ratings issued using McArthur's Forest Fire Danger Index (FFDI), the component representing fuel availability is based on soil moisture deficit. The Keetch-Byram Drought Index (KBDI) and Mount's Soil Dryness Index (SDI) are two models used in operations to calculate the soil moisture deficit. The KBDI and SDI are simplified, empirical water balance models that do not take into account the majority of physical factors which affect soil moisture dynamics such as soil type, vegetation type, terrain or aspect. They over-simplify evapotranspiration and runoff processes, potentially leading to large errors in estimated soil moisture state. In this study, we show that soil moisture from numerical weather prediction models and satellite remote sensing can be more accurate than the above indices.

We discuss the development of a high-resolution prototype system based on Joint UK Land Environment Simulator (JULES) land surface model to estimate soil moisture deficit for fire danger rating. This system, called the JULES based Australian Soil Moisture Information (JASMIN), estimates soil moisture at a spatial resolution of 5 km. JASMIN can predict surface soil moisture, which is closely related to dead fuel moisture content, and root-zone soil moisture that provides information on live fuel moisture content. Verification against ground based soil moisture observations show that this prototype system is significantly more skilful than the traditional models. We also discuss the calibration done on JASMIN soil moisture, which in its native form cannot be used in existing fire warning system.

### The influences on Australian East Coast Lows in present and future climates

Acacia Pepler
Bureau of Meteorology
Friday, 5th May 2017
2:00 PM - 3:00 PM

Abstract: East Coast Lows (ECLs) are midlatitude cyclones that develop on the east coast of Australia, and regularly cause severe weather including major flooding, large waves, strong winds, and coastal erosion. Because of their impacts, there is a need to better understand both how and why they develop, and how they may change in coming decades.

Observational datasets and regional climate model simulations are used to understand the frequency, characteristics, and impacts of East Coast Lows in both the present and future climates. Future projections of East Coast Lows and their impacts will be discussed, as well as how their development is influenced by key features of the east coast region - the warm East Australian Current and the Great Dividing Range.

### Dynamical seasonal forecasting for decision support in marine management

Claire Spillman
Bureau of Meteorology
Wednesday, 3rd May 2017
10:00 AM - 11:00 AM

Abstract: Seasonal forecasting has great scope for use in marine applications, particularly those with a management focus. Seasonal forecasts from dynamical ocean-atmosphere models of high risk conditions in marine ecosystems can be very useful tools for managers, allowing for proactive management responses. The Australian Bureau of Meteorology's seasonal forecast model POAMA currently produces operational real-time global forecasts of sea surface temperatures, with tailored outlooks produced for coral reef, aquaculture and wild fisheries management in Australian waters.

Operational realtime seasonal forecasts for coral bleaching risk on the Great Barrier Reef predict warm conditions that may lead to coral bleaching several months in advance, and play an important role in the Great Barrier Reef Marine Park Authority's Early Warning System. Early warnings of potential bleaching risk can assist reef managers to prepare for the likelihood of an event, focusing resources, briefing stakeholders and increasing awareness of bleaching onset. In marine farming and fishing operations in Australia, seasonal forecasting is being used to reduce uncertainty and manage business risks. Further, habitat distribution forecasts can be generated by combining these environmental forecasts with biological habitat preference data, providing industry with species-specific information. POAMA will be upgraded to the new higher resolution ACCESS-S seasonal prediction system in 2017-18, in collaboration with the UK Met Office.

Dynamical forecasts potentially offer improved performance relative to statistical forecasts, particularly given baseline shifts in the environment due to climate change. Seasonal forecasts are most useful when management options are available for implementation in response to the forecasts. Improved management of marine resources, with the assistance of such forecast tools, is likely to enhance future planning, industry resilience and adaptive capacity under climate change.

### The Met Office global models: how to change the complete system and still have something that works

Michael Thurlow
UK Met Office
Monday, 1st May 2017
11:00 AM - 12:00 NOON

Abstract: Looking back over the recent history of global NWP there are two main developments that can be summarised as make it big and make it complicated. Both are driven by the HPC power available to use at each stage and our scientific belief that improved algorithms and coupling between systems will provide improved forecasts and, hopefully, happy customers. Unfortunately our software systems took a back seat in this drive forward and it became very clear that the suite control systems that we had been using would no longer be able to manage the workload. In this talk I will outline the Met Office's experience in moving to the multiply-dependent world of current NWP systems through the adoption of Cylc in 2013 and subsequent scientific developments. I will also have a look forward to where we are planning to go with the global model in the next few years and the challenges that are thrown up as a consequence.

### Model Reduced 4D-Var in Application to a North Sea Ecosystem Model

Joanna Pelc
NASA Goddard Space Flight Center
Wednesday, 5th Apr 2017
10:00 AM - 11:00 AM

Abstract: The model reduced 4D-Var (Vermeulen and Heemink, 2006, Pelc et al., 2012) is applied to a North Sea ecosystem model BLOOM/GEM (Blauw et al., 2009). Obtaining gradients for ecological models is commonly known to be challenging. However, since the method is performed in the reduced space, the implementation of the adjoint of the tangent linear approximation of the original model is not required. Moreover, the estimation of the initial states is performed in the reduced space which decreases the size of the control vector by several orders of magnitude.

First the technique is tested in a twin experiment framework, and next the method is performed with measurements obtained from the satellite images of chlorophyll-a (MERIS). Biological parameters' calibration is combined with the estimation of the initial condition of the model. The initial concentration fields are estimated in the substantially reduced space, decreasing the size of the control vector from more than ten thousands down to a size of only a few hundreds.

### Large ensemble analysis of stratosphere-troposphere coupling on the seasonal time scale

Martin Jucker
University of Melbourne
Wednesday, 22nd Mar 2017
10:00 AM - 11:00 AM

Abstract: This talk will focus on stratosphere-troposphere coupling on the seasonal time scale. It will examine the life cycle of the most prominent events of strong stratosphere-troposphere coupling observed in the atmosphere, the so-called Sudden Stratospheric Warmings (SSWs). After a general introduction to stratospheric dynamics and its significance in seasonal forecasting, a large number of events sampled from an idealised General Circulation Model (GCM) will be analysed statistically. It is shown that composites of SSWs which have a long-term effect on the state of the troposphere (downward 'propagating') strongly depend on the exact definition of what is considered 'propagating'. However, there are two characteristics which seem consistently linked to propagating events, namely a somewhat more barotropic response at the onset date and longer persistence in the lower stratosphere after the onset. We will show that rather than tropospheric forcing (such as upward Eliassen-Palm fluxes), internal stratospheric variability is the decisive factor in setting up the environment for an SSW to occur.

### Drift in Decadal Predictions -- Ideas for Parametric Post-processing

Henning Rust
Free University of Berlin
Friday, 17th Mar 2017
1:30 PM - 2:30 PM

Abstract: A climate projection is a climate model simulation with prescribed boundary conditions but starting from an arbitrary climate state on the model's attractor. The idea of decadal climate predictions is instead to use the same boundary conditions but initialise the climate model with a climate state close to the observed state. As this is not necessarily a state on the climate model's attractor, simulations show an artificial behaviour thought to be the model's attempt to get back to its preferred states. This is artificial in the sense that it is neither a reaction to changing boundary conditions nor part of natural climate variability. From a data analysis perspective, this behaviour is conceived as a model bias changing with forecast lead-time. The term frequently used for this behaviour is "drift". This talk is about a concept to quantify drift and about parametric strategies to reduce its influence on the predictions by post-processing either the ensemble mean ("drift correction") or the full probabilistic prediction ("re-calibration").

### A hierarchical spatial statistical model for mapping Australian temperature data

Sandy Burden
University of Wollongong
Wednesday, 15th Mar 2017
10:00 AM - 11:00 AM

Abstract: Producing high resolution maps of complex environmental processes, such as maximum daily temperature in Australia, is a challenging problem. Temperature observations are noisy, observation locations are irregular, and in some regions observations are sparse. Further, since the temperature process is subject to many sources of variability at multiple scales, it has a complex spatial dependence structure that must be accounted for when producing estimates at unobserved locations. One approach is to formulate a hierarchical spatial statistical model for Australian temperatures. Hierarchical models provide a versatile framework within which to consider the problem, since they can account for complex dependence structures and uncertainty in the temperature process and its parameters, as well as measurement error and noise in the observations. The models are estimable using multiple sources of data with varying support, and importantly, they can quantify the uncertainty of estimates. In this talk I describe a general approach for modelling an environmental process using a hierarchical spatial statistical model with data-driven parameter estimates. I illustrate the approach using Australian temperature data with the aim of producing a map of the monthly mean maximum daily temperatures. In the talk I include details of the analysis procedure and data pre-processing, and I consider some of the strengths and limitations of the approach along with potential extensions of the analysis.

### Turbulence Near Deep Convection

Dragana Rajak
Bureau of Meteorology
Wednesday, 8th Mar 2017
10:00 AM - 11:00 AM

Abstract: Deep convective clouds are one of the most important sources of turbulence that affect aviation. This type of turbulence does not only occur within clouds but also in the clear air regions outside the cloud (so-called near-cloud turbulence (NCT)). NCT can occur significant distances from thunderstorms making it invisible and hard to detect using standard on-board or ground-based radars and is therefore a significant hazard to aircraft. Mechanisms that generate NCT are not yet fully understood and are the focus of this study. Here, turbulence generation by organised convection (viz. supercells and mesoscale convective systems) using three-dimensional (3D) simulations conducted with the Weather Research and Forecasting model will be examined. Firstly, results from high-resolution idealised simulations will be shown, with a focus on the role of 3D cloud-induced flow perturbations on turbulence generation and their sensitivity to different background conditions like wind shear. Secondly, a case study of an actual turbulence encounter that occurred in the clear air outside a convective system on 3 June 2005 will be examined. All simulations show regions of turbulence that extend large distances (>50-100 km) away from the active deep convection (i.e. regions with high radar reflectivity). In idealised simulations, turbulent regions are related to strong upper-level storm outflow and the associated enhanced vertical shear. In the case study simulations, turbulence is due to a large amplitude breaking gravity wave that propagates ahead of the convective system, but also due to combination of large-scale flow and upper-level storm outflow. The results of these simulations and their implications for turbulence avoidance by aircraft will be discussed.

### The Unified Model 100 years after L. F. Richardson

Nigel Wood
UK Met Office
Friday, 24th Feb 2017
2:00 PM - 3:00 PM

Abstract: The joint Met Office, NERC, and STFC GungHo project (2011-2016) proposed a new dynamical core for the Unified Model that targets efficient use of next on High Performance Computers (HPCs). That new dynamical core is based on a finite-element discretization applied to the cubed-sphere. However, the current Unified Model software infrastructure is unsuitable for such a change to the model and is unlikely to meet the challenges of the next generation of supercomputers. Therefore the GungHo project additionally proposed a new modelling infrastructure. The LFRic project (named after L.F. Richardson) is implementing that design as the replacement for the Unified Model. As well as having to support the finite element data structures required by GungHo, LFRic must be flexible to changes in both HPC architecture and model formulation (and hence is based on unstructured meshes), whilst delivering good computing performance on future exascale supercomputers. To achieve these goals, the scientific code has been separated from the underlying computational infrastructure, so that changes can be made to either without impacting the other. Code auto-generation techniques are used to optimize the model for different system architectures. This split of responsibility is termed the Separation of Concerns''. Scaling experiments on the new Met Office supercomputer using auto-generated code confirm the viability of the approach. This talk will give an overview of the design of both GungHo and LFRic, and the rationale for those designs. The planned time line, and the challenges that that presents us with, will also be discussed.

### An Expert System Approach for Detecting Areas of High Ice Water Content for Aviation

Julie Haggerty
NCAR
Wednesday, 22nd Feb 2017
1:00 PM - 2:00 PM

Abstract: Ingestion of large amounts of ice particles by jet engines, known as ice crystal icing (ICI), appears to be the culprit in over 150 engine power-loss and damage events during the past two decades. Usually occurring in convective weather conditions at high altitudes, heated probes also appear to be susceptible to this threat. Using information from prior ICI events, a prototype real-time diagnostic tool known as the Algorithm for Prediction of HIWC Areas (ALPHA) was developed. ALPHA applies fuzzy logic methodology to define ranges of interest for a set of meteorological predictors of HIWC conditions. Input fields from satellite, model, and radar are blended to yield a 3-dimensional field estimating the likelihood of HIWC conditions.

Data from field campaigns in 2014-2015 provide an extensive set of research quality measurements in the conditions typically associated with ICI events. The High Altitude Ice Crystal - High Ice Water Content (HAIC-HIWC) field campaigns in Darwin, Australia, and Cayenne, French Guiana, and the HIWC Radar Study in Ft. Lauderdale, USA were designed to enhance knowledge of ice crystal icing processes in deep convective clouds. The availability of a substantive data set with flight level measurements of ice water content enables the use of a machine learning approach to objectively define the parameter set used in the ALPHA fuzzy logic implementation. A randomly selected portion of the data set is used for algorithm training, while the remainder of the data set is reserved for independent evaluation of the resulting algorithm. This presentation will describe optimization of ALPHA and performance statistics for detecting HIWC conditions.

### Robust contribution of decadal anomalies to the frequency of central-Pacific El Niño

Jing-Jia Luo
Bureau of Meteorology
Wednesday, 22nd Feb 2017
10:00 AM - 11:00 AM

Abstract: During year-to-year El Niño events in recent decades, major sea surface warming has occurred frequently in the central Pacific. This is distinct from the eastern Pacific warming pattern during canonical El Niño events. Accordingly, the central-Pacific El Niño exerts distinct impacts on ecosystems, climate and hurricanes worldwide. The increased frequency of the new type of El Niño presents a challenge not only for the understanding of El Niño dynamics and its change but also for the prediction of El Niño and its global impacts at present and future climate. Previous studies have proposed different indices to represent the two types of El Niño for better understanding, prediction and impact assessment. Here, we find that all popularly used indices for the central-Pacific El Niño show a dominant spectral peak at a decadal period with comparatively weak variance at interannual timescales. Our results suggest that decadal anomalies have an important contribution to the occurrence of the central-Pacific El Niño over past decades. Removing the decadal component leads to a significant reduction in the frequency of the central-Pacific El Niño in observations and in Coupled Model Intercomparison Project Phase 5 simulations of preindustrial, historical and future climate. The paper was published at Scientific Reports on 5 Dec 2016.

### Why are GCMs bad at simulating the MJO?

Gilles Bellon
University of Auckland
Tuesday, 14th Feb 2017
2:30 PM - 3:30 PM

Abstract: Most General Circulation Models still have difficulties to simulate realistic Madden-Julian Oscillation events. Two leading hypotheses have been proposed to explain this widespread deficiency. The first one posits that the GCMs do not simulate a realistic vertical profile of diabatic heating associated with clouds and cloud envelopes. The second hypothesis considers that the GCMs are unable to simulate the large-scale, spatial organisation of convection necessary to initiate an MJO event because their parametrisations of convection lack the sub-grid and inter-grid processes that are essential to its development. The presentation will present studies that address questions inspired by these hypotheses: how do diabatic-heating profiles simulated by GCMs compare to observations? On what scales should we expect convection to organise?

### Regulating the cumulative impacts of groundwater withdrawals with an eye to water data

Rebecca Nelson
University of Melbourne
Tuesday, 14th Feb 2017
11:00 AM - 12:00 NOON

Abstract: The regulation of groundwater extraction in Australia has shifted dramatically through an intense era of water reforms spanning three decades. A key outstanding issue is controlling withdrawals with an eye to their cumulative impacts on groundwater resources and dependent ecosystems. Such control is complicated not just by the incremental additive effects of many small withdrawals, but also by interactive and synergistic effects. This complexity is intensified further by data paucity, potentially significant time lags, and simultaneous background changes to natural systems, such as those caused by climate change. Obtaining, sharing and using water data are key challenges associated with managing groundwater, and particularly regulating the cumulative impacts of groundwater extraction. For example, much of the attention on regulating cumulative impacts has focused on ensuring that traditionally unregulated types of groundwater withdrawals (about which there is little data) are controlled or at least monitored, and on special-purpose regulation of clusters of withdrawal activities, as in coal seam gas extraction (which may produce data that does not automatically feed into regulatory structures for more common activities). This talk sets out the regulatory context for managing the cumulative impacts of groundwater extraction, to provide context for how water data is used - and could be used - in this challenging scenario. It focuses on Australia, where attention to cumulative impacts has been relatively recent, and contrasts our frameworks with that of the United States.

### Plains Elevated Convection At Night (PECAN) field campaign: Motivation and highlights

Tammy M. Weckwerth
NCAR/EOL
Monday, 13th Feb 2017
11:00 AM - 12:00 NOON

Abstract: The Plains Elevated Convection At Night (PECAN) field campaign was designed to improve our understanding and forecast skill of the North American Great Plains nocturnal maximum in warm season precipitation. This nocturnal rainfall maximum is puzzling because the convective activity is out of phase with the solar heating maximum. There were five primary science components of PECAN: 1) convection initiation, 2) bores, 3) low level jets, 4) mesoscale convective systems (MCSs) and 5) storm and MCS-scale numerical weather prediction. For six weeks in June-July 2015, over 350 researchers, including 130 students, came together to operate over 100 instruments designed to sample the U. S. Great Plains atmosphere prior to and during these nocturnal precipitation events. In addition to fielding standard operational and research instruments, such as radars, soundings and surface stations, PECAN created ten unique mobile and fixed integrated profiling systems, termed PISAs (PECAN Integrated Sounding Arrays), that were utilized for the field campaign. All of the PISAs had the capability to sample vertical profiles of wind, temperature and moisture albeit through different combinations of instruments. I will present an overview of the motivation, goals and instruments of PECAN, discuss the field campaign and summarize some early science highlights.

### Effects of the changing heating profile associated with melting layers in a climate model

Hongyan Zhu
Bureau of Meteorology
Wednesday, 25th Jan 2017
10:00 AM - 11:00 AM

Abstract: This study investigates the impact of modifying the melting behaviour at the freezing level in the GA2.0 version of the Met Office UM. It finds that by allowing snow to melt over a greater depth, biases in rainfall over the Maritime Continent (MC) are improved, and there is an indication of benefits to the MJO. This study uses moistening diagnostics under weak temperature gradient theory to explain how and why changes to the treatment of melting influence tropical rainfall biases. The modified melting experiment increases the lower tropospheric diabatic heating rate per unit column-integrated convective heating in the MC, which helps to increase lower tropospheric vertical moisture advection per unit column convective heating, making conditions more favorable for convection there. Changes of the opposite sense occur in tropical ocean regions of the west Pacific and Indian Ocean. Changes in lower tropospheric radiative heating per unit convection produced by the different treatment of melting are particularly influential in engendering mean precipitation changes between the experiments. Differences in precipitation in the MC region between the control and melting experiments and opposite changes in oceanic regions to the east and west are linked through changes in the Walker circulation, making it unclear which region is most influential for forcing the improvement in the pattern of precipitation biases. Sensitivity experiments that artificially enhance convection in one region through imposition of SST anomalies produce a negative precipitation response in the other region.

### Wind and wave extremes from very large forecast ensembles

Oyvind Breivik
Norwegian Met Institute
Wednesday, 14th Dec 2016
10:00 AM - 11:00 AM

Abstract: Global return values of marine wind speed and significant wave height are estimated from very large aggregates of archived ECMWF ensemble forecasts at +240-h lead time. Long lead time ensures that the 50 ensemble members in each forecast cycle represent independent draws from the model climate. Compared with ERA-Interim, a reanalysis, the ensemble yields higher return estimates for both wind speed and significant wave height. Confidence intervals are much tighter due to the large size of the dataset. The period (9 yrs) is short enough to be considered stationary even with climate change. Furthermore, the ensemble is large enough for non-parametric 100-yr return estimates to be made from order statistics. These direct return estimates compare well with extreme value estimates outside areas with tropical cyclones. Like any method employing modelled fields, it is sensitive to tail biases in the numerical model, but we find that the biases are moderate outside areas with tropical cyclones. We are now in the process of extending the method to selection of peaks from the forecast range +216-240 hours. This brings the total length of the data set to the equivalent of 600 years of data. Preliminary results suggest that previous return estimates are robust.

### Turbulence near deep convection

Dragana Rajak
Bureau of Meteorology
Monday, 12th Dec 2016
11:00 AM - 12:00 NOON

Abstract: Deep convective clouds are one of the most important sources of turbulence that affect aviation. This type of turbulence does not only occur within clouds but also in the clear air regions outside the cloud (so-called near-cloud turbulence (NCT)). NCT can occur significant distances from thunderstorms making it invisible and hard to detect using standard on-board or ground-based radars and is therefore a significant hazard to aircraft. Mechanisms that generate NCT are not yet fully understood and are the focus of this study.

Here, turbulence generation by organised convection (viz. supercells and mesoscale convective systems) using three-dimensional (3D) simulations conducted with the Weather Research and Forecasting model will be examined. Firstly, results from high-resolution idealised simulations will be shown, with a focus on the role of 3D cloud-induced flow perturbations on turbulence generation and their sensitivity to different background conditions like wind shear. Secondly, a case study of an actual turbulence encounter that occurred in the clear air outside a convective system on 3 June 2005 will be examined. All simulations show regions of turbulence that extend large distances (>50-100 km) away from the active deep convection (i.e. regions with high radar reflectivity). In idealised simulations, turbulent regions are related to strong upper-level storm outflow and the associated enhanced vertical shear. In the case study simulations, turbulence is due to a large amplitude breaking gravity wave that propagates ahead of the convective system, but also due to combination of large-scale flow and upper-level storm outflow. The results of these simulations and their implications for turbulence avoidance by aircraft will be discussed.

### Evolution of forecast evaluation methods: A 2016 perspective

Barbara Brown
NCAR
Friday, 2nd Dec 2016
2:00 PM - 3:00 PM

Abstract: Forecast evaluation has been an important - and often controversial - topic in meteorology practically since the first weather forecasts were issued more than 100 years ago. While many verification methodologies have persisted through much of this history, the last few decades have seen a renaissance in the development of new approaches, as well as greater understanding of capabilities that have existed since the early days of forecasting. Evaluation approaches for specific types of forecasts such as nowcasts have consistently been associated with numerous special issues, including the extremeness of the weather conditions, the use of remotely sensed data, and the importance of user-relevant approaches. This talk will consider new developments in forecast evaluation - and the verification community - in recent years. Special focus will be given to new spatial methods, methods for extreme event forecasts, and user-relevant approaches that bring together the usefulness of the forecasts with their quality.

### Evaluation of ACCESS-S1 for prediction of Australian climate

Eun-Pa Lim and Li Shi
Bureau of Meteorology
Wednesday, 30th Nov 2016
10:00 AM - 11:00 AM

Abstract: In 2015 the Bureau of Meteorology (BoM) made a major strategic decision to advance its seasonal prediction capability, and so, BoM has been developing a new seasonal prediction system, called ACCESS-S1 (the Australian Community Climate and Earth-System Simulator-Seasonal prediction system). ACCESS-S1 built upon the latest climate model Global Seasonal forecast system version 5 using the Global Coupled configuration 2 (GloSea5-GC2) from the UK Met Office (UKMO) along with locally developed system enhancements. The ACCESS-S1 system is planned to become operational, replacing the BoM's current operational system POAMA2 by mid-2017.

The UKMO's GloSea5-GC2 includes a number of state-of-the-art features compared to POAMA2, such as substantially higher horizontal and vertical resolution, improved model physics and parameterization, a multi-level land surface model, and an interactive multi-level sea-ice model.

Using a locally developed ensemble generation scheme, 22 member ensemble hindcasts initialised on the 25th of January, April, July and October and on the 1st of February, May, August and November have been generated for the period 1990-2012 from the ACCESS-S1 system. For this presentation, we assessed the performance of ACCESS-S1 in simulating the climate mean states, predicting large-scale climate drivers such as ENSO, IOD and SAM and predicting Australian climate.

Compared with the POAMA2, ACCESS-S1 demonstrates significantly reduced mean climate bias globally and over Australia. Also, the large-scale climate modes such as ENSO, IOD and SAM are better predicted by ACCESS-S1. Importantly, Australian rainfall, Tmax and Tmin are better predicted than POAMA2 at short lead times from multi-week to 1 month. However, there is still a large scope for improvement, including overestimation of the magnitude of ENSO and IOD and underestimation of the teleconnection strength of ENSO, IOD and SAM to Australian climate.

### The Southern Ocean westerlies influencing large-scale connectivity

Stephanie Downes
University of Tasmania
Friday, 25th Nov 2016
10:00 AM - 11:00 AM

Abstract: Two major multi-model efforts have dominated the climate modelling community over the past five years, namely the Coordinated Ocean-ice Reference Experiments Phase II (CORE-II) and the Coupled Model Intercomparison Project Phase 5 (CMIP5). This talk will merge models from both efforts and show how the temporal evolution of major Southern Ocean metrics are linked to the westerly winds. Specifically, the Antarctic Circumpolar Current (ACC) and the lower branch of the meridional overturning circulation (MOC). Models where the winds have less variability show a stronger relationship between the ACC and lower MOC, but only for overturning in density space. Sadly many models do not archive circulation in density coordinates (particularly CMIP5 models), making this relationship difficult to diagnose. The strength of the lower MOC and ACC are governed by lateral density gradients. Come along and learn more about multi-model efforts and what diagnostics matter for future Southern Ocean analyses

### 'Climate sensitivity', climate change and cloud feedbacks ... Where does the science stand?

Rob Colman
Bureau of Meteorology
Wednesday, 16th Nov 2016
10:00 AM - 11:00 AM

Abstract: How much will Australia's and the globe's temperature change over the coming century? In large part this is dictated by the climate sensitivity. Climate sensitivity is a standard measure, long used by the climate community in defining how large the temperature response will be to climate change. The Equilibrium Climate Sensitivity (ECS), in particular, is defined as the climate change that occurs when the atmospheric CO2 levels are doubled, once the climate system comes to a new equilibrium. In 1979 it was estimated to be between 1.5 and 4C. Nearly 40 years later the estimated range is the same. Climate feedbacks and in particular cloud feedbacks are known to be the key factors underlying this uncertainty.

This talk will address a few questions concerning climate sensitivity and future global and Australian climate change:

• What does climate sensitivity mean for global and Australian temperature projections?
• Why is the range so large?
• Why has the range not reduced after almost 40 years of advances in climate science?
• What do models/observations/theory tell us?
• What is the future for understanding and constraining climate sensitivity and climate projections?

### Interactions of climate variability and change in simulations and observations

Dr Ed Hawkins
Monday, 14th Nov 2016
10:00 AM - 11:00 AM

Abstract: Earth's climate is changing. Temperatures are increasing globally, but at varying rates, both spatially and over time. Arctic sea ice is melting overall, but rather unevenly. These features reflect the complex interaction of long-term anthropogenic changes and natural climate fluctuations. Making predictions of the future, with appropriate estimates of uncertainty, requires an appreciation of this complexity, especially as much of the uncertainty on regional spatial scales is irreducible. This talk will discuss our historical understanding of climate variations and how we are using both observations and climate simulations to provide estimates of the uncertainty in future climate predictions, and how we might communicate these issues.

Biography: Ed Hawkins is a climate scientist in the National Centre for Atmospheric Science at the University of Reading. He is a NERC Advanced Fellow and was a Contributing Author to the 5th Assessment Report of the IPCC. He also runs a popular climate blog (http://www.climate-lab-book.ac.uk) and can usually be found on twitter (@ed_hawkins).

### Utilization of Doppler Weather Radar data at NCMRWF

Dr Amar Jyothi
NCMRWF
Wednesday, 2nd Nov 2016
10:00 AM - 11:00 AM

Abstract: The characterization of quality of data from Indian DWR (Doppler weather Radars) in terms of their Quality Index (QI) is essential for assimilating the data into NWP (Numerical Weather Prediction) models. The QI maps give a quick overview about the inherent static errors in the DWR data. Various quality control algorithms are applied for the generation of QI corresponding to both dynamic and static errors in the data. The near real time access to the DWR data at NCMRWF enables the generation of an accumulated gridded radar rainfall product. The quality of DWR derived VVP profiles used for the assimilation in NCUM model will be presented. The VVP profiles provide reasonably accurate estimates of the vertical wind structure in the troposphere over radar locations which can be effectively used in the numerical weather prediction system. Some of these results will be presented.

### Ocean Processes around Aotearoa New Zealand and Some Connections to Marine Stakeholders (aka NZ Oceanography 101)

Dr Craig Stevens
NIWA
Thursday, 13th Oct 2016
10:00 AM - 11:00 AM

Abstract: From the perspective of Aotearoa New Zealand's shelf seas, the Zealandia system (the drowned continent of which the islands of New Zealand is the above-water component) is fed from the north by the East Australia Current and the Southland Current from the south. The Tasman Sea to the west is a poorly defined and only weakly energetic region. In this talk I will highlight on-going efforts to observe these oceanic drivers, some science highlights, and connect with places to where marine stakeholders are using, or should be using, this knowledge.

### Variability and long-term trends in severe coastal flooding in south-eastern Australia since the mid-19th century, associated storms and death tolls

Scott Power
Bureau of Meteorology
Wednesday, 5th Oct 2016
10:00 AM - 11:00 AM

Abstract: Millions of Australians live in a 1500 km stretch of coastal catchments in south-east Australia from Brisbane, through the rapidly growing Gold Coast and Northern Rivers district, through Sydney, to the state border. Major flooding in this region causes death, economic loss and major disruptions to the lives of its inhabitants. Using a range of historical sources ranging from hydrological and meteorological data, through to newspaper articles, we identified over 250 severe floods since 1860. We considered a flood to be major if it caused inundation of a river within approximately 50 km of the coast or if there was non-riverine flooding over land near the coast, extending 20 km or more along the coast. Some of the most extreme events identified occurred in the 19th century and early-to-mid 20th century. If such storms occurred today, they would have catastrophic impacts due to the massive increase in urban development in the study region since that time.

The El Niño-Southern Oscillation (ENSO) is shown to strongly modulate the frequency of severe coastal flooding and the number of associated death tolls. For example, the average numbers of coastal floods and deaths associated with freshwater drowning in La Niña years are 92% and 220% higher than the corresponding averages in El Niño years, respectively. The average number of deaths per flood in La Niña years is 3.2, which is 66% higher than the average in El Niño years. Fortunately, death tolls have been relatively low since 1960, despite an increase in the number of floods and a very marked increase in population density. This presumably reflects improvements in weather forecasting, emergency services and an increased awareness of the risks associated with flooding in the broader community.

There has been a statistically significant, 50% increase in the frequency of major floods since the late 19th century. While this is consistent with an expectation that anthropogenic global warming may lead to a projected increase in flood risk in response to global warming over the coming century in many regions, further research is needed to clarify the relative importance of possible contributors, including anthropogenic global warming and land use change.

### The "possible" future of rainfall in Victoria and its implications

Bertrand Timbal
Bureau of Meteorology
Wednesday, 28th Sep 2016
10:00 AM - 11:00 AM

Abstract: Victoria and more broadly South-Eastern Australia (SEA) has experienced reduced rainfall in the last 30 years compared to the historical record. This reduction has been in the cool part of the year from Apr to Oct while the rest of the year has experienced a small increase. There is a great interest in the future of SEA rainfall, and whether these patterns will continue as they have had consequences for water management, agriculture and the natural environment. This has been extensively studied over the last 10 years with dedicated research programs such as the South Eastern Australia Climate Initiative (SEACI) from 2006 to 2012 and the Victorian Climate Initiative (VicCI) from 2013 to 2016. In this presentation, we will review some of the key advances we made over the 10 years of research on 1) improving our understanding of current and future changes in the large-scale atmospheric circulation affecting SEA climate, 2) increasing our understanding of the uncertainties in the range of future rainfall and 3) the implications for projections of future runoff across the region

### Understanding and predicting the seesaw between the Somali and Maritime Continent Cross-Equatorial Flows (CEFs)

Chen Li
IAP, University of Chinese Academy of Sciences
Monday, 26th Sep 2016
10:00 AM - 12:00 NOON

Abstract: The Somali and Maritime Continent Cross-Equatorial Flows (CEFs) represent a key component of the Indian Summer Monsoon and the Australia-East Asia Monsoon systems, and play an important role in linking the interhemispheric climate and environment. During boreal summer (June-July-Aug), a significant negative correlation (seesaw) has been found between the Somali CEF and the Maritime Continent (MC) CEF on interannual time scale. This CEF-seesaw shows a close relation to a dipole rainfall anomaly between the two monsoon systems. In this talk, I am going to introduce the CEF-seesaw phenomena, its underlying mechanism based on centennial observations (ERA-20C) and UM-GA6 experiments, and the predictability of the two CEFs and their seesaw based on POAMA-2, SINTEX-F and ACCESS-S1 hindcasts. Multi-decadal change of the CEF-seesaw over the past century, its possible mechanism, and the impact of global warming on the seesaw will be also discussed.

### Simulation Analysis-based Risk Evaluation (SABRE) Fire. Operational Stochastic Fire Spread Decision Support Capability in the Queensland Fire and Emergency Services

Ben Twomey
Andrew Sturgess
Queensland Fire and
Emergency Services
Friday, 23rd Sep 2016
11:00 AM - 12:30 PM

Abstract: In this presentation we introduce the Queensland Fire and Emergency Services (QFES) approach to leveraging simulation-based outputs for operational decision making incorporating lessons learned from the FIRE-DST. QFES is creating an all hazards, operational decision support system known as Simulation Analysis-based Risk Evaluation (SABRE). The first operational capability within SABRE is known as SABRE Fire, and it has been undergoing prototyping and operational evaluation on live bushfire incidents since Oct 2015. While sharing some similar, high level goals with the FIRE-DST, SABREs conceptual architecture and aims differ significantly.

SABRE aims to permit open and easy definition and management of uncertainties in input data for any suitable hazard simulation. Like the FIRE-DST, it employs Monte Carlo sampling to generate many trials of a deterministic simulation (Phoenix) with inputs randomly drawn from definable probability densities vice single, best estimate forecast values. It then translates the multiple outputs into intuitive and interactive decision support tools based on risk while ensuring this is done consistently, robustly and within useful operational timelines.

### Multi-year potential predictability in temperature and precipitation in the southern hemisphere

Christine Chung
Bureau of Meteorology
Wednesday, 14th Sep 2016
10:00 AM - 11:00 AM

Abstract: Decadal variability is evident in rainfall, temperature, severe weather and flood frequency around the globe. It is therefore important that we understand the cause of this variability and the extent to which it can be predicted. Here we assess southern hemisphere decadal climate variability and its predictability potential in an ensemble of CMIP5 models. Global hotspots of subsurface ocean decadal variability are identified, revealing the Southern Tasman Sea as having the strongest predictability potential. We find that subsurface temperature variability in the Southern Tasman Sea primarily arises in response to preceding changes in southern hemispheric winds. This variability is multi-year to decadal in character and it is coherent with surface temperature in large parts of the southern hemisphere almost a decade later. Variability in the Southern Tasman Sea is also coherent with Antarctic precipitation 1-3 years later in some models. This study suggests that the southern Tasman Sea is a useful indicator of future southern hemisphere climatic conditions more broadly.

I will also briefly discuss some results from a study by Meehl et al (2016) into the mechanisms behind accelerated Antarctic sea-ice expansion between 2000 and 2014. The negative phase of the Interdecadal Pacific Oscillation in global coupled climate models is characterized by anomalies similar to the observed sea-level pressure and near-surface 850 hPa wind changes near Antarctica since 2000 that are conducive to expanding Antarctic sea-ice extent. These atmospheric circulation changes are shown to be mainly driven by precipitation and convective heating anomalies related to the Interdecadal Pacific Oscillation in the equatorial eastern Pacific, with additional contributions from convective heating anomalies in the South Pacific convergence zone and tropical Atlantic regions.

### A high-resolution reanalysis for Australia background and progress to date

Peter Steinle, Dörte Jakob,
Chun-Hsu Su
Bureau of Meteorology
Monday, 12th Sep 2016
10:00 AM - 11:00 AM

Abstract: The Bureau of Meteorology has commenced activities to produce a high-resolution reanalysis covering the Australian continent and the surrounding region, including parts of southeast Asia, New Zealand, and south to near the edge of the Antarctic sea-ice. The Bureau has committed considerable resources to complete the reanalysis for a period covering approximately the last 25 years and is actively seeking partners willing to invest in this project.

The regional reanalysis will use the Australian Community Climate and Earth-System Simulator (ACCESS) model, at 12 km horizontal resolution (that of the current generation operational ACCESS-R system), and provide gridded data sets every hour. Reanalyses at finer scale subdomains are also undertaken. Tasmanian emergency management agencies have secured Federal Government funding to support generation of a 1.5 km resolution subdomain over the state, and NSW Rural Fire Service has secured funding to support the ACCESS-C NSW domain at that resolution. Initial work has now commenced to deliver reanalyses for these subdomains at high spatial resolution. The reanalysis dataset will underpin the computation of average and high-end fire weather conditions that can be expected to support fire behaviour and risk modelling studies.

### Data fusion based drought index (DFDI)

Monash University
Wednesday, 7th Sep 2016
10:30 AM - 11:30 AM

Abstract: DFDI is a new index to address existing shortcomings in spatio-temporally evaluating water stress conditions by employing individual drought indices for a specific location. The proposed methodology is able to monitor water stress conditions of terrestrial ecosystems by objectively linking different aspects of the ecosystem such as water availability and vegetation conditions. The developed index, called Data Fusion-based Drought Index (DFDI), makes use of advanced statistical methods, and also considers eco-meteorological characteristics, such as landuse, land-cover, and climate of an area to determine the water stress conditions at each time step for each specific location. More information about the details of DFDI would be available at following recently published paper: Azmi, M., Rüdiger, C., Walker, J. (2016), A data fusion-based drought index, Water Resources Research, 52:2222-2239, DOI: 10.1002/2015WR017834.

### The AusPollen partnership project: partnering to address knowledge gaps and needs for monitoring variability in airborne pollen exposure across Australia

Janet Davies
QUT
Wednesday, 24th Aug 2016
04:00 PM - 05:00 PM

Abstract: Allergic rhinitis (hay fever) and episodic asthma attacks triggered by grass pollen, regularly affect up to three million Australians. Natural factors such as ecology, land cover, weather and wind direction influence seasonal and daily levels of airborne pollen. Even moderate levels of airborne grass pollen reduce patient quality of life. Despite the high prevalence, medical and socioeconomic burden of allergic rhinitis and asthma, Australia is one of the few developed countries without a national pollen monitoring program. The interdisciplinary Australian Aerobiology Working Group formed in 2013 to analyse existing pollen count data sourced from 17 sites in different regions of Australasia and three French comparator sites. The currently available pollen count data sets showed distinct regional and seasonal variability in airborne pollen levels. Novel methods for integrating remote sensing and pollen aerobiology were examined. This current multi-site prospective AusPollen project was co-developed with partner organisations to address key gaps in information on exposure to environmental triggers for patients with allergic respiratory disease. The partners; Asthma Australia, ASCIA, The Bureau of Meteorology, MeteoSwiss (Switzerland), Commonwealth Scientific and Industrial Research Organisation and Stallergenes, are all key stakeholders in care and education of allergy and asthma patients, and in the establishment of operational forecast systems for airborne pollen. The goal of this AusPollen Partnership Project will extend research evidence by implementing and evaluating a functional pollen alert apps and inform elements of a pollen forecast system can be developed into an operational national forecast service for all of Australia. It is anticipated that the AusPollen will lead to reduced symptoms, improved quality of life, and will empower patients to self-manage their condition.

### Visualizing and Verifying Probabilistic Forecasts of the Madden-Julian Oscillation

Andrew Marshall
Bureau of Meteorology
Thursday, 11th Aug 2016
11:00 AM - 12:00 NOON

Abstract: We describe a new approach for presenting probabilistic forecasts of the Madden-Julian Oscillation (MJO) based on the community standard Realtime Multivariate MJO index, using forecasts from version 2 of the Predictive Ocean Atmosphere Model for Australia. In particular, this proposed new metric overcomes the difficulty of interpreting a dispersive ensemble member plume in the standard deterministic MJO forecast display. Using the example real-time forecast initialised on 8 Mar 2015, we exploit available ensemble forecast information in a probabilistic phase-space display to present the likelihood of the MJO occurring in any given phase over five-day intervals out to 40 days lead time. Beyond real-time MJO monitoring and prediction, this new metric also provides a tool for forecast verification using probability-based skill scores.

### Local to climate scale drivers of coastal storm environments — The Coastal Convective Interactions Experiment (CCIE)

Joshua Soderholm
University of Queensland
Wednesday, 10th Aug 2016
10:00 AM - 11:00 AM

Abstract: Prediction of convective storm environments relies principally upon the broad-scale meteorology (e.g., synoptic boundaries and air masses) in contrast to mesoscale (2 —200 km) processes within the planetary boundary layer. In populated coastal regions, the considerable diverse physical setting leads to a number of complex PBL circulations and processes, presenting a significant uncertainty for convection nowcasting. On larger scales, quantifying the influence of climate variability upon storm frequency can offer a potential seasonal prediction tool. These challenges will be addressed through the presentation of major findings from the Coastal Convective Interactions Experiment. This experiment incorporates meteorological and climatological analyses to deliver a comprehensive understanding of the coastal storm environment of South East Queensland. Highlights include (i) the design of Australia's first subtropical thunderstorm field campaign (including CP-2 and UQ-XPOL), (ii) analysis of thunderstorm hotspots drivers and ENSO influence using an 18 year radar-derived climatology, and (iii) the diurnal precondition mechanisms which favour the development of severe convective storms.

### Some Forecast-Science Highlights: 1975 — 2015

Noel Davidson
Bureau of Meteorology
Thursday, 4th Aug 2016
02:45PM - 03:45PM

Abstract: There have been truly amazing changes in the science and technology upon which the Bureau has based its continual improvement over the last 40 years. This talk will summarize my perspective on the evolutionary, occasionally revolutionary changes, based mainly on collaborations with numerous talented colleagues. Topics and discussion will touch on: Trends in actual weather prediction over the last 40 years, improvements in short-term TC prediction, short-term fluctuations in the Australian Monsoon, tropical and tropical cyclone initialization and prediction, extratropical-tropical interaction, diagnosis of Tropical Cyclone behaviour, inner-core structure change of Tropical Cyclones, environmental interactions and dynamics of extreme rain events, and some ongoing forecast problems.

### Attribution of extreme monthly heat events using POAMA

Guomin Wang
Bureau of Meteorology
Wednesday, 3rd Aug 2016
10:00 AM - 11:00 AM

Abstract: Here we present a method for the attribution of extreme climate events using an initialised climate prediction system POAMA. To illustrate the use of this method, we attribute the causes of two recent month long record heat events that occurred in Oct 2014 and 2015 over Australia. The events were reforecast twice, one initialised with real world analysed ocean-land-atmosphere states and current CO2 concentration and another with counterfactual 'natural' world ocean-land-atmosphere states and the CO2 concentration level of 1960. The attribution forecast results suggest that global warming associated with increased CO2 is a major factor in causing the heat events. Despite the warm ocean state in 2014 and strong El Niño conditions in 2015 further sensitivity experiments suggest that the atmospheric circulation anomalies played a more important role than the direct impact from the ocean in promoting extreme heat across Australia.

### Impact of the QBO on Predictability of the MJO

Harry Hendon
Bureau of Meteorology
Wednesday, 27th Jul 2016
10:00 AM - 11:00 AM

Abstract: The MJO during boreal winter is observed to be significantly stronger during the easterly phase of the QBO than during the westerly phase for the period 1982-2014. Using 33 years of hindcasts from the POAMA coupled model forecast system and other available hindcasts from the S2S archive, we show that this strengthened MJO activity during the easterly QBO phase translates to improved prediction of the MJO by up to 9 days as depicted by the RMM indices. Forecast skill is higher for events of similar initial magnitude in QBOE years compared to QBOW years, thus ruling out higher skill in QBOE years simply because the MJO was overall stronger.

Although it is as yet unclear as to whether there is an impact of the QBO on the MJO during the forecasts, a simple forecast sensitivity experiment suggests that the QBO-easterly phase favors an eastward expanded Indo-Pacific convective region. Thus, stronger, longer lasting MJO's should be favoured in QBOE years due to an expanded region of warm pool convection. However, the mechanism of impact of the QBO on the MJO, including the seasonality of the impact, is as yet established.

### NESP Clean Air and Urban Landscapes Hub - Introduction and update

Prof. Peter Rayner and
Dr Cathy Oke
University of Melbourne
Monday, 25th Jul 2016
11:00 AM - 12:00 NOON

Abstract: The Clean Air and Urban Landscapes Hub (CAUL) is part of the National Environmental Research Program. NESP is intended "to improve our understanding of Australia's environment through collaborative research that delivers accessible results and informs decision making". CAUL's research is focussed on environmental quality in cities. This talk from Hub Leader Professor Peter Rayner and Knowledge Broker Dr Cathy Oke will give an overview of CAUL's programme and progress and open a discussion of its place in the larger research landscape. See www.nespurban.edu.au

### No sea ice, lots of sea ice and why the ENSO/sea ice link is tricky (and some other stuff such as trends in SST)

Phillip Reid
Bureau of Meteorology
Wednesday, 20th Jul 2016
10:00 AM - 11:00 AM

Abstract: Here we continue our examination of Antarctic sea ice, beginning with an investigation of the "no sea ice" case and why it is important. We then tie in some of our previous seminars on ENSO events and sea-ice extent, showing that ENSO, and in particular El Nino, is a major disruptor of sea-ice seasonality processes. We look at a few recently published papers on sea-ice trends and drivers. Sea surface temperatures get mentioned, yet again.

### A High Resolution NWP Modelling Study of Fog at Perth Airport

Belinda Roux
Bureau of Meteorology
Monday, 11th Jul 2016
02:00 PM - 03:00 PM

Abstract: Fog has a major impact on airline operations in terms of safety and economics yet reliable forecasting of fog remains a significant challenge. This is a particular issue at Perth Airport in the southwest of Western Australia as the airport is remote and the nearest airport suitable for large aircraft is more than two hours flying time. This study aims to examine the performance of the Bureau of Meteorology's soon to be operational city scale model (ACCESS-C2) for providing guidance on the physical processes associated with fog at Perth Airport.

Case studies have been conducted of a number of fog events at Perth Airport. Analysis of the model output concentrates on how well the model represents the mesoscale dynamics, which have an impact on the airflow, temperature and moisture and hence the development of fog. ACCESS-C2 generally captures the observed mesoscale circulations and temperatures in the domain around Perth Airport. The model output is sensitive to the representation of land-surface characteristics and topography. The results presented here suggest the Darling escarpment east of Perth Airport has a significant impact on the occurrence of fog at the airport. Observed variations in the topography contribute to complex local circulations which may affect the location and timing of fog. A number of issues have been identified that need to be addressed to ensure the ACCESS model will provide better forecasts of fog onset and clearance.

### Expansion of the Southern Hemisphere Hadley Circulation

Hanh Nguyen
Bureau of Meteorology
Monday, 6th Jul 2016
10:00 AM - 11:00 AM

Abstract: The Hadley circulation (HC) is a thermally driven circulation that features ascent of equatorial air to a height of about 15km, transport aloft toward the poles, descent at the subtropics, and a return flow near the surface. These features offer an explanation for the persistence and extent of the trade winds and the subtropical high pressure belt that dominate the climate of the tropics and subtropics.

Over recent years, growing interest has been focused on the expansion and strengthening of the HC in the context of a changing climate, with potentially important implications for the surrounding climatic zones. As the Victorian Climate Initiative (VicCI) is coming to an end this EOFY, I will give an overview of my advancements on understanding the mechanisms behind changes in the HC in the Southern Hemisphere under the SEACI then VicCI flags.

### Why is the tropical cyclone boundary layer not well mixed?

Jeff Kepert,
Juliane Schwendike and
Hamish Ramsay
Bureau of Meteorology
Wednesday, 29th Jun 2016
10:00 AM - 11:00 AM

Abstract: Plausible diagnostics for the top of the tropical cyclone boundary layer include (i) the top of the layer of strong frictional inflow and (ii) the top of the "well-mixed" layer; that is, the layer over which potential temperature $\theta$ is approximately constant. Observations show that these two candidate definitions give markedly different results in practice, with the inflow layer being roughly twice the depth of the layer of nearly constant '$\theta$'. Here, we will present an analysis of the thermodynamics of the tropical cyclone boundary layer derived from an axisymmetric model. We show that the marked dry static stability in the upper part of the inflow layer is due largely to diabatic effects. The radial wind varies strongly with height, and therefore so does radial advection of '$\theta$'.

This process also stabilizes the boundary layer, but to a lesser degree than diabatic effects. We also show that this differential vertical advection contributes to the observed superadiabatic layer adjacent to the ocean surface, where the vertical gradient of the radial wind is reversed, but that the main cause of this unstable layer is heating from turbulent dissipation. The top of the "well-mixed" layer is thus distinct from the top of the boundary layer in tropical cyclones. The top of the inflow layer is a better proxy for the top of the boundary layer, but is not without limitations. These results may have implications for boundary-layer parameterisations that diagnose the boundary layer depth from thermodynamic, or partly thermodynamic, criteria.

### Mitigating urban heat during heatwaves to improve human thermal comfort

Stephanie Jacobs
Monash University
Monday, 29th Jun 2016
10:00 AM - 11:00 AM

Abstract: Heatwaves cause a higher number of Australia-wide fatalities than all other natural disasters combined. In some cities, heatwaves have also been shown to exacerbate the urban heat island putting urban communities at a greater risk of heat stress during extreme heat events. Considering that 90% of Australia's population lives in urban areas, mitigating urban heat during heatwaves is essential for saving lives.

We use the Weather Research Forecasting model coupled to the Princeton Urban Canopy Model to simulate Melbourne's deadliest heatwave, Jan 28-30 2009. To improve simulation accuracy, we apply the AWAP soil moisture data to the regional simulations. The implementation of more realistic soil moisture compared to traditional reanalysis products considerably improves the simulation of the maximum temperatures of the heatwave. We then investigate the temperature response to two urban heat mitigation strategies: the implementation of white roofs across the city and irrigation. Preliminary results highlight that the cooling capacity of each mitigation strategy is different, but that the implementation of white roofs can lead to a reduction in 2m temperature and apparent temperature at the surface.

### The next generation of national climate projections in Australia

Michael Grose
CSIRO
Friday, 17th Jun 2016
10:00 AM - 11:00 AM

Abstract: www.climatechangeinaustralia.gov.au was released in 2015, so now we have an opportunity to take stock of what has been done in the climate projections area in the past and what could be done better in the next generation. We are facing an ever-increasing need for reliable climate change information in a form that can be used in decisions and planning. Climate services and products for Australia can be enhanced according to three measures: credibility, salience and legitimacy. When assessing the timing and content of future national climate projections, the next IPCC assessment reports and CMIP6 will provide a strong user 'pull'. However, when assessing the most prudent actions and investment, we should also take a critical look at progress in scientific understanding, opportunities to enhance the communication of projections, and the socio-political landscape. In this talk I'll discuss climate projections, list some lessons from the last 30 years, and run through some of the issues we have identified about the future. I'm keen to gather further input and links for a report on the topic, which aims to provide a useful account and resource for activities in this area over the next decade. I'm keen to listen and gather ideas on this topic, to make the report a useful shared resource.

Possible future issues include:
- 'Game changers', such as multi-year to decadal prediction;
- Scientific advances in constraining ranges of change, understanding future changes to important phenomena and many more;
- A changing socio-political landscape where projections may gain new legitimacy (raising legal and liability issues);
- New users and new uses for climate change projections;
- Improvements in making projections locally-relevant;
- New communication methods, including a stronger link to decision-making frameworks;
- Areas of special interest, including urban, coastal and agricultural issues;
- New lessons from national-level projects overseas;
- Large practical issues, such as 'big data'

### A new classification of storms and associated extrema (Topic 1, from 10:00 am); Pyrocumulonimbus lightning and fire ignition on Black Saturday (Topic 2, from 10:20 am)

Andrew Dowdy
Bureau of Meteorology
Wednesday, 15th Jun 2016
10:00 AM - 11:00 AM

Abstract Topic 1 (10:00 am): Cyclone, front and thunderstorm datasets are examined in combination with each other, allowing a range of different storm classification types to be examined. Of the various storm types examined, the highest risk of extreme precipitation and extreme wind speeds is associated with a triple storm type characterised by the simultaneous occurrence of a cyclone, front and thunderstorm at the same location. A number of notable historical events, such as the Pasha Bulker east coast low of June 2007, are found to be consistent with this triple storm type. The storm classification method is applied to produce maps of the primary causes of extreme precipitation and extreme wind speed, based on an eleven-year period of available data. The physical properties of the storm types are examined at various levels throughout the troposphere.

Abstract Topic 2 (10:20 am): The atmospheric response to the Black Saturday fires is investigated with a focus on convective processes associated with the fire activity. In particular, it is demonstrated that multiple fire plumes generated a number of distinct pyrocumulonimbus (pyroCb) clusters and that there was a large amount of lightning activity within these pyroCbs. The timing of this lightning activity is found to offer broad implications for constraining fire behavior exacerbation, given that deep convection can produce extreme and chaotic variations in surface wind conditions near the fire. An observed fire ignition late on Black Saturday is examined in relation to the pyrogenic lightning, finding that the pyroCbs produced relatively little rainfall on the ground which exacerbated the risk of a sustained fire ignition from dry lightning (i.e., lightning that occurs with relatively little accompanying precipitation). The findings are intended to produce a greater understanding of pyroCb events and associated fire-atmosphere interactions to help enhance fire response capabilities, including through improved forecast guidance for the potential of dangerous fire weather conditions associated with convective systems, as well as in relation to the risk of new fire ignitions from pyrogenic lightning.

### Informing tropical cumulus parameterization development using long-term radar observations in Darwin

Alain Protat
Bureau of Meteorology
Wednesday, 8th Jun 2016
10:00 AM - 11:00 AM

Abstract: The representation of convective cloud properties in global and high-resolution models is an essential step towards accurate quantitative precipitation forecasts. It has been recently demonstrated that global models tend to produce a reasonably good estimate of the rainfall accumulation, but for wrong reasons (the "raining too little too often" problem). This calls for a better understanding of convective cloud processes and their variability as a function of the large-scale environment. In this talk I will review results obtained on the characterization of tropical convective properties using a research C-band dual-polarization radar (CPOL) located around Darwin, Australia.

This talk will highlight (i) the variability of the statistical properties of convective clouds as a function of the large-scale atmospheric "regime", (ii) the composite life cycle of the cumulus cloud modes and their microphysical properties, (iii) the diurnal cycle, (iv) the mechanisms involved in the transition from shallow to deep convection, with some focus on the role of the congestus stage, and (v) the convective mass flux, its components, and its variability. We'll finally show how well those processes are currently reproduced by the model and what plans we have to improve the representation of convection in global models.

### Assessing the characteristics of rapidly weakening tropical cyclones

Kimberly M. Wood
Mississippi State University
Monday, 30th May 2016
09:30 AM - 10:30 AM

Abstract: Rapid intensity changes of tropical cyclones (TCs), where intensity refers to the 1-minute sustained 10-m wind speed, strongly contribute to overall intensity forecast errors. Since appropriate preparations by marine interests and coastal communities depend on accurate forecasts of intensity, much work has been done to improve predictability of rapid intensification (RI; defined as a 30 kt [15.4 m s-1] increase in 24 h).

However, far less research has been done on the rapid weakening (RW) of tropical cyclones while they remain over ocean even though RW has been observed near land. Using National Hurricane Center (NHC) best track data from the North Atlantic and eastern North Pacific basins, RW is defined as a 30 kt decrease of intensity in 24 h. Both official NHC forecasts and the Statistical Hurricane Intensity Prediction Scheme (SHIPS) model generally overestimate the intensity of TCs undergoing RW. Reanalysis data reveal the dominant contributors to RW tend to be decreasing sea surface temperatures, decreasing mid-level moisture, and increasing vertical wind shear. Ongoing work is investigating the spatial component of these negative factors as a means to reduce future forecast errors for RW events. This presentation will also discuss the utility of satellite data in identifying RW as well as the frequency and characteristics of RW in other basins, particularly near Australia.

### The Global Climate Observing System (GCOS) Network Manager Providing an end to end network support service at a global scale

Tim Oakley
WMO
Tuesday, 3rd May 2016
11:00 AM - 12:00 NOON

Abstract: GCOS is a long-term, user-driven operational system capable of providing the comprehensive observations required for monitoring the climate system and is sponsored by three UN bodies; WMO, IOC of UNESCO and UNEP, and by the International Council for Sciences (ICSU). This multiple sponsorship guarantees that we cover the broad range of climate information required on a global scale and that our work is supported by an international community.

In order to understand where our climate is headed, we need to know where it has been in the past and where it is now and our observational networks are fundamental in helping to answer these questions. As an international network manager I link the funding from our sponsors to the GCOS priority areas and provide technical support across the observational chain. GCOS does not own any of the observing equipment but through the GCOS Cooperation Mechanism (GCM) it aims to support National services and institutes, in the design, installation and operational management of their systems. This presentation will provide an update on the status of GCOS, and the future priorities which will be included in its new Implementation plan (2017 - 2021), in particular those relevant to the WMO Regional Association V (South West Pacific).

### Soil dryness in fire danger rating: Time for a change ?

Vinod Kumar and
Bureau of Meteorology
Wednesday, 27th Apr 2016
10:00 AM - 11:00 AM

Abstract: The fuel availability estimates in McArthur Forest Fire Danger Index used in Australia for issuing fire warnings is based on soil moisture deficit, calculated as either the KeetchByram Drought Index (KBDI) or Mount's Soil Dryness Index (MSDI). These indices are essentially simplified, empirical water balance models designed to estimate soil moisture depletion in the duff layer and root-zone. In this study, through validation against observations, we highlight certain limitations of these indices in estimating soil moisture at different layers it is designed for. These two indices are calculated at ~ 5km resolution using the AWAP data for whole Australia. We also verify soil moisture from a multi-layer land surface model (LSM), concurrently with the traditional indices. Past studies have suggested that a multiple layer soil moisture model is more appropriate to predict fire potential.

The LSM soil moisture dataset used for this study is obtained from the Australian Bureau of Meteorology's operational global numerical weather prediction (NWP) system called ACCESS. ACCESS has undergone an update from its initial release during the study period (2009-2014), and hence both soil moisture products available at 80 and 40 km resolution are used. The observations used for this study are from the OzNet, OzFlux and CosmOz networks. The verification results show that both KBDI and MSDI are poor indicators of shallow layer (~ 030 cm) soil moisture, and the NWP model skills are better. We also show some preliminary work that is carried toward the verification of deep layer soil moisture. The verifications for deeper soil are restricted to OzNet and OzFlux, as only these two networks provide observations matching the maximum soil profile defined by KBDI/MSDI (depth ~ 1m). We highlight some of the quality control issues and limitations associated with these deep soil moisture observations for verification purpose.

A prototype high resolution soil moisture analysis system is presented that uses an offline land surface model that can be driven by data from many sources; such as surface observations of rainfall, temperature, dew-point temperature, wind speed, surface pressure as well as satellite derived measurements of rainfall, surface soil moisture, downward surface shortwave radiation, skin temperature, leaf area index and tree heights. The system is computationally efficient and can be run at 5km resolution over the whole of Australia. The system uses rainfall analyses from AWAP together with temperature, humidity and surface pressure analyses from MSAS. The analysis system estimates soil moisture on four soil layers over the top 3 meters of soil, the surface layer has a thickness of 10 cm. The system takes into account the effect of different vegetation types, root depth, stomatal resistance and spatially varying soil texture. The analysis system has a one hour time-step with daily updating. Verification against ground based soil moisture observations from the OzNet and CosmOz are presented.

### The effect of turbulent plume dynamics on ember transport

William Thurston
Bureau of Meteorology
Wednesday, 6th Apr 2016
10:00 AM - 11:00 AM

Abstract: Spot fires occur when embers are launched by bushfire plumes into the background wind, which then carries the embers a significant distance from the fire front. If the embers land in a suitable fuel bed and are still burning a spot fire may be ignited. Spotting is a hazardous phenomenon because it leads to unpredictable fire behaviour and accelerated fire spread. Therefore a better understanding of the processes that contribute to long-range spotting is essential for the prediction of fire spread. Here we aim to assess the contribution of turbulent plume dynamics to the process of long-range spotting.

We use a two-stage modelling approach to calculate the landing positions of potential firebrands launched by bushfire plumes. Firstly, we use the UK Met Office large-eddy model to perform numerical simulations of idealised bushfire plumes under a range of background wind speeds. Secondly, the three-dimensional, time-varying velocity fields produced by the LEM are used to drive a Lagrangian particle-transport model, which calculates the path of more than 1.5 million potential firebrands for each plume. In order to assess the contribution of the in-plume turbulence to the firebrand transport, the time-varying particle-transport calculations are then repeated using a steady-state plume velocity, calculated from the one-hour mean plume fields.

Plumes in weak winds contain a counter-rotating vortex pair, which leads to large lateral spread in firebrand landing position. Plumes in strong winds are more turbulent and bent over, leading to more longitudinal spread in firebrand landing position and a greater maximum spotting distance. In-plume turbulence is shown to substantially increase the lateral and longitudinal spread in firebrand landing position, and in the case of plumes in strong background winds increase the maximum spotting distance by a factor of two. Systematic studies such as this will inform the development of improved physically based spotting models.

### In-flight detection of volcanic ash - The Airborne Volcanic Object Imaging Detector (AVOID)

Dr Fred Prata
Nicarnica Aviation, Norway
Monday, 7th Mar 2016
02:00 PM - 03:00 PM

Abstract: Volcanic ash is a known hazard to aviation and has caused numerous unforeseen commercial aircraft encounters in recent and past years. Much effort has been placed on detection of volcanic ash using satellite instruments and forecasting skill has improved greatly since the eruption of Eyjafjallajökull, Iceland in Apr/May 2010. Closer to home, eruptions of Merapi (2010), Puyehue-Cordon Caulle (2011), Kelut (2014) and Rinjani (2015) all caused aviation problems of varying severity. It seems clear that an in-flight capability for detection of volcanic ash from commercial aircraft would be of some value and also complement existing satellite techniques as well as provide useful input data to ash forecasting models.

In this talk I will briefly outline the background to the development of an in-flight infrared imaging system capable of detecting ash from distances of up to 100 km a development that started in Australia. A unique proof-of-concept experiment demonstrating detection from an AIRBUS A340 test aircraft was successfully completed over the Bay of Biscay, in late 2013. An artificial ash cloud was generated for the experiment using an AIRBUS A400M tanker aircraft. Results from this experiment will be presented. The ash detector is now being certified and installed onto two easyJet A320 aircraft and the first commercial trials are due to take place in late 2016.

### The new Met Office science strategy, science partnerships and convective scale modelling

Jon Petch
UK Met Office
Monday, 29th Feb 2016
10:30 AM - 11:30 AM

Abstract: The Met Office has recently released a new science strategy for 2016-2021. I will give an overview of the direction of travel, highlight the current status of science partnerships and talk about work to better organise the development of the convective scale atmospheric model. This will include a brief summary of outcomes on the convective scale workshop held in Feb in Singapore, current issues and recent model improvements.

### Exploratory Modelling of Transformative Change in Societal Systems of Service Provision

Dr. Fjalar de Haan
University of Melbourne
Wednesday, 24th Feb 2016
10:00 AM - 11:00 AM

Abstract: Service-provision systems like energy supply, water management and health care have social, institutional and often infrastructural aspects. This makes modelling these systems an interdisciplinary affair with unique challenges. A key challenge amongst those is the deep uncertainties involved - be they related to complexity or human agency.

This talk will present a model to simulate transformative change in service-provision systems. The particular model set-up and results discussed focus on the uptake of green water servicing solutions, though the framework is more generally applicable. Much attention will be devoted to the application of methods of exploratory modelling to deal with deep uncertainties including massive scenario generation and cluster analysis.

### Developments in volcanic dispersion modelling: the legacy of Eyjafjallajokull

Claire Witham
UK Met Office
Tuesday, 23rd Feb 2016
10:00 AM - 11:00 AM

Abstract: The eruption of Eyjafjallajokull in 2010 was a pivotal point for the London VAAC and in the development of modelling and forecasting volcanic plumes. In this seminar I will take a look at the many advances in dispersion modelling that have been made by the Met Office and others over the following years, both for volcanic ash and for volcanic gases.

It is important to recall that the models themselves are only one component, with the source information and the meteorological input also being key sources of uncertainty. Progress has been made in all these areas, for example in modelling ash particle properties, the use of plume dynamics models to better constrain mass eruption rates and in the improvement of NWP resolutions. I will present both the gains and limitations in each of these areas and look at the challenges for increasing sophistication in the forecast environment.

### An Observational Study on the Evolution Process of the Positive and Negative Indian Ocean Dipole

Dr. Takanori Horii
Japan Agency for Marine-Earth Science and Technology (JAMSTEC)
Wednesday, 17th Feb 2016
10:00 AM - 11:00 AM

Abstract: The Indian Ocean dipole (IOD) is a seasonal to interannual ocean-atmosphere phenomenon occurring in the tropical Indian Ocean. In this presentation, I will talk about the different evolution process of positive and negative IOD based on observational data. We mainly used moored-buoy data in the eastern Indian Ocean to clarify the processes that produced anomalous sea surface temperature (SST) in the eastern Indian Ocean associated with IOD events. Heat balance analysis demonstrated that zonal heat advection mainly controlled the mixed layer temperature variation during positive IOD events. In contrast, in the case of warm SST anomalies in the eastern Indian Ocean during negative IOD, air-sea heat fluxes and horizontal heat advections mainly account for the mixed layer temperature variation. Here, reduced latent heat loss had a major role in producing the warm SST anomalies. These results suggest that the IOD includes different feedback processes in its positive and negative phases. I will also introduce our recent study on intraseasonal coastal upwelling signal along the southern coast of Java, which may play a key role in IOD, observed by Indonesian tidal station data.

### Predicting rainfall change under global warming: Emerging opportunity for atmosphere-ocean dynamics

Prof. Shang-Ping Xie
Scripps Institution of Oceanography
Wednesday, 10th Feb 2016
10:00 AM - 11:00 AM

Abstract: Global mean temperature has risen for the past century and is projected to rise even more in response to the increasing atmospheric CO2 concentrations. Precipitation change is of vital importance to societies but precipitation projections are intrinsically challenging as they change sign from one region to another.

Recent studies show that in the tropics, radiatively forced changes in precipitation and atmospheric circulation are tightly coupled to spatial patterns of ocean surface warming. Mechanisms identified for ocean warming pattern formation include those important in natural variability such as Bjerknes and WES feedbacks but there are also mechanisms unique to climate change such as the dynamic thermostat due to the equatorial upwelling. Developing predictive understanding of ocean-atmosphere coupling holds the key to building confidence in regional climate projections, much as in the historical development of El Nino prediction. The talk examines the interactions between radiatively forced changes in the ocean and atmosphere, and discusses the implications for regional climate change.

### 2011-15 - the world's hottest five-year period on record

Blair Trewin
Bureau of Meteorology
Wednesday, 3rd Feb 2016
10:00 AM - 11:00 AM

Abstract: 2011-15 was the world's hottest five-year period on record, with temperatures 0.57 C above the 1961-90 reference period, and included 2015, the world's hottest year on record, with global temperatures returning to within the envelope predicted by climate models after having been near or below the bottom of that envelope during the La Nina-dominated period between 2008 and 2012.

In this seminar, I will discuss significant features of the world's climate over the last five years, in addition to the high temperatures. This includes significant multi-year droughts in Brazil, the western United States, eastern Australia, and southern Africa, as well as major floods, heatwaves and tropical cyclones, and the extent to which the risk of these events has (or has not) been altered as a result of anthropogenic climate change. An assessment of the cryosphere in both the Arctic and Antarctic, and the current status of global sea levels, will also be included.

### Recent Results on Hurricanes and Climate

Suzana Camargo
Columbia University
Friday, 29th Jan 2016

Abstract: In this talk I will discuss recent results from 2 different studies on hurricanes and climate. In the first one, we analyzed tropical cyclone genesis indices (TCGIs), which are functions of the large-scale environment that are designed to be proxies for the probability of tropical cyclone (TC) genesis. While the performance of TCGIs in the current climate can be assessed by direct comparison to TC observations, their ability to represent future TC activity based on projections of the large-scale environment cannot. We examined the performance of TCGIs in high-resolution atmospheric model simulations forced with sea surface temperatures (SST) of future, warmer climate scenarios, using a perfect model framework. We investigated whether the TCGIs derived for the present climate can, when computed from large-scale fields taken from future climate simulations, capture the simulated global mean decreases in TC frequency projected by this model. The TCGIs differ in their choice of environmental predictors, and several choices of predictors perform well in the present climate. However, some TCGIs that perform well in the present climate do not accurately reproduce the simulated future decrease in TC frequency.

In a second study, the impact of the Montreal Protocol on the potential intensity of tropical cyclones over the next 50 years is investigated with the Whole Atmosphere Community Climate Model (WACCM), a state-of-the-art, stratosphere-resolving atmospheric model, coupled to land, ocean, and sea-ice components, and with interactive stratospheric chemistry. An ensemble of WACCM runs from 2006 to 2065 forced with a standard future scenario is compared to a second ensemble in which ozone depleting substances are not regulated (the so-called `World Avoided'). It is found that by the year 2065, changes in the potential intensity of tropical cyclones in the World Avoided are nearly three times as large as for standard scenario. The Montreal Protocol thus provides a strong mitigation of the adverse effects of intensifying tropical cyclones. The relative importance of warmer sea surface temperatures (ozone depleting substances are important greenhouse gases) and cooler lower stratospheric temperatures (accompanying the massive destruction on the ozone layer) are carefully examined.

### The Ocean Rain And Ice-phase precipitation measurement NetworkOceanRAIN and the installation onboard RV Investigator

Christian Klepp
CliSAP/CEN, University of Hamburg
Wednesday, 20th Jan 2016
10:00 AM - 11:00 AM

Abstract: Systematic high quality oceanic in-situ precipitation measurements of rain, snow and mixed-phase are requested on an international science level and are essential for improved understanding and validation of hydrological processes in satellite, re-analysis and model data. OceanRAIN, the shipboard "Ocean Rainfall And Ice-phase precipitation measurement Network" for surface validation is to date the only systematic long-term optical disdrometer-based oceanic shipboard precipitation data collection effort to establish a comprehensive statistical basis of precipitation for all climate related hotspots over the global oceans. The selected research ships do not circumvent high impact weather, allowing for a collection of the full precipitation spectrum including extremes. The fast growing OceanRAIN data base comprises to date more than 6 million minutes of precipitation measurements from 11 ships since the project start in 2010. Since Jan 2016 the Australian RV Investigator is equipped with the OceanRAIN instrumentation within the CAPRICORN project.

OceanRAIN combines different shipboard data streams of navigational data, standard meteorological measurements and the automated optical disdrometer systems of type ODM470. OceanRAIN samples precipitation occurrence, intensity and accumulation discriminated for rain, snow and mixed-phase precipitation. The ODM470 was especially designed for shipboard operation under high and frequently varying wind speeds and rough sea states. Through minute-based particle size distributions, a rain and snowfall (including mixed-phase) retrieval calculates the rate given the precipitation phase. The precipitation phase in turn is estimated by an automatic precipitation phase distinction algorithm using a combination of air temperature, relative humidity and particle size. This presentation describes the instruments, algorithms, methodology, and data products of OceanRAIN. The procedure of the data processing chain is outlined, including calibration, shipboard operation, data acquisition and quality control. Applications are presented for regional subsets of all climatic regions including the cold-season high-latitudes.

OceanRAIN aims at increasing knowledge about oceanic precipitation and improving error characterization of GPM (Global Precipitation Measurement) era precipitation retrievals, adding to the continual improvement of the satellite retrieval algorithms. Furthermore, benchmarking of existing satellite-based climatologies, re-analysis and model data is envisaged. The growing data base can be utilized for statistical and process study applications on different temporal and spatial scales, microphysical studies of rain and snow formation, and allows insight to the point-to-area problem of precipitation.