ESCI Technical Report - Standardised Method - 1
ESCI Technical Report - Standardised Method - 1
ESCI Technical Report - Standardised Method - 1
Andrew Dowdy1, Andrew Brown1, Acacia Pepler1, Marcus Thatcher2, Tony Rafter2, Jason
Evans3, Hua Ye1, Chun-Hsu Su1, Samuel Bell1 and Christian Stassen1
1
Bureau of Meteorology, Docklands, Australia
2
CSIRO Oceans and Atmosphere, Aspendale, Australia
3
University of New South Wales, Sydney, Australia
July 2021
Andrew Dowdy1, Andrew Brown1, Acacia Pepler1, Marcus Thatcher2, Tony Rafter2,
Jason Evans3, Hua Ye1, Chun-Hsu Su1, Samuel Bell1 and Christian Stassen1
Bureau of Meteorology, Docklands, Australia
1
3
University of New South Wales, Sydney, Australia
July 2021
Authors: Andrew Dowdy, Andrew Brown, Acacia Pepler, Marcus Thatcher, Tony Rafter, Jason Evans,
Hua Ye, Chun-Hsu Su, Samuel Bell and Christian Stassen
Title: Extreme temperature, wind and bushfire weather projections using a standardised method
ISBN: 978-1-925738-32-2
ISSN: 2206-3366
i
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Bureau of Meteorology
GPO Box 1289, Melbourne
Victoria 3001, Australia
andrew.dowdy@bom.gov.au
© 2021 Bureau of Meteorology. To the extent permitted by law, all rights are reserved and no part of
this publication covered by copyright may be reproduced or copied in any form or by any means except
with the written permission of the Bureau of Meteorology.
The Bureau of Meteorology advise that the information contained in this publication comprises general
statements based on scientific research. The reader is advised and needs to be aware that such
information may be incomplete or unable to be used in any specific situation. No reliance or actions
must therefore be made on that information without seeking prior expert professional, scientific and
technical advice. To the extent permitted by law and the Bureau of Meteorology (including each of its
employees and consultants) excludes all liability to any person for any consequences, including but not
limited to all losses, damages, costs, expenses and any other compensation, arising directly or indirectly
from using this publication (in part or in whole) and any information or material contained in it.
ii
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Contents
Abstract ...................................................................................................................... 1
1. Introduction and overview ................................................................................ 2
2. Methodology ...................................................................................................... 4
2.1 Standardised method for projections information ..................................................... 4
2.2 Examples of method outputs being used ................................................................. 6
3. Extreme temperature projections .................................................................... 8
3.1 Introduction ............................................................................................................... 8
3.2 Summaries for physical processes ........................................................................... 8
3.3 Summaries for historical climate ............................................................................. 15
3.4 Summaries for future climate .................................................................................. 15
3.5 Lines of evidence table ........................................................................................... 17
3.6 Projections and confidence information .................................................................. 18
4. Extreme wind projections ............................................................................... 20
4.1 Introduction ............................................................................................................. 20
4.2 Summaries for physical processes ......................................................................... 20
4.3 Summaries for historical climate ............................................................................. 25
4.4 Summaries for future climate .................................................................................. 27
4.5 Lines of evidence table ........................................................................................... 31
4.6 Projections and confidence information .................................................................. 32
5. Extreme fire weather projections ................................................................... 34
5.1 Introduction ............................................................................................................. 34
5.2 Summaries for physical processes ......................................................................... 35
5.3 Summaries for historical climate ............................................................................. 40
5.4 Summaries for future climate .................................................................................. 41
5.5 Lines of evidence table ........................................................................................... 42
5.6 Projections and confidence information .................................................................. 43
6. Conclusion ...................................................................................................... 45
Acknowledgements.................................................................................................. 45
References ................................................................................................................ 45
iii
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
ABSTRACT
The influence of anthropogenic climate change on extreme temperatures, winds and bushfire
weather in Australia is assessed here using a standardised method for projections information.
These assessments consider a comprehensive range of factors based on observations, modelling
and physical process understanding. Those factors are reviewed using a standardised method to
collate lines of evidence and then guide the production of projections data and confidence
assessments. Projections are produced based on global climate model data as well as dynamical
downscaling data using three regional climate modelling approaches (CCAM, BARPA and
NARCliM/WRF), with environmental diagnostics also used for severe convective winds from
thunderstorms. The projections data are all calibrated using quantile matching methods trained
on observations-based data, with a particular focus on the accurate representation of extremes.
The resultant projections data include nationally consistent maps corresponding to the 10-year
average recurrence interval (i.e., return period) around the middle of this century, with a focus of
the discussion on regions around southern and eastern Australia during summer as needed for
some risk assessment applications. The projections data are also available for other seasons and
time periods throughout this century, as well as for other metrics of extreme or average conditions.
The results for southern and eastern Australia during summer show more extreme temperatures
(very high confidence), more severe winds (low confidence) and more dangerous bushfire
conditions (high confidence in southern Australia; medium confidence in eastern Australia)
attributable to increasing greenhouse gas emissions.
1
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
2
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
exceedance). This included maps of the most likely future projected change in values
corresponding to the 10-year ARI, together with estimates of the 10th and 90th percentile range of
plausible 10-year ARI values as a confidence assessment measure. National maps of those
quantities are presented here, together with confidence assessment information, based on
considering various lines of evidence. The resultant maps and data layers (with supporting
confidence assessment information) are intended for a broad range of user groups including in
sectors for which extreme temperatures, winds or bushfires are relevant. The projections are
presented for the future climate around the middle of this century, as well as for the historical
climate, but are also available for other time periods throughout this century as well as for other
metrics of extreme or average conditions throughout Australia (with data available on request).
The following Section 2 describes the steps for applying the standardised method, then
provides some examples of practical uses of the outputs for improved resilience to climate
hazards. Sections 3-5 document the application of the method for extreme temperature, wind and
fire weather, respectively.
3
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
2. METHODOLOGY
4
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
For quantities that have a reasonably robust range of evidence, with good agreement between
those different lines of evidence (e.g., about two thirds of the Lines of Evidence Table having
a consistent sign of future change), then model output may be the best option for producing
the Projections Likelihood Information, while still considering the various uncertainties and
strengths/weaknesses of different modelling approaches for helping to guide the production
of the data products. For quantities with lower confidence (i.e., more limited evidence and/or
lower agreement between lines of evidence), then a more qualitative best estimate could be
appropriate. For example, in some cases with very high uncertainty the best estimate for the
Projections Likelihood Information might simply be ‘an increase is more likely than a
decrease’ for a particular region, if that is the best information that can be provided based on
the balance of available knowledge from the Lines of Evidence Table. It is acknowledged
that given the broad range of different information sources and data types (e.g., direct model
output or statistical diagnostic methods) this step of the process may require some degree of
expert judgement to be used.
The Projection Likelihood Information can include confidence assessment information, such
as based on the degree of evidence and agreement from the Lines of Evidence table. For
example, estimates of the 10th and 90th percentile range of plausible change is one measure
that could be used to help indicate the degree of confidence in a projected future change, as
well as noting various other approaches that could be used for some applications, including
the framework shown in Table 2.1 together with various words that have a range of
quantitative probabilities associated with them to accompany the provision of projections.
5
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
For this study, the method is applied for the projected change in climate from the time
period 1986–2005 (i.e., a commonly used historical reference period for CMIP5 data (CSIRO &
BoM 2015)) to the time period 2040–2059 (i.e., a time period centred on the middle of this century
around 2050 as requested by energy sector stakeholders). The information collected here for Step
1 for the Lines of Evidence Table is intended to be relevant for the National Energy Market
(NEM) region around southeast Australia, including listing any regional variations that might be
important to consider.
The RCP8.5 scenario, representing a high emissions pathway for anthropogenic
greenhouse gases, is used for the future projections for a number of reasons. Of the set of modelled
greenhouse gas emission pathways provided in CMIP5 (which start to deviate from each other
after 2005), observed climate change trends for temperature indicate that the high emissions
pathway RCP8.5 has been followed more closely than low emissions pathways (e.g., RCP2.6)
(IPCC 2013; Schwalm et al. 2020). Additionally, although there is potential for reductions in
greenhouse gas emissions and the associated rate of temperature increase later this century,
RCP8.5 is used here for the application of this method given that it takes many years after changes
in emissions for an emergent change in a climate trend, noting the focus for this application on
the period from now until around the middle of this century. However, for applications in which
projections are needed based on lower emissions pathways than RCP8.5, methods could be used
for scaling these projected changes according to the global warming magnitude for a particular
time period or emissions pathway, such as has been recently demonstrated (NESP 2020).
Table 2.1: Confidence can be assessed based on the degree of evidence and agreement,
consistent with IPCC guidelines. The degree of confidence can then be used together with the
projections data to help provide likelihood estimates (i.e., probability of occurrence) consistent
with Mastrandrea et al. (2011).
Limited evidence Medium evidence Robust evidence
6
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
applications in Australia. Another example is the inclusion of the10-yr ARI maps for
temperature in AEMO's 2020 Integrated System Plan (ISP).
Enhanced resilience modelling: Randomised failures are currently used as synthetic input to
energy sector modelling for matching supply and demand, including for assessing future
changes in the resilience of the NEM. As suggested by energy sector groups, the outputs of
the standardised method can be used to refine these failure rates, to help design and plan for
a network that is more resilient to future climate change based on considering a
comprehensive range of evidence.
Enhanced reliability modelling: The outputs can be used for providing guidance to
accompany the projections data provided as input for the NEM reliability modelling,
including insight on whether some datasets might be preferentially weighted over others for
some variables/regions.
Enhanced guidance for stakeholders on climate risk and hazard scenarios including
compound events: The comprehensive review and synthesis framework of the standardised
method is being used to help examine some details for compound event scenarios, intended
for use in subsequent risk assessment applications and 'stress testing' activities on climate
hazards.
Broader applicability: Although the results presented here are primarily intended to meet the
needs of the electricity sector in Australia, they are also intended to have broader benefits
including for other sectors, given the relevance of extreme temperatures, winds and bushfire
conditions to other sectors. These projections and confidence assessments are also providing
a foundation for the initial stages of the Australian Climate Service (ACS) recently
established for producing and providing climate information in Australia.
7
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
3.1 Introduction
The standardised method for projections information is applied here for extreme values
of daily maximum temperature at a height of 2 m during summer, with a focus on the 10-yr ARI
values for regions around southern and eastern Australia. The application of this method follows
the two steps described in Section 2.1.
For Step 1 of the method, short summaries are presented below (not listed in order of
importance) for different aspects relating to future changes in extreme temperature during
summer, with regional variations noted where relevant. The summaries are then used to populate
the Lines of Evidence Table (Table 3.1), with key details from the summaries noted succinctly in
the rows of that table, including the degree of influence that this aspect has on extreme
temperature and its implied direction of projected change (either an increase, decrease, little
change or increased uncertainty).
For Step 2 of the method, the results from the Lines of Evidence Table are used for
guidance in producing the Projections Likelihood Information. For this study, this includes the
best estimate of the most probable projections for extreme temperature (presented here as maps
showing the 10-yr ARI values) as well as estimates of the 10th and 90th percentile range of
plausible change in the 10-yr ARI values (as a measure for indicating the degree of confidence in
the projections). As discussed in Section 2.1, the RCP8.5 emissions pathway from CMIP5 is
considered relevant for use in providing projections towards the middle of this century, with a
focus here on a historical reference period 1986–2005 and a projected future period 2040–2059.
8
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
spring but also summer (CSIRO & BoM 2015; Ukkola et al. 2020), with this higher frequency of
drier soils expected due to higher rates of atmospheric evaporative demand and increased periods
of drought (including meteorological drought defined based on rainfall deficit measures). It is
also noted that there are considerable uncertainties around climate models simulations of how soil
moisture can influence temperature through land-atmosphere coupling processes. For example,
climate models may overestimate the coupling between soil moisture and extreme temperatures
in wet areas of the globe, so potentially overestimate this aspect to some degree relating to
increases in extreme temperatures in some cases (Ukkola et al., 2018). There are also uncertainties
in the influence of climate change on the direction and magnitude of soil moisture change,
including relating to uncertainties in changes to rainfall, potential evaporation and the use of soil
water by vegetation under increasing levels of CO2 (Jovanovic et al. 2008; Ukkola et al. 2020).
In summary, soil moisture can be an important influence on temperature extremes, while
noting some uncertainties in the ability of climate models to simulate some processes that are
relevant for soil moisture. Projections indicate more frequent periods of dry soil moisture on
average in the future during summer in southern and eastern Australia, which will act to increase
the risk of extreme temperatures, with medium confidence.
Subtropical ridge
An intense subtropical ridge (STR) of mean sea-level pressure is associated with an
increase in the mean maximum temperature and the frequency of days above the 90th percentile
in southern Australia in all seasons (Pepler et al. 2018). This relationship is strongest in winter
and spring, including in southern regions such as Victoria and Tasmania. During summer, an
intense STR is associated with more hot days in Tasmania but fewer hot days on the east coast
including Brisbane. Observations and reanalysis data show the STR has grown more intense in
recent decades, which has contributed to observed declines in southeast Australian rainfall
9
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
(Timbal & Drosdowsky, 2013), but it is unknown whether the intensification of the STR has also
contributed to past changes in maximum temperature or hot days.
The STR seasonal cycle is relatively well simulated in CMIP5 and is projected, with high
confidence, to intensify in the future (CSIRO & BoM, 2015). In this regard, CMIP5 models
represent a significant improvement over CMIP3 models. Despite confidence in the projection of
STR intensification, it is uncertain how this future intensification will impact future extreme
temperatures in Australia, while noting that the subtropical ridge is typically associated with
descending air and relatively clear skies (i.e., reduced cloudiness) that could potentially be one
contributing factor for increased temperature extremes. Although CMIP5 models have limited
ability to replicate the STR influence on Australian rainfall (CSIRO & BoM, 2015), the STR
relationships with temperature are mostly independent of the STR-rainfall relationships (Pepler
et al. 2018) and it is a current knowledge gap in the literature as to how well the CMIP5 models
replicate the STR relationship with extreme temperatures.
In the Southern Hemisphere, the STR intensification and other measures of tropical
expansion have been linked to climate change (Nguyen et al. 2015; Grise et al. 2019) with some
contribution from ozone depletion in the summer months as well as natural variability (Garfinkel
et al., 2015; Waugh et al., 2015). Climate models consistently project a future southward shift and
intensification of the subtropical ridge (Kent et al. 2013; Grose et al. 2015). However, this may
be masked by the influence of ozone hole recovery during the summer months in coming decades
to some degree (IPCC 2013).
In summary, the STR has historically had a significant influence on the occurrence of
extreme temperatures, with more intense STR associated with hotter summer temperatures
particularly in southern Australia. Although CMIP5 models do a reasonable job of simulating the
STR, including an increase in intensity being likely in the future, the impact of the STR on future
extreme temperatures is somewhat uncertain. As STR is a large-scale feature with links to
broader-scale processes such as tropical expansion, RCMs may offer relatively limited
improvement over GCMs in representing the STR. However, RCMs may be better able to
simulate the impacts of the STR on local climate extremes, due to better simulation of interactions
between the large scale and local factors such as cloud cover.
Cold Fronts
Frontal systems are major drivers of extreme temperature events in southern Australia.
Strong northwesterly winds prior to cold fronts can enhance the advection of extreme heat from
inland Australia towards the southeast regions during summer. Some studies suggest relatively
little change in the frequency of fronts in southeast Australia and a slight decrease in their mean
intensity over recent decades (Rudeva & Simmonds, 2015), while some studies also indicate the
frequency of fronts has decreased in some regions of southeast Australia such as for the eastern
seaboard (Pepler et al. 2021).
Climate models are generally able to simulate the average annual frequency of fronts in
the Australian region during winter, but relatively few studies have examined this during summer
(Catto et al. 2015; Blázquez & Solman 2017). Climate model projections have a weak increase in
the frequency and intensity of fronts in southern Australia, but the available projections do not
distinguish the cold fronts associated with northwesterly winds from warm fronts and stationary
fronts (Catto et al. 2014; Blázquez & Solman 2019). Using the older CMIP3 climate models, a
simple temperature-based proxy for very extreme cold fronts associated with summer temperature
extremes and bushfires indicated a likely future increase in the frequency of frontal systems
(under both medium and high emissions scenarios), increasing from ~0.5 events per year in the
current climate to 1-2 events per year by the end of the 21st century (Hasson et al. 2009).
Considering studies such as these, considerable uncertainties remain in relation to extreme
10
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
temperature events associated with fronts in the during summer and how these events could
potentially change in the future.
In summary, observations indicate frontal activity has undergone little change in southern
and eastern Australia during summer. A future projected increase appears more likely than a
decrease in the number of fronts that occur in southeast Australia, but there is considerable
uncertainty for future projected changes in frontal systems and their impact on extreme
temperatures based on the current knowledge. Given that fronts are synoptic-scale systems which
GCMs can simulate reasonably well, there may not be a large benefit from using RCMs to
examine future frontal system activity compared to other factors like clouds and solar radiation.
However, RCMs could potentially provide value for some aspects relating to fronts such as their
interaction with terrain and associated extreme weather impacts for localised regions in some
cases.
Tropical cyclones
The occurrence of tropical cyclones (TCs) in northern Australia has been linked with the
intensification of heat extremes in southern Australia, including in southeast Australia during
summer (Parker et al. 2013; Quinting & Reeder 2017; Quinting et al. 2018). For example, the
extreme heat experienced around the time of the Black Saturday fires in 2009, which set new
temperature records for daily maximum air temperature for Melbourne and surrounding locations,
was associated with the presence of a TC (Parker et al., 2013). Observations indicate a decrease
in occurrence frequency of TCs for the Australian region over recent decades (Dowdy 2014;
Chand et al. 2019).
Future projections of TCs during summer for the Australian region indicate a small
decrease in their frequency (medium confidence) (Bell et al. 2019). However, the frequency of
intense category 4 and 5 TCs may not change or increase slightly, along with some poleward
migration (low confidence) (CSIRO & BoM 2015; Knutson et al. 2020; NESP 2020). In general,
GCMs have insufficient spatial and temporal resolution to adequately simulate tropical cyclones.
11
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
RCMs generally have finer resolution and better resolve tropical cyclones, although RCMs still
do not fully capture all relevant processes. For this reason, additional methods for cyclone
projections can also be useful to consider, such as synthetic cyclone tracks, in addition to dynamic
modelling.
12
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Figure 3.1: Correlations between temperature and climate measures. This is presented in the
upper row of panels for daily maximum temperature (using average summer values for the
months December to February: DJF) and measures representing different modes of variability
including ENSO (using the NINO3.4 index), SAM (using the SAM index) and IOD (using the
DMI index). Similar correlations are also shown in the lower row of panels, but for the number
of days with temperature above the 99.5th percentile during summer. These correlations are all
based on the period from 1979 to 2019, using one value for each summer period (DJF).
NINO3.4 and DMI data are attained from the NASA ESRL
(https://psl.noaa.gov/gcos_wgsp/Timeseries/) while SAM data are from
https://legacy.bas.ac.uk/met/gjma/sam.html. Pearson's correlation coefficient, r, is shown with
stippling corresponding to statistically significant values at the 95% confidence level (2-tailed).
13
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Positive SAM is associated with a decreased likelihood of extreme heat during the spring,
but correlations are more mixed during the summer months (Hendon et al. 2007; Marshall et al.
2013; Perkins et al. 2015). The relationship between SAM and average values of daily maximum
temperature during summer is broadly similar in spatial patterns (e.g., sign of correlation, from
Fig. 3.1) to the case for the relationship between SAM and the occurrence of more extreme values
of daily maximum temperature, with generally weak correlations or a negative correlation in
central eastern regions (particularly for mean temperature). A strong negative SAM is also
associated with sudden stratospheric warmings (as occurred in the 2019 Austral spring), which
can cause extreme heat during spring and early summer (Lim et al. 2019), potentially associated
with some of the negative correlations apparent in Fig. 3.1 for the central east region.
SAM has been becoming more positive in recent decades, particularly during the summer
months (Marshall, 2003), which has been linked to a combination of increased greenhouse gases
as well as ozone depletion and natural variability (Garfinkel et al. 2015; Waugh et al. 2015).
CMIP5 models project a robust shift towards more positive values of SAM in all seasons during
the 21st century (Lim et al. 2016), although this may be masked to some degree by the influence
of ozone hole recovery during the summer months in coming decades (Banerjee et al. 2020). In
summary, climate models can simulate SAM well, but projections of a positive trend in SAM
would likely cause little change in the risk of heat extremes during summer apart from potentially
central east (noting a negative correlation with temperature as well as links with sudden
stratospheric warmings (Lim et al. 2019) for which future projected changes are not currently
known).
14
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
temperatures during heatwaves reflect the standard UHI addition to the temperature of the
surrounding areas. The increased night-time temperatures mean that systems have less
opportunity to cool overnight which poses a hazard for some systems including human health.
Cities will likely experience similar temperature increases due to global warming as their
surrounding regions but will remain warmer due to the UHI. It is uncertain whether the intensity
of the UHI will change as the planet warms, with any changes sensitive to changes in other factors
such as green space (i.e., vegetated areas including tree cover), soil moisture and circulation
(Fischer et al. 2012; Zhao et al. 2018). However, in regions which are currently on the urban
fringe, future population growth and urban expansion is expected to result in additional increases
in hot extremes beyond that expected from climate change alone (Argueso et al. 2015; Wouters
et al. 2017). In summary, the UHI effect means that extreme heat events are more severe in urban
regions, regardless of climate change, and urban areas are often not well simulated in coarse
resolution GCMs, although this can be better resolved in RCMs with dedicated urban
parameterisations. It is unclear if this effect will change in the future, but future warming is
expected to be larger in areas which are also experiencing urbanisation.
15
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
- Dynamical downscaling using the CCAM modelling approach (conformal cubic atmospheric
model; Thatcher and McGregor (2011)) applied to 5 GCMs (ACCESS1-0, CanESM2, GFDL-
ESM2M, MIROC5 and NorESM1-M);
- Dynamical downscaling using the recently developed BARPA modelling approach (Bureau
of Meteorology Atmospheric Regional Projections for Australia) applied here to one GCM
(ACCESS1-0 GCM for eastern Australia);
- Dynamical downscaling using the NARCliM modelling approach (NSW and ACT Regional
Climate Model; Evans et al. 2014) applied to 3 GCMs (ACCESS1-0, ACCESS1-3 and
CanESM2) with 2 configurations of each (providing 6 different ensemble members);
- Calibrated data based on the QME method applied to four GCMs (ACCESS1-0, CNRM-
CM5, GFDL-ESM2M and MIROC5 GCMs).
For further details on the selection and assessment of these models see CSIRO & BoM
(2015) and Thatcher et al. (2021). It is generally recommended to consider results from a broad
range of modelling approaches (rather than only relying on a single method) when trying to
sample the uncertainty space for plausible future changes, such that a focus on this report is on
the combined results from this 16-member ensemble of calibrated projections datasets (i.e., 5
from CCAM, 1 from BARPA, 6 from NARCliM and 4 from GCMs).
To calculate the values corresponding to the 10-year ARI, a Generalised Extreme Value
(GEV) approach was used. This is based on 20-year time slices: using 1986–2005 for the historic
period and 2040–2059 for the future climate projection for the RCP8.5 emission pathway (noting
that these projections data are also available for other time periods throughout this century and
historical periods, as well as for RCP4.5).
The projections from these different modelling approached are presented in Fig. 3.2 (i.e.,
based on the GCMs, CCAM, BARPA and NARCliM ensembles, all with QME calibration
applied). The projections for each of these modelling approaches show clear increases in extreme
temperatures for the future climate.
In addition to these results based on CMIP5, some results have recently been published
based on some CMIP6 projections (Grose et al. 2020). Those results show broadly similar changes
for temperature extremes in Australia to those based on CMIP5 projections, noting that
subsequent studies will continue to examine this further including based on a larger set of CMIP6
models than was available for that study.
16
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Figure 3.2: Projected change in values corresponding to the 10-year ARI for daily maximum
temperature at a height of 2 m. This is shown based on GCMs (left panels), CCAM (second to
left panels), BARPA (second to right panels) and NARCliM (right panels), all calibrated using
the QME method. Maps are shown for Australia based on the model ensemble average in each
case. This is presented for the historical climate based on 1986–2005 (upper panels) and future
simulated climate based on 2040–2059 under a high emissions pathway RCP8.5 from CMIP5
(lower panels).
Physical processes
Soil moisture Moderate influence. More frequent dry soil with medium confidence.
Influence on temperature potentially overestimated. Regional models
likely to add value.
Cloud cover and Moderate influence. Low confidence in little change or a small
solar radiation increase. Regional models likely to add value.
17
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Tropical cyclones Small influence. Fewer in the future (medium confidence); regional
models likely to add value.
Urban effects Important for local heat extremes. Urban heat island adds a few degrees
and stays reasonably consistent in future (high confidence); increased
temperature extremes in areas of future urban growth.
Historical climate
Seasonal cycle Models reproduce the seasonal cycle and spatial variability (high
confidence).
Historical trend Strong increase from observations (high confidence). Models reproduce
the trend well (high confidence).
Future climate
RCM: BARPA Strong increase (high confidence). Based on one model to date.
Convection- Uncertain future change due to lack of available data and analysis.
permitting models
18
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
NARCliM) can help with the simulation of some of these processes. Therefore, the relatively high
level of agreement between RCM approaches helps add some confidence for projected future
increases. Based on this overall assessment considering this wide range of factors, there is very
high confidence in the projected direction of change, with a future increase in 10-year ARI
temperatures being very likely (i.e., 90-100% probability).
Based on the above points and details in the Lines of Evidence Table, projected changes
for 10-year ARI temperatures for the 2050 climate are considered here based on the 16-member
ensemble of calibrated projections datasets, combined based on equally weighting each member
of this ensemble. The ensemble median is used as a central estimate of the most probable
projected change (Fig. 3.3). As an estimate of the range of plausible values from the 16 ensemble
members, the second lowest value from the ensemble is used for the 10th percentile and the second
highest value from the ensemble is used for the 90th percentile. These values are calculated
individually at each grid cell location for the median and percentile estimates.
The results show that the future projected temperatures are higher than for the historical
period, including for the lower estimate corresponding to the 10th percentile of the model
ensemble in the future, as well as for the median and upper estimate (90th percentile). This
highlights the considerable degree of agreement between these diverse modelling approaches.
Figure 3.3: Projected values corresponding to the 10-year ARI for daily maximum temperature
at a height of 2 m, based on a 16-member ensemble of calibrated model projections. Maps are
shown through Australia for the historical period (based on 1986-2005; upper panel), as well as
for the future simulated climate (based on 2040-2059 under a high emissions pathway RCP8.5:
lower panels) including a central estimate with lower and upper estimates also provided.
19
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
4.1 Introduction
The standardised method for projections information is applied in this section with a
focus on extreme winds during summer (DJF) in regions around southern and eastern Australia.
Destructive winds in Australia can be caused by severe thunderstorms (mesoscale weather
systems characterised by strong and deep moist convection) and by larger-scale synoptic systems
such as tropical cyclones or extratropical cyclones (including east coast lows) and associated
frontal systems. In particular, severe thunderstorms have been responsible for most of the surface
wind gusts which exceed the 10-year ARI near the major population centres including in southern
and eastern Australia (Holmes 2002), such that the focus here is on severe convective wind gusts,
with other synoptic-scale phenomena also considered here in some sections for completeness.
For Australia, wind gusts are defined by a 3-second average wind speed. Severe
convective wind gusts (SCWs) are considered for the purposes of this study as exceeding 25 m.s-
1
, at a height of 10 meters above ground level, caused by thunderstorm outflow. This threshold
(equivalent to exceeding 90 km.hr-1) is consistent with the threshold used for severe weather
forecasting and operational warnings produced by the Australian Bureau of Meteorology. While
gusts of around 25 m.s-1 may not always be destructive, it is noted that this definition is based on
exceeding that value and therefore also includes higher wind speeds (e.g., around 45 m.s-1) which
have a higher chance of causing property damage. This covers a range of ARI values consistent
with wind speeds such as provided in current Australian standards, including spanning a range
broadly similar to the values for the 10-yr ARI in southern and eastern Australia assuming flat,
open terrain (Holmes, 2002). The atmospheric environments which produce this range of wind
gusts (roughly around 25-45 m.s-1) are typically characterised by unstable atmospheric conditions
(i.e., conducive for convection) as well as likely to include conditions favourable for convective
organisation which can lead to increased severity of hazards (such as can be associated with strong
wind shear between vertical levels (Taszarek et al. 2017)). Tornadoes are a special class of severe
convective winds that are not considered here, including due to their very rare occurrence at a
given location and their very small spatial scale, as well as noting that the design standards widely
used in Australia do not intend structures to withstand the occurrence of a tornado.
20
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
21
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
potential for increases in the frequency of severe convective wind environments into the late
century (Spassiani 2020), which is similar to historical findings for Europe (Rädler et al. 2018).
There have also been future projections of severe convective wind speeds for Tasmania (Cechet
et al. 2012), applying a severe thunderstorm diagnostic to historical observed wind speeds. In
addition to considering such studies, here we also present projections of SCW environments,
following the method of Brown & Dowdy (2021), as detailed in Section 4.4.
Thunderstorm initiation
Given an environment favourable for severe convection (i.e., thermodynamically
unstable conditions, including such as described in the above subsections), thunderstorm initiation
can be triggered in various ways including associated with the occurrence of synoptic systems
(cyclones, fronts and jet streams), atmospheric waves, orographic influences (sea-breezes and
mountains) and small-scale moisture fluctuations (Weckwerth 2000). There are considerable
uncertainties for the influence of climate change on triggering mechanisms such as these for
thunderstorm initiation, with some details as follows.
Projection studies tend to indicate that changes in synoptic initiation mechanisms such as
mid-latitude extratropical cyclones (including east coast lows: ECLs) are not clear for Australia
during the summer months (Catto et al. 2014; Pepler et al. 2016; Dowdy et al. 2019a). Cyclone-
related convection is sensitive to coastal sea surface temperatures (SSTs) which can be a source
of warm and moist air relevant for thunderstorm occurrence (Chambers et al. 2015), noting that
the Tasman Sea east of Australia is a region of accelerated ocean warming including with the
strengthening of the East Australia Current (EAC) (CSIRO and BoM 2015) and that this region
has shown an increase in thunderstorm activity based on environmental conditions over recent
decades from reanalysis data (Dowdy 2020a). Projections related to fronts were discussed in
Section 3.2, indicating considerable uncertainty, with little or no change being the most plausible
outcome.
There is relatively little information on long-term changes to orographic flows such as
sea breezes; however, the strength of the sea breeze is strongly related to the land-sea temperature
contrast, which is expected to increase into the future. One study found an increase in the
frequency and intensity of sea breezes in Adelaide between 1955-2007 (Masouleh et al. 2019).
Regional model simulations at 20 km resolution have been shown to provide a reasonable
simulation of the sea breeze in the Mediterranean region (Drobinski et al. 2018), although
convective parameterisations are less skilful in simulating sea breeze-related CIN (Birch et al.
2015).
In summary, there are a range of processes that are important for thunderstorm initiation.
However, the influence of climate changes on those processes is highly uncertain, including
during summer in southern and eastern Australia. SST increases might potentially help provide
enhanced moisture sources (noting the strengthening EAC due to climate change), although there
is a need for further research to understand how relevant that association might be between the
EAC thunderstorm activity.
22
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
feasible that ENSO may potentially modulate convective initiation mechanisms in some regions.
For example, reduced cloud cover and enhanced sea-breeze circulation in south-east Queensland
during El Niño conditions might potentially increase the frequency of severe thunderstorm events
(Soderholm et al. 2017). It is likely that there is not a strong relationship between ENSO and
synoptic-scale initiation mechanisms, including little or no relationship found between ENSO and
fronts in southern Australia or between ENSO and ECLs in eastern Australia (Rudeva &
Simmonds; Power and Callaghan 2016; Dowdy et al. 2019a). The relationship between ENSO
and SCW environments is shown here in Fig. 4.1a, suggesting very little relationship with ENSO
in eastern Australia during the summer (e.g., only very small regions with significant correlations,
noting that 5% of the region would have a significant correlation on average due to random chance
alone given the use of the 95% confidence level). In summary, the influence of ENSO on SCWs
appears to be relatively weak while noting considerable uncertainties based on limited data and
analysis to date. This is also the case for the relationship between ENSO and severe thunderstorm
occurrence, as well as between ENSO and synoptic initiation mechanisms (including fronts and
cyclones in southern and eastern Australia during summer).
The IOD does not appear to have a strong influence on thunderstorm activity in Australia
during summer (including in southern and eastern Australia) as detailed in Dowdy (2020a), while
noting that study was not specifically focussed on severe thunderstorms which could potentially
have different characteristics to thunderstorms in general. The influence of the IOD on severe
thunderstorms in Australia is currently uncertain based on a lack of previous analyses, although
the IOD may relate to extreme wind gust variability in general, with potential for higher
occurrence frequencies during negative IOD phases (Azorin-Molina et al. 2021). The influence
of the IOD on SCW environments is not significant during the summer in southeastern Australia
but a significant negative correlation is show in Fig. 4.1c in northeast regions.
Similar to the IOD and ENSO, the influence of SAM on severe thunderstorms in Australia
is largely uncertain. No consistent relationship has been found previously with thunderstorm
environments (Dowdy 2020a). From Fig. 4.1b it appears that the negative phase of SAM is more
conducive than the positive phase for SCW environments in eastern Australia. In addition,
enhanced westerlies and associated cold fronts during the negative phase of SAM (Rudeva &
Simmonds, 2015) may increase the frequency of synoptic initiation mechanisms in some southern
regions, and extreme wind gusts from station data have also been shown to be more frequent in
this phase (Azorin-Molina et al., 2021).
Figure 4.1: Correlations for summer between the number of days with a favourable severe
convective wind environment and seasonally-averaged indicators of a) ENSO (Niño3.4 index)
23
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
b) IOD (Dipole Mode Index) and c) SAM (Marshall Index) for 1979-2018. The thunderstorm
environments are calculated from the ERA5 reanalysis (Hersbach et al. 2020) based on the
method of Brown & Dowdy (2021). Hatched regions indicate a significant relationship at the
95% confidence level (e.g., about 5% of the region could be expected hatched on average due to
random chance alone). NINO3.4 and DMI data are attained from the NASA ESRL
(https://psl.noaa.gov/gcos_wgsp/Timeseries/) while SAM data are from
https://legacy.bas.ac.uk/met/gjma/sam.html.
24
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
occurrence frequency of Category 4-5 TCs in the future for Australia, including for the east coast
during summer.
Model assessment
GCMs, reanalyses and commonly used downscaling approaches available for Australia
are unable to resolve the small spatial scales required for simulation of SCWs. Therefore, models
are assessed here in terms of their ability to correctly represent the environments which are
favourable for SCW occurrence, as well as the spatial and temporal variability of these
environments. The ability of environmental model diagnostics to represent the variability of
observed events is also discussed. Additionally, some details on fine-scale (convection-
permitting) modelling are also provided in Section 4.4.
For Australia, GCMs are generally able to represent the spatial distribution of severe
thunderstorm environments, although significant biases may exist for individual models in the
seasonal and diurnal cycle, related to the representation of near-surface moisture (Allen et al.
2014). In other regions, climate model representation of thunderstorm environments has been
shown to vary greatly with individual models (Seeley & Romps 2015), while some models have
been shown to replicate historical trends in environments for sufficiently large climate signals
(Pistotnik et al. 2016). Individual model biases for severe thunderstorm environments may be
addressed to some extent using a multi-model ensemble with bias correction.
Reanalysis models used for historical analyses can reliably represent atmospheric
environments based on observed sounding data (Brown & Dowdy 2021), although some key
elements such as CIN may remained unresolved due to insufficient vertical resolution (King &
Kennedy 2019). SCW diagnostics from these models can broadly represent the seasonal and
diurnal cycle of measured wind events in Australia (Brown & Dowdy 2021). Diagnostics have
also been shown to have a statistically significant correlation with the observed inter-annual
variability of SCW events, which has also been found for other small-scale convective hazards in
other regions, such as tornado events in the United States (Gensini & Brooks 2018). In addition,
25
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
environmental model diagnostics have been shown to explain most of the variability in
convection-permitting modelled thunderstorms (Hoogewind et al. 2017).
In summary, reanalysis data can provide a good representation of thunderstorm
environments as well as SCW environments, such as broadly representing features of the
variability of observed events. Significant biases exist in the representation of these environments
within individual climate models, although biases may be somewhat addressed using multi-model
ensembles with bias correction.
26
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Figure 4.2: Long-term changes in the frequency of days with favourable SCW environments
during the summer, based on ERA5 reanalysis data. Changes are based on four diagnostics, (a)
BDSD, (b) total totals (T-Totals), (c) severe hazards in reduced buoyancy environments
(SHERBE) and (d) the derecho composite parameter (DCP). The change in the mean number of
days per season is shown, calculated as the difference from the period 1979:1998 to the period
1999:2018. Significant changes are represented by hatching based on Student’s t-test with a
90% confidence level (two-tailed).
27
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Figure 4.3: Projected future changes in the frequency of favourable SCW environments during
summer shown as (a-d) a change in the number of days per season and (f-i) percentage changes.
The changes are calculated from the period 1979:2005 to the period 2081:2100 based on a high
emissions pathway (RCP8.5) using an ensemble of 12 GCMs. The ensemble median response is
shown. Changes where at least 10 (out of 12) models agree on the sign of change, as well as
where the seasonal mean number of environments in the historical period is greater than one,
are shown with hatching. These results are intended for broad-scale guidance on some of the
plausible changes that could occur for SCW occurrence in a warmer world, including on
direction of change and estimated range of potential future change as represented by these
metrics.
Convection-permitting modelling
Convection-permitting (a.k.a. convection-allowing) modelling has been used in a
relatively limited number of studies as an alternative to the large-scale environmental approaches
commonly used for projections of severe thunderstorms and associated hazards. Although being
very computationally expensive, this type of modelling can have the advantage of simulating
some factors which are more challenging to represent in environmental approaches. This can
potentially include better simulation of CIN and some triggering mechanisms such as the
influence of localised orographic features, as well as potential for improved representation of
some other aspects of thunderstorm characteristics (e.g., potentially providing some estimates of
intensity and morphology in some cases).
28
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
29
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Figure 4.4: Modelled wind gust speed vs observed wind gust speed, presented for different
quantiles of daily maximum wind gusts at 12 locations. Results are presented from the
convection-permitting mid-latitude model run of BARPAC-M (a) as well as from its host model
for the regional configuration of BARPA-R (b). As one example of projected future changes
based on BARPAC-M, the 20-year maximum wind gust is shown under historical (1985-2005)
and future (2039-2058) conditions, with the blue line representing the quantile-matching of
wind speeds between those two periods using data for individual grid points (land only). The
dotted line represents no change from historical to future, with values below and above that line
representing decreases and increases, respectively, in the occurrence frequencies of wind speeds
in the ranges shown.
30
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Physical processes
Historical trend in Little change or fewer through inland eastern Australia with
SCW small region of potential increase in southeast (low confidence,
environments with uncertainty in modelling methods and limited observations).
Future climate
31
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
GCMs: CMIP5 More SCW environments for southern and eastern Australia (low
confidence due to uncertainty in model diagnostics).
RCMs and Some indication of a potential increase, but with very limited
convection- available data and analysis to date (highlighting a need for more
permitting models research).
32
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
modelling (Table 4.2) is intended to be useful for some planning and risk management purposes.
The central estimates of the model ensemble could also be useful in some cases, showing that the
most likely projections for the future is little change or a small increase in frequency.
Table 4.2: Projected percentage changes in severe convective wind environment frequency
(days per season) during summer, based on 12 CMIP5 GCMs, as well as using four diagnostics
(Brown & Dowdy 2021). This results in a 48-member ensemble, with the median, 10th and 90th
percentile changes shown. The changes are calculated from the period 1979:2005 to the period
2081:2100 based on a high emissions pathway (RCP8.5), averaged over Eastern and Southern
Australia (using the regions defined in CSIRO & BoM (2015)).
33
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
5.1 Introduction
Bushfires can be considered as a form of compound event given the range of factors that
influence their occurrence, including based on the combined influence of various weather factors
(from various near-surface conditions to higher-level atmospheric processes including
convection). Bushfire occurrence is also influenced by various other factors including vegetation
conditions (such as relating to fuel load and type) and ignition sources (such as associated with
human activities or with lightning), some of which can be challenging to model (given current
climate modelling capabilities). The primary focus of the analysis presented here is on dangerous
weather conditions for bushfires, with other factors also considered to some degree (i.e., relating
to vegetation conditions and ignition sources).
Bushfire weather conditions are often represented by indices as a useful way of
combining various weather conditions known to influence fire behaviour (e.g., near-surface
humidity, wind speed, temperature and rainfall). Examples of such indices include the Forest Fire
Danger Index (FFDI) commonly used in Australia (McArthur 1967) as well as the Fire Weather
Index (FWI) originally developed in Canada but now widely used throughout the world (Van
Wagner 1987; Field et al. 2017). The FFDI and FWI are both based on near-surface measures of
humidity, wind speed, temperature and rainfall, with a broadly similar order of sensitivity to these
four individual weather conditions (Dowdy et al. 2009). Indices have also been developed for
grass fires, while noting that grass fires were not identified as a key hazard of interest by
stakeholders for this research. Indices are also available for various other fuel types including a
multi-index system currently in development for Australia (known as the Australia Fire Danger
Rating System: AFDRS).
Indices such as the Continuous-Haines index (C-Haines) are based on conditions at
higher levels of the atmosphere and can be useful for indicating risk factors associated with the
occurrence of extreme fire events (including very dangerous fires that generate thunderstorms in
their fire plumes known as pyrocumulonimbus or pyroCb clouds) (Mills & McCaw 2010; Dowdy
et al. 2019b). Many of the more disastrous fire events in recent decades have been associated with
the occurrence of pyroCbs, including for the Canberra fires in 2003 and the Black Saturday fires
in 2009 as well as during the 2019/2020 Black Summer fires (Fromm et al. 2006; Cruz et al. 2012;
McRae et al. 2013; Dowdy et al. 2017; Australian Government 2020). Such events are often
associated with extreme fire weather conditions occurring simultaneously at near-surface levels
(e.g., as indicated by the FFDI) as well as at higher levels (e.g., as indicated by the C-Haines
index) (Dowdy and Pepler 2018; Di Virgilio et al. 2019).
The standardised method for projections information is applied here for extremely
dangerous fire weather conditions during summer in regions around southern and eastern
Australia. The combined influence of multiple different weather conditions known to influence
fire behaviour is considered here. Factors considered include near-surface weather variables such
as humidity, wind speed, temperature and drought measures relating to fuel availability, as well
as other atmospheric phenomena such as the influence of synoptic systems, mesoscale convective
processes as well as large-scale atmospheric and oceanic modes of variability. Although the focus
here is on fire weather, other factors relating to bushfire occurrence are also discussed including
ignition and fuel conditions.
34
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
35
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Fire weather indices such as the FFDI and FWI include drought measures in their
formulation that are more similar to measures of agricultural drought than meteorological drought
in that they include the influence of other weather conditions in addition to rainfall. For example,
temperature is used together with rainfall as input to the Keetch-Byram Drought Index (KBDI)
(Keetch & Byram 1967) as often used as an input for the Drought Factor used in the FFDI (used
to indicate a proxy estimate of fuel availability based on moisture content). In contrast, relative
humidity, temperature and wind speed are used for the multiple different fuel moisture measures
that the formulation of the FWI System includes (Van Wager et al. 1974).
As noted in the section above on individual weather factors, mean temperatures as well
as the frequency of extreme temperature events are projected to increase in the future with high
confidence, together with a general decrease in relative humidity, as well as little change or a
small decrease in wind speed. Considering these factors together with the projected increase in
meteorological drought (including increased frequency, intensity and duration) suggests a likely
increase in the frequency of very dry fuel conditions. However, there are considerable
uncertainties around projected changes in different types of drought as well as fuel moisture
responses to climate change, including as noted in Section 3 in relation to soil moisture
projections. Regional models may add value for some of these factors (e.g., more detail on land
surface processes, rainfall and orographic dependencies).
36
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
regional model approaches (Dowdy et al. 2019b). Those projections indicate an increase in the
number of days with very high fire weather conditions (based on FFDI above 25) as well as an
increase in the number of days with FFDI above the 95th percentile for 1990-2009), noting lower
agreement between models in some parts of eastern Australia. Similarly, future increases were
also projected for the number of days with FFDI above 50 and for the number of days with FFDI
above the 99th percentile for 1990-2009 (Dowdy 2020b). In addition to the projections presented
in those studies, plausible variation above and below such values is indicated from previous
studies based on different metrics and different modelling approaches using FFDI. For example,
relatively large increases have been derived using monthly mean climate changes from 3 GCMs
to scale observations and calculate changes in severe fire weather days with FFDI > 50 (CSIRO
& BoM 2015), as well as other studies that indicate less confidence in large increases in FFDI in
the future (Clarke et al. 2016).
Projections of future climate have also been produced based on other fire weather indices,
including a global study that used the FWI (Abatzoglou et al. 2019) and reported no emergent
climate change signal in general for Australia based on the methods they presented. Although
increases were projected in some regions they were not statistically significant at a high
confidence level noting the high interannual variability that can occur in weather and climate
conditions in Australia (such as due to the influence of large-scale modes of variability including
ENSO, discussed in sections below). Examples such as that based on FWI with little change
indicated, together with the range of FFDI projections from various studies noted above, show
that considerable differences can occur between different studies and highlight the benefit of
considering results from a broad variety of datasets and methods (as is a goal of this standardised
method).
Very dangerous types of fire events have also been examined in relation to climate
change, including extreme pyro-convection conditions (i.e., associated with thunderstorms that
form in fire plumes: pyroCbs). PyroCbs occurred for the Black Saturday fires in 2009 and the
Canberra fires in 2003 fires as well as many examples during the 2019/20 Black Summer fires
(Fromm et al. 2006; McRae et al. 2013; Dowdy et al. 2017; Australian Government 2020).
Significant trends have been found for extreme pyro-convection risk factors including based on
historical data (Dowdy & Pepler 2018) and future projections (Di Virgilio et al. 2019; Dowdy et
al. 2019b). These studies indicate increased risk factors for parts of southern and southeast
Australia as well as decreases in some cases for other regions including in parts of eastern
Australia. However, a range of uncertainties around future changes in convective systems is also
noted, such as the contrasting roles of increasing water vapour content and decreasing lapse rates
that can have various influences on risk factors associated with fire behaviour and/or potential for
convective systems to develop (with details also available in Section 4 around uncertainties in
future projected changes for convective systems).
37
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
contributing to the build-up of extremely hot and dry air, while noting it is not currently known
if this process would change in the future. Blocking (quasi-stationary) highs over the Tasman Sea
can advect hot and dry air from inland regions towards the more densely populated regions closer
to the south and east coasts, as well as interact with approaching cold fronts from the south to
produce strong northwesterly winds, corresponding to a very dangerous set of fire weather
conditions for southern and eastern Australia (Hasson et al. 2009; Reeder et al. 2015; Dowdy et
al. 2017).
The passage of the front (or pre-frontal trough) comprises shifts in wind direction which
can change the direction of fire movement, i.e., the northern flank can become the new head fire
leading to rapid increases in the rate of area burnt. This can cause significant challenges for
firefighters (Cruz et al. 2012). While future projections of cold fronts are generally uncertain, as
detailed in Section 3, one study based on the older generation of climate models (CMIP3) found
a projected increase in frequency of such extreme events from 0.5 to 1-2 per year by the end of
the 21st century (Hasson et al. 2009).
38
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
(relating to negative SAM conditions to some degree) can be associated with more severe fire
weather conditions in central eastern Australia during spring (Lim et al. 2019), with this not
expected to be represented in these results focussed on summer in Fig. 5.1.
Figure 5.1: Correlations for the number of days with FFDI > 99.5th percentile during summer
and measures representing different modes of variability including ENSO (using the NINO3.4
index), SAM (using the SAM index) and IOD (using the DMI index). These correlations are all
based on the period from 1979 to 2019, using one value for each summer period (DJF).
NINO3.4 and DMI data are attained from the NASA ESRL
(https://psl.noaa.gov/gcos_wgsp/Timeseries/) while SAM data are from
https://legacy.bas.ac.uk/met/gjma/sam.html. Pearson's correlation coefficient, r, is shown with
stippling corresponding to statistically significant values at the 95% confidence level (2-tailed).
39
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Section 3.2) suggests that an increase in lightning ignitions might be more likely than a decrease
in the future climate, while noting considerable uncertainties around this.
Changes in vegetation characteristics including amount (fuel load) and type can also
influence fire hazards throughout Australia, noting that this is particularly important for grassfires
in the more northern and central regions of Australia (McKeon et al. 2009). There are potential
increases in fuel loads for various vegetation types associated with projected increases in carbon
dioxide concentrations, often referred to as the 'fertilisation effect' (Clarke et al. 2016), where
higher concentrations of atmospheric carbon dioxide promote vegetation growth (Donohue et al.,
2013). Global drylands have generally been greening over recent decades and the fertilisation
effect has been identified as a causal factor in this greening (Burrell et al. 2020). Consequently,
an increase in some fuel-related fire risk factors may be considered more likely than a decrease,
while noting considerable uncertainties given the relatively limited ability of current climate
models to accurately simulate future changes in some risk factors relating to fuel characteristics.
Similarly, there are also large uncertainties around potential future changes in fuel type, such as
whether or not vegetation may shift to types that tend to burn more frequently during this
transition period to a warmer world, with no studies currently available on this topic for Australia.
Model assessment
The ability of climate models to simulate aspects such as the seasonal cycle, observed
trends, spatial detail and extremes is important for helping to understand the degree of confidence
in future projected changes based on these models. Assessments presented in various studies
(CSIRO & BoM 2015; Di Virgilio et al. 2019; Dowdy et al. 2019b) indicate that global models
40
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Figure 5.2: Projections for values corresponding to the 10-year ARI for daily fire weather
conditions as represented by the FFDI (with the FFDI intended as a useful means of combining
different weather factors known to influence fire behaviour in Australia). This is shown based
on GCMs (left panels), CCAM (second to left panels), BARPA (second to right panels) and
NARCliM (right panels), all calibrated using the QME method. Maps are shown through
Australia based on the model ensemble average in each case, presented for the historical period
(based on 1986–2005; upper panels) as well as for the future simulated climate (based on 2040–
2059 under a high emissions pathway RCP8.5 from CMIP5; lower panels).
41
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Physical processes
Individual weather Strong influence. More extreme temperatures and heatwaves, lower
factors relative humidity; small decrease in wind speed.
Combined near- Strong influence. Projected increase, but not statistically significant, and
surface weather only based on one study.
conditions, FWI
Historical climate
Seasonal cycle Models reproduce the seasonal cycle and spatial variability well (high
confidence).
42
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Historical trend Increase from observations (medium confidence). Models reproduce the
trend well (medium confidence).
Future climate
RCM: CCAM Increase (high confidence in general; medium near east coast).
RCM: NARCliM Increase (high confidence in general; medium near east coast).
RCM: BARPA Increase (high confidence in general; based on one model to date).
Additional factors
Lightning ignitions Strong influence. Influence of climate change largely uncertain but
increase more likely than decrease (low confidence).
Fuel load and type Strong influence. Influence of climate change largely uncertain but
increased fuel load more likely than decrease (low confidence).
43
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
conditions in 2050 are developed here based on combining data from various calibrated modelling
approaches including GCMs (4 ensemble members), CCAM (5 ensemble members), BARPA (1
ensemble member) and NARCliM (6 ensemble members). FFDI data are available from these
models and are the primary data source used here. The contrasting modelling approaches are
combined based on equally weighting the changes.
The ensemble median is used as a central estimate of the most probable projected change
(Fig. 5.3). As an estimate of the range of plausible values, the second lowest value from the
ensemble is used for the 10th percentile and the second highest value is used from the ensemble
is used for the 90th percentile, with these values calculated individually at each grid cell location.
However, given some of the uncertainties and variations between different modelling approaches
and studies as noted in this section (including projections based on the FWI showing smaller
changes than for FFDI), the lower bound of the range provided here has been modified to reflect
the potential for lower values. This is done based on reducing any projected increases for the 10th
percentile by a factor of two (as a qualitative estimate based on expert judgement). For example,
at a given grid-cell location, if the 10th percentile for the future period was higher by a value of 8
as compared to the 1986–2005 value, it would be changed to only be a value of 4 higher than the
1986–2005 value at that location. Projections for any regions that show decreases for the 10th
percentile are not changed. Only the 10th percentile is changed to allow for lower values, but no
lines of evidence suggest these FFDI projections data systematically underestimate future
increases such that the 90th percentile is unchanged and is considered a plausible upper estimate
for the future projected changes for these fire weather conditions.
Figure 5.3: Projected change in values corresponding to the 10-year ARI for daily fire weather
conditions during summer. Maps are shown through Australia for the historical period (based on
1986–2005; upper panel), as well as for the future simulated climate (based on 2040–2059 under
a high emissions pathway RCP8.5: lower panels) including a central estimate with lower and
upper estimates also provided. The data are based on the FFDI, with some modifications based
on considering the broader lines of evidence from Table 5.1.
44
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
6. CONCLUSION
The influence of anthropogenic climate change on extreme temperatures, winds and fire
weather was assessed using this standardised method for projections information. Calibrated data
from GCMs and RCMs were used for temperature and fire weather, with environmental
diagnostics also used for severe convective winds from thunderstorms. The projections presented
here are more extreme than examined in previous studies (e.g., 10-yr ARI projections for fire
weather and severe convective winds), with care taken to communicate uncertainties and
document the comprehensive lines of evidence considered here.
The nationally consistent calibrated projections presented here, including based on new
RCM data from BARPA, CCAM and NARCliM as well as GCMs, are intended to be of use for
a broad range of applications. This includes for applications such as improved planning and
helping to build resilience in relation to the influence of anthropogenic climate change on future
hazards in Australia. Data are available on request.
The resultant productions data include nationally consistent maps corresponding to the
10-year average recurrence interval (ARI) around the middle of this century, with a focus of the
discussion on regions around southern and eastern Australia during summer, as needed for some
risk assessment applications. The projections are also available for other seasons and time periods
throughout this century, as well as for other metrics of extreme or average conditions. The results
for southern and eastern Australia during summer show more extreme temperatures (very high
confidence), more severe winds (low confidence) and more dangerous bushfire conditions (high
confidence in southern Australia; medium confidence in eastern Australia) attributable to
increasing greenhouse gas emissions.
ACKNOWLEDGEMENTS
This research was funded through the Energy Sector Climate Information (ESCI) project
supported by the Australian Government, Bureau of Meteorology (BoM), CSIRO and AEMO.
The authors would like to thank Chiara Holgate (BoM) and Justin Peter (BoM) for providing
reviews on earlier versions of this report as well as others in the ESCI project team for
contributing to various aspects around the information covered in this report, including Ben
Jones (AEMO), Peter Steinle (BoM), Greg Roff (BoM), Ian Watterson (CSIRO), Chi-Hsiang
Wang (CSIRO) and Leanne Webb (CSIRO).
REFERENCES
Abatzoglou, J. T., Williams, A. P., Boschetti, L., Zubkova, M., & Kolden, C. A. (2018). Global
patterns of interannual climate–fire relationships. Global Change Biology, 24(11), 5164-
5175.
Abram, N. J., Henley, B. J., Gupta, A. S., Lippmann, T. J., Clarke, H., Dowdy, A. J., Sharples,
J. J., Nolan, R. H., Zhang, T., Wooster, M. J., & Wurtzel, J. B. (2021). Connections of
climate change and variability to large and extreme forest fires in southeast
Australia. Communications Earth & Environment, 2(1), 1-17.
Allen, J., Karoly, D., & Mills, G. (2011). A severe thunderstorm climatology for Australia and
associated thunderstorm environments. Aust. Meteorol. Oceanogr. J., 61(3), 143–158.
45
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Allen, J. T., Karoly, D. J., & Walsh, K. J. (2014). Future Australian Severe Thunderstorm
Environments. Part I: A Novel Evaluation and Climatology of Convective Parameters from
Two Climate Models for the Late Twentieth Century. J. Clim., 27, 3827–3847,
doi:10.1175/JCLI-D-13-00425.1
Allen, J. T., Karoly, D. J., & Walsh, K. J. (2014). Future Australian Severe Thunderstorm
Environments. Part II: The Influence of a Strongly Warming Climate on Convective
Environments. J. Climate, 27, 3848–3868.
Allen, J. T., & Karoly, D. J. (2014). A climatology of Australian severe thunderstorm
environments 1979 – 2011: Inter-annual variability and ENSO influence. Int. J. Climatol.,
34, 81–97.
Allen, J. T., & Allen, E. R. (2016). A review of severe thunderstorms in Australia. Atmos. Res.,
178–179, 347–366, doi:10.1016/j.atmosres.2016.03.011
Allen, J. (2018). Climate Change and Severe Thunderstorms. Oxford Research Encyclopedia of
Climate Science. Retrieved 13 Jan. 2021, from
https://oxfordre.com/climatescience/view/10.1093/acrefore/9780190228620.001.0001/acref
ore-9780190228620-e-62
Arblaster, J. M., & Alexander, L. V. (2012). The impact of the El Nio-Southern Oscillation on
maximum temperature extremes. Geophysical Research Letters, 39(20), 2–6,
https://doi.org/10.1029/2012GL053409
Argüeso, D., Evans, J. P., Fita, L., & Bormann, K. J. (2014). Temperature response to future
urbanization and climate change. Climate Dynamics, 42(7–8), 2183–2199,
https://doi.org/10.1007/s00382-013-1789-6
Argüeso, D., Evans, J. P., Pitman, A. J., & Di Luca, A. (2015). Effects of city expansion on heat
stress under climate change conditions. PloS One, 10(2), e0117066,
doi:10.1371/journal.pone.0117066
Australian Government (2020). Royal Commission into National Natural Disaster
Arrangements. Australian Government, ISBN: 978-1-921091-45-2.
Azorin-Molina, C., McVicar, T. R., Guijarro, J. A., Trewin, B., Frost, A. J., Zhang, G., Minola,
L., Son, S., Deng, K., & Chen, D. (2021). A decline of observed daily peak wind gusts with
distinct seasonality in Australia, 1941-2016, Journal of Climate, 1-63.
https://doi.org/10.1175/JCLI-D-20-0590.1
Banerjee, A., Fyfe, J. C., Polvani, L. M., Waugh, D., & Chang, K. L. (2020). A pause in
Southern Hemisphere circulation trends due to the Montreal Protocol. Nature, 579(7800),
544–548. https://doi.org/10.1038/s41586-020-2120-4
Bates, B. C., Chandler, R. E., & Dowdy, A. J. (2015). Estimating trends and seasonality in
Australian monthly lightning flash counts. Journal of Geophysical Research, 120(9), 3973–
3983. https://doi.org/10.1002/2014JD023011
Bell, S. S., Chand, S. S., Tory, K. J., Dowdy, A. J., Turville, C., & Ye, H. (2019). Projections of
southern hemisphere tropical cyclone track density using CMIP5 models. Climate
Dynamics, 52(9), 6065-6079.
Berg, A., Sheffield, J., & Milly, P. C. D. (2017). Divergent surface and total soil moisture
projections under global warming. Geophysical Research Letters, 44(1), 236–244,
https://doi.org/10.1002/2016GL071921
46
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Birch, C. E., Roberts, M. J., Garcia-Carreras, L., Ackerley, D., Reeder, M. J., Lock, A. P., &
Schiemann, R. (2015). Sea-Breeze Dynamics and Convection Initiation: The Influence of
Convective Parameterization in Weather and Climate Model Biases. Journal of Climate,
28(20), 8093-8108, https://doi.org/10.1175/JCLI-D-14-00850.1
Blázquez, J., & Solman, S. A. (2017). Fronts and precipitation in CMIP5 models for the austral
winter of the Southern Hemisphere. Climate Dynamics, 50, 2705,
https://doi.org/10.1007/s00382-017-3765-z
Blázquez, J., & Solman, S. A. (2019). Relationship between projected changes in precipitation
and fronts in the austral winter of the Southern Hemisphere from a suite of CMIP5 models.
Climate Dynamics, 52(9–10), 5849–5860, https://doi.org/10.1007/s00382-018-4482-y
Bony, S., et al. (2006). How well do we understand and evaluate climate change feedback
processes? Journal of Climate, 19, 3445–3482, doi:10.1175/JCLI3819.1
Boschat, G., Pezza, A., Simmonds, I., Perkins, S., Cowan, T., & Purich, A. (2015). Large scale
and sub-regional connections in the lead up to summer heat wave and extreme rainfall
events in eastern Australia. Climate Dynamics, 44(7–8), 1823–1840,
https://doi.org/10.1007/s00382-014-2214-5
Brooks, H. E., Lee, J. W., & Craven, J. P. (2003). The spatial distribution of severe
thunderstorm and tornado environments from global reanalysis data. Atmospheric Research,
67–68, 73–94, doi:10.1016/S0169-8095(03)00045-0
Brooks, H. E. (2013). Severe thunderstorms and climate change. Atmospheric Research, 123,
129–138, http://dx.doi.org/10.1016/j.atmosres.2012.04.002
Brown, A., & Dowdy, A. (2019). Extreme wind gusts and thunderstorms in South Australia
analysed from 1979-2017. Bureau Research Report No. 034. Available at:
http://www.bom.gov.au/research/publications/researchreports/BRR-034.pdf (accessed 25
November 2020).
Brown, A., & Dowdy, A. (2021). Severe convection-related winds in Australia and their
associated environments. Journal of Southern Hemisphere Earth Systems Science,
https://doi.org/10.1071/ES19052
Bureau of Meteorology (BOM), & CSIRO (2020). State of the Climate 2020. Available from:
http://www.bom.gov.au/state-of-the-climate/
Burrell, A. L., Evans, J. P., & De Kauwe, M. G. (2020). Anthropogenic climate change has
driven over 5 million km 2 of drylands towards desertification. Nature Communications,
11(1), 1-11.
Cai, W., T. Cowan & M. Raupach (2009). Positive Indian Ocean Dipole events precondition
southeast Australia bushfires. Geophysical Research Letters 23, 3357-3360.
https://doi.org/10.1029/2009GL039902
Cai, W., Wang, G., Dewitte, B., Wu, L., Santoso, A., Takahashi, K., Yang, Y., Careric, A., &
McPhaden, M. J. (2018a). Increased variability of easten Pacific El Nino under greenhouse
warming. Nature, 564, 201-206. https://doi.org/10.1038/s41586-018-0776-9
Cai, W., Wang, G., Gan, B., Wu, L., Santoso, A., Lin, X., Chen, Z., Jia, F., & Yamagata, T.
(2018b): Stabilised frequency of extreme positive Indian Ocean Dipole under 1.5C
warming. Nature Communications, 9, 1419. https://doi.org/10.1038/s41467-018-03789-6
Cai, W., Wu, L., Lengaigne, M., Li, T., McGregor, S., Kug, J.-S., et al. (2019). Pantropical
climate interactions. Science, 363(6430), eaav4236. https://doi.org/10.1126/science.aav4236
47
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Catto, J. L., Nicholls, N., Jakob, C., & Shelton, K. L. (2014). Atmospheric fronts in current and
future climates. Geophysical Research Letters, 41(21), 7642–7650,
https://doi.org/10.1002/2014GL061943
Catto, J. L., Jakob, C., & Nicholls, N. (2015). Can the CMIP5 models represent winter frontal
precipitation? Geophysical Research Letters, 42(20), 2015GL066015,
https://doi.org/10.1002/2015GL066015
Cechet, R. P., Sanabria, L. A., Divi, C. B., Thomas, C., Yang, T., Arthur, W. C., Dunford, M.,
Nadimpalli, K., Power, L., White, C. J. & Bennett, J. C. (2012). Climate Futures for
Tasmania: severe wind hazard and risk technical report.
Chambers, C. R. S., Brassington, G. B., Walsh, K. et al. (2015). Sensitivity of the distribution of
thunderstorms to sea surface temperatures in four Australian east coast lows. Meteorology
and Atmospheric Physics, 127, 499–517. doi:10.1007/s00703-015-0382-4
Chand, S. S., Dowdy, A. J., Ramsay, H. A., Walsh, K. J., Tory, K. J., Power, S. B., Bell, S. S.,
Lavender, S. L., Ye, H., & Kuleshov, Y. (2019). Review of tropical cyclones in the
Australian region: Climatology, variability, predictability, and trends. Wiley
Interdisciplinary Reviews: Climate Change, 10(5), p.e602.
Chapman, S., Thatcher, M., Salazar, A., Watson, J. E. M., & McAlpine, C.A. (2019). The
impact of climate change and urban growth on urban climate and heat stress in a subtropical
city. International Journal of Climatology, 39, 3013– 3030. https://doi.org/10.1002/joc.5998
Chen, J., Dai, A., Zhang, Y., & Rasmussen, K. L. (2020). Changes in convective available
potential energy and convective inhibition under global warming. Journal of Climate, 33(6),
2025-2050.
Chew L. W., Liu X., Li X-X, & Norford, L. K. (2020). Interaction between heat wave and urban
heat island: A case study in a tropical coastal city, Singapore. Atmospheric Research, 247,
105134, https://doi.org/10.1016/j.atmosres.2020.105134
Clarke, H., Lucas, C., & Smith, P. (2013). Changes in Australian fire weather between 1973 and
2010. International Journal of Climatology, 33(4), 931-944.
Clarke, H., Pitman, A. J., Kala, J., Carouge, C., Haverd, V., & Evans, J. (2016). An
investigation of future fuel load and fire weather in Australia. Climatic Change, 139 (3),
591-605.
Clarke, H., & Evans, J. P. (2019). Exploring the future change space for fire weather in
southeast Australia. Theoretical and Applied Climatology, 136(1), 513-527.
Clemesha, R. E. S., Guirguis, K., Gershunov, A., Small, I. J., & Tardy, A. (2018). California
heat waves: their spatial evolution, variation, and coastal modulation by low clouds. Climate
Dynamics, 50, 4285–4301, https://doi.org/10.1007/s00382-017-3875-7
Cruz, M. G., Sullivan, A. L., Gould, J. S., Sims, N. C., Bannister, A. J., Hollis, J. J., & Hurley,
R. J. (2012). Anatomy of a catastrophic wildfire: the Black Saturday Kilmore East fire in
Victoria, Australia. Forest Ecology and Management, 284, 269-285.
CSIRO & Bureau of Meteorology (2015). Climate Change in Australia Information for
Australia’s Natural Resource Management Regions: Technical Report, CSIRO and Bureau
of Meteorology, Australia.
Dawson, A., Palmer, T. N., & Corti, S. (2012) Simulating regime structures in weather and
climate prediction models. Geophysical Research Letters, 39, L21805,
doi:10.1029/2012GL053284
48
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Di Virgilio, G., Evans, J. P., Blake, S. A., Armstrong, M., Dowdy, A. J., Sharples, J., & McRae,
R. (2019). Climate change increases the potential for extreme wildfires. Geophysical
Research Letters, 46(14), 8517-8526.
Diffenbaugh, N. S., M. Scherer, & R. J. Trapp (2013). Robust increases in severe thunderstorm
environments in response to greenhouse forcing. Proceedings of the National Academy of
Sciences, 110, 16361–16366, doi:10.1073/pnas.1307758110
Donohue, R. J., M. L. Roderick, T. R. McVicar, & G. D. Farquhar (2013). Impact of CO2
fertilization on maximum foliage cover across the globe’s warm, arid environments,
Geophysical Research Letters, 40, 3031–3035, doi:10.1002/grl.50563
Doswell, C. A., & Evans, J. S. (2003). Proximity sounding analysis for derechos and supercells:
An assessment of similarities and differences. Atmospheric Research, 67–68, 117–133.
Dowdy, A. J., Mills, G. A., Finkele, K., & de Groot, W. (2009). Australian fire weather as
represented by the McArthur forest fire danger index and the Canadian forest fire weather
index. Centre for Australian Weather and Climate Research Tech. Rep, 10, p.91.
Dowdy, A. J., & Mills, G. A. (2012). Atmospheric and fuel moisture characteristics associated
with lightning-attributed fires. Journal of Applied Meteorology and Climatology, 51(11),
2025-2037. https://doi.org/10.1175/JAMC-D-11-0219.1
Dowdy, A. J. (2014). Long‐term changes in Australian tropical cyclone numbers. Atmospheric
Science Letters, 15(4), 292-298. https://doi.org/10.1002/asl2.502
Dowdy, A., Grose, M., Timbal, B., Moise, A., Ekstrom, M., Bhend, J., & Wilson, L. (2015).
Rainfall in Australia’s eastern seaboard: a review of confidence in projections based on
observations and physical processes. Australian Meteorological and Oceanographic
Journal, 65(1), 107–126. https://doi.org/10.22499/2.6501.008
Dowdy, A. J. (2016). Seasonal forecasting of lightning and thunderstorm activity in tropical and
temperate regions of the world. Scientific Reports, 6, 1–10, doi:10.1038/srep20874
Dowdy, A. J., & J. L. Catto (2017). Extreme weather caused by concurrent cyclone, front and
thunderstorm occurrences. Scientific Reports, 1–8, doi:10.1038/srep40359
Dowdy, A. J., Fromm, M. D., & McCarthy, N. (2017). Pyrocumulonimbus lightning and fire
ignition on Black Saturday in southeast Australia. Journal of Geophysical Research:
Atmospheres, 122(14), 7342-7354. https://doi.org/10.1002/2017JD026577
Dowdy, A. J. (2018). Climatological variability of fire weather in Australia. Journal of Applied
Meteorology and Climatology, 57(2), 221-234. https://doi.org/10.1175/JAMC-D-17-0167.1
Dowdy, A. J., & Pepler, A. (2018). Pyroconvection risk in Australia: Climatological changes in
atmospheric stability and surface fire weather conditions. Geophysical Research Letters,
45(4), 2005-2013. https://doi.org/10.1002/2017GL076654
Dowdy, A. J., Pepler, A., Di Luca, A., Cavicchia, L., Mills, G., Evans, J., et al. (2019a). Review
of Australian east coast low pressure systems and associated extremes. Climate Dynamics,
53, 4887, https://doi.org/https://doi.org/10.1007/s00382-019-04836-8
Dowdy, A. J., Ye, H., Pepler, A., Thatcher, M., Osbrough, S. L., Evans, J. P., Di Virgilio, G., &
McCarthy, N. (2019b). Future changes in extreme weather and pyroconvection risk factors
for Australian wildfires. Scientific Reports, 9(1), 1-11. https://doi.org/10.1038/s41598-019-
46362-x
Dowdy, A. J. (2020a). Climatology of thunderstorms, convective rainfall and dry lightning
environments in Australia. Climate Dynamics, doi:10.1007/s00382-020-05167-9
49
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Dowdy, A. J. (2020b). Seamless climate change projections and seasonal predictions for
bushfires in Australia. Journal of Southern Hemisphere Earth Systems Science, 70(1), 120-
138. https://doi.org/10.1071/ES20001
Drobinski, P., Bastin, S., Arsouze, T. et al. (2018). North-western Mediterranean sea-breeze
circulation in a regional climate system model. Climate Dynamics, 51, 1077–1093.
https://doi.org/10.1007/s00382-017-3595-z
Evans, J. P., Ji, F., Lee, C., Smith, P., Argüeso, D., & Fita, L. (2014). Design of a regional
climate modelling projection ensemble experiment–NARCliM. Geoscientific Model
Development, 7(2), 621-629.
Fasullo, J. T., Otto-Bliesner, B. L., & Stevenson, S. (2018). ENSO’s Changing Influence on
Temperature, Precipitation, and Wildfire In a Warming Climate. Geophysical Research
Letters, 1–10, https://doi.org/10.1029/2018GL079022
Field, R. D., Spessa, A. C., Aziz, N. A., Camia, A., Cantin, A., Carr, R., De Groot, W. J.,
Dowdy, A. J., Flannigan, M. D., Manomaiphiboon, K., & Pappenberger, F. (2015).
Development of a global fire weather database. Natural Hazards and Earth System
Sciences, 15(6), 1407-1423.
Fischer, E. M., Oleson, K. W., and Lawrence, D. M. (2012). Contrasting urban and rural heat
stress responses to climate change, Geophysical Research Letters, 39, L03705,
doi:10.1029/2011GL050576
Fita L., Evans, J. P., Argüeso, D., King, A., & Liu, Y. (2016). Evaluation of the regional climate
response in Australia to large-scale climate modes in the historical NARCliM simulations.
Climate Dynamics, 1–15, https://doi.org/10.1007/s00382-016-3484-x
Freund, M. B., Brown, J. R., Henley, B. J., Karoly, D. J., & Brown J. N. (2020). Warming
patterns affect El Nino diversity in CMIP5 and CMIP6 models. Journal of Climate,
10.1175/JCLI-D-19-0890.1
Fromm, M., Tupper, A., Rosenfeld, D., Servranckx, R., & McRae, R. (2006). Violent pyro‐
convective storm devastates Australia's capital and pollutes the stratosphere. Geophysical
Research Letters, 33(5).
Garfinkel, C. I., Waugh, D. W., & Polvani, L. M. (2015). Recent Hadley cell expansion: The
role of internal atmospheric variability in reconciling modeled and observed trends.
Geophysical Research Letters, 42(24), 10824–10831,
https://doi.org/10.1002/2015GL066942
Gartland, L. (2011). Heat Islands: Understanding and mitigating heat in urban areas. Routledge.
ISBN 9781849712989
Gensini, V. A., Ramseyer, C., & Mote, T. L. (2014). Future convective environments using
NARCCAP. International Journal of Climatology, 34, 1699–1705, doi:10.1002/joc.3769
Gensini, V. A., & Brooks, H. E. (2018). Spatial trends in United States tornado frequency. npj
Climate and Atmospheric Science, 1(1), 1-5. https://doi.org/10.1038/s41612-018-0048-2
Gibson, P. B., Perkins-Kirkpatrick, S. E., Uotila, P., Pepler, A. S., & Alexander, L. V. (2017).
On the use of self-organizing maps for studying climate extremes. Journal of Geophysical
Research, 122(7), 3891–3903, https://doi.org/10.1002/2016JD026256
Grise, K. M., Davis, S. M., Simpson, I. R., Waugh, D. W., Fu, Q., Allen, R. J., et al. (2019).
Recent tropical expansion: Natural variability or forced response? Journal of Climate, 32(5),
1551–1571, https://doi.org/10.1175/JCLI-D-18-0444.1
50
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
51
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
Imran, H. M., Kala, J., Ng, A. W. M. & Muthukumaran, S. (2019). Impacts of future urban
expansion on urban heat island effects during heatwave events in the city of Melbourne in
southeast Australia. Quarterly Journal of the Royal Meteorological Society, 145, 2586–
2602, https://doi.org/10.1002/qj.3580
IPCC (2013). Climate Change 2013: The Physical Science Basis. Contribution of Working
Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change
[Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y.
Xia, V. Bex & P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA, 1535pp.
Jakob, D. (2010). Challenges in developing a high-quality surface wind-speed data-set for
Australia. Australian Meteorological and Oceanographic Journal, 60, 227-236,
10.22499/2.6004.001.
Jovanovic, B., Jones, D. A., & Collins, D. (2008). A high-quality monthly pan evaporation
dataset for Australia. Climatic Change, 87, 517–35.
Jovanovic, B., Collins, D., Braganza, K., Jakob, D. and Jones, D.A. (2011). A high-quality
monthly total cloud amount dataset for Australia. Climatic Change, 108(3), 485-517.
Keetch, J. J., & Byram, G. M. (1968). A drought index for forest fire control. USDA Forest
Service, Research Paper SE-38.
Kendon, E. J., et al. (2017). Do convection-permitting regional climate models improve
projections of future precipitation change? Bulletin of the American Meteorological Society,
98, 79–93, doi:10.1175/BAMS-D-15-0004.1
Kent, D. M., Kirono, D. G. C., Timbal, B., & Chiew, F. H. S. (2013). Representation of the
Australian sub-tropical ridge in the CMIP3 models. International Journal of Climatology,
33(1), 48–57, https://doi.org/10.1002/joc.3406
King, A. T., & A. D. Kennedy (2019). North American supercell environments in atmospheric
reanalyses and RUC-2. Journal of Applied Meteorology and Climatology, 58, 71–92,
doi:10.1175/JAMC-D-18-0015.1
Knutson, T., Camargo, S. J., Chan, J. C., Emanuel, K., Ho, C. H., Kossin, J., Mohapatra, M.,
Satoh, M., Sugi, M., Walsh, K., & Wu, L. (2020). Tropical cyclones and climate change
assessment: Part II: Projected response to anthropogenic warming. Bulletin of the American
Meteorological Society, 101(3), E303-E322.
Kuchera, E. L. & Parker, M. D. (2006). Severe Convective Wind Environments. Weather
Forecasting, 21(4), 595–612.
Leslie, L. M., Leplastrier, M., & Buckley, B. W. (2008). Estimating future trends in severe
hailstorms over the Sydney Basin: A climate modelling study. Atmospheric Research, 87,
37–51.
Lim, E.-P., Hendon, H. H., Arblaster, J. M., Delage, F., Nguyen, H., Min, S.-K., & Wheeler, M.
C. (2016). The impact of the Southern Annular Mode on future changes in Southern
Hemisphere rainfall. Geophysical Research Letters, 2016GL069453,
https://doi.org/10.1002/2016GL069453
Lim, E.-P., Hendon, H. H., Boschat, G., Hudson, D., Thompson, D. W. J., Dowdy, A. J., &
Arblaster, J. M. (2019). Australian hot and dry extremes induced by weakenings of the
stratospheric polar vortex. Nature Geoscience, 12(11), 896–901,
https://doi.org/10.1038/s41561-019-0456-x
52
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Lim, E. P., Hendon, H. H., Butler, A. H., Thompson, D. W., Lawrence, Z., Scaife, A. A.,
Shepherd, T. G., Polichtchouk, I., Nakamura, H., Kobayashi, C., & Comer, R. (2021). The
2019 Southern Hemisphere stratospheric polar vortex weakening and its impacts. Bulletin of
the American Meteorological Society, 1-50.
Loughran, T. F., Pitman, A. J., & Perkins-Kirkpatrick, S. E. (2019). The El Niño–Southern
Oscillation’s effect on summer heatwave development mechanisms in Australia. Climate
Dynamics, 52(9–10), 6279–6300. https://doi.org/10.1007/s00382-018-4511-x
Lucas, C., (2010). A High-quality Historical Humidity Database for Australia. CAWCR
Technical Report 24, Bureau of Meteorology, Docklands, Victoria, Australia.
Maloney, E. D., Adames, Á. F., & Bui, H. X. (2019). Madden–Julian oscillation changes under
anthropogenic warming. Nature Climate Change, 9(1), 26–33,
https://doi.org/10.1038/s41558-018-0331-6
Marjani, S., Alizadeh-Choobari, O., & Irannejad, P. (2019). Frequency of extreme El Niño and
La Niña events under global warming. Climate Dynamics, 53(9), 5799–5813,
https://doi.org/10.1007/s00382-019-04902-1
Marshall, A. G., Hudson, D., Wheeler, M. C., Alves, O., Hendon, H. H., Pook, M. J., & Risbey,
J. S. (2013). Intra-seasonal drivers of extreme heat over Australia in observations and
POAMA-2. Climate Dynamics, 43(7–8), 1915–1937, https://doi.org/10.1007/s00382-013-
2016-1
Marshall, G. J. (2003). Trends in the Southern Annular Mode from observations and reanalyses.
Journal of Climate, 16(24), 4134–4143, https://doi.org/10.1175/1520-
0442(2003)016<4134:TITSAM>2.0.CO;2
Masouleh, Z. P., Walker, D. J., & McCauley Crowther, J. (2019). A Long-Term Study of Sea-
Breeze Characteristics: A Case Study of the Coastal City of Adelaide, Journal of Applied
Meteorology and Climatology, 58(2), 385-400.
Mastrandrea, M. D., et al. (2011). The IPCC AR5 guidance note on consistent treatment of
uncertainties: a common approach across the working groups. Climatic Change, 108(4),
675-691.
McArthur, A. G. (1967). Fire behaviour in eucalypt forests. Forestry and Timber Bureau, No
107, Canberra, Australia.
McKeon, G. M., et al. (2009). Climate change impacts on northern Australian rangeland
livestock carrying capacity: a review of issues, Rangel. J., 31, 1–29.
McRae, R. H., Sharples, J. J., Wilkes, S. R., & Walker, A. (2013). An Australian pyro-
tornadogenesis event. Natural Hazards, 65(3), 1801-1811.
McVicar, T. R., Van Niel, T. G., Li, L. T., Roderick, M. L., Rayner, D. P., Ricciardulli, L., &
Donohue, R. J. (2008). Wind speed climatology and trends for Australia, 1975–2006:
Capturing the stilling phenomenon and comparison with near-surface reanalysis output.
Geophysical Research Letters, 35(20), L20403, https://doi.org/10.1029/2008GL035627
Mills, G. A., & McCaw, W. L. (2010). Atmospheric stability environments and fire weather in
Australia: extending the Haines Index. CAWCR Technical Report, Centre for Australia and
Weather Climate Research, Melbourne, Australia.
Myers, T. A., & Norris, J. R. (2015). On the Relationships between Subtropical Clouds and
Meteorology in Observations and CMIP3 and CMIP5 Models. Journal of Climate, 28, 8,
2945-2967, https://doi.org/10.1175/JCLI-D-14-00475.1
53
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
NESP (2020). Earth Systems and Climate Change Hub. 2020. Scenario analysis of climate-
related physical risk for buildings and infrastructure: climate science guidance. Technical
report by the National Environmental Science Program (NESP) Earth Systems and Climate
Change Science (ESCC) Hub for the Climate Measurement Standards Initiative, ESCC Hub
Report No.21, ISBN: 978-0-6489444-1-6.
Nguyen, H., Lucas, C., Evans, A., Timbal, B., & Hanson, L. (2015). Expansion of the Southern
Hemisphere Hadley cell in response to greenhouse gas forcing. Journal of Climate, 28(20),
8067–8077, https://doi.org/10.1175/JCLI-D-15-0139.1
Parker, T. J., Berry, G. J., & Reeder, M. J. (2013). The influence of tropical cyclones on
heatwaves in Southeastern Australia. Geophysical Research Letters,
https://doi.org/10.1002/2013GL058257
Parker, T. J., Berry, G. J., Reeder, M. J., & Nicholls, N. (2014). Modes of climate variability
and heat waves in Victoria, southeastern Australia. Geophysical Research Letters, 41(19),
6926–6934, https://doi.org/10.1002/2014GL061736
Patterson, M., Bracegirdle, T., & Woollings, T. (2019). Southern Hemisphere atmospheric
blocking in CMIP5 and future changes in the Australia‐New Zealand sector. Geophysical
Research Letters, 5, 1–10, https://doi.org/10.1029/2019gl083264
Pepler, A. S., Alexander, L. V, Evans, J. P., & Sherwood, S. C. (2015). Zonal winds and
southeast Australian rainfall in global and regional climate models. Climate Dynamics, 46,
123–133, https://doi.org/10.1007/s00382-015-2573-6
Pepler, A. S., Di Luca, A., Ji, F., Alexander, L. V., Evans, J. P., & Sherwood, S. C. (2016),
Projected changes in east Australian midlatitude cyclones during the 21st century.
Geophysical Research Letters, 43, 334– 340, doi:10.1002/2015GL067267
Pepler, A., Ashcroft, L., & Trewin, B. (2018). The relationship between the subtropical ridge
and Australian temperatures. Journal of Southern Hemisphere Earth Systems Science, 68(1),
201–214, https://doi.org/10.22499/3.6801.011
Pepler, A. S., Dowdy, A. J., & Hope, P. (2021). The differing role of weather systems in
southern Australian rainfall between 1979 – 1996 and 1997 – 2015. Climate Dynamics,
https://doi.org/10.1007/s00382-020-05588-6
Perkins, S. E., Argüeso, D., & White, C. J. (2015). Relationships between climate variability,
soil moisture, and Australian heatwaves. Journal of Geophysical Research: Atmospheres,
120(16), 2015JD023592, https://doi.org/10.1002/2015JD023592
Perkins-Kirkpatrick, S. E., White, C. J., Alexander, L. V., Argüeso, D., Boschat, G., Cowan, T.,
et al. (2016). Natural hazards in Australia: heatwaves. Climatic Change, 139(1), 101–114.
https://doi.org/10.1007/s10584-016-1650-0
Pistotnik, G., P. Groenemeijer, & R. Sausen (2016). Validation of Convective Parameters in
MPI-ESM Decadal Hindcasts (1971-2012) against ERA-Interim Reanalyses.
Meteorologische Zeitschrift, 25, 753–766, doi:10.1127/metz/2016/0649.
Power, S., Casey, T., Folland, C., Colman, A., & Mehta, V. (1999). Inter-decadal modulation of
the impact of ENSO on Australia. Climate Dynamics, 15(5), 319–324.
Power, S. B., & Callaghan, J. (2016). Variability in severe coastal flooding, associated storms,
and death tolls in Southeastern Australia since the mid-nineteenth century. Journal of
Applied Meteorology and Climatology, 55(5), 1139–1149, https://doi.org/10.1175/JAMC-
D-15-0146.1
54
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
55
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER PROJECTIONS USING A STANDARDISED METHOD
56
EXTREME TEMPERATURE, WIND AND BUSHFIRE WEATHER USING A STANDARDISED METHOD
Van Wagner, C. E. (1974). Structure of the Canadian forest fire weather index, Vol. 1333,
Ontario: Environment Canada, Forestry Service.
Voigt, A., Albern, N., Ceppi, P., Grise, K., Li, Y., & Medeiros, B. (2020). Clouds, radiation,
and atmospheric circulation in the present‐day climate and under climate change. Wiley
Interdisciplinary Reviews: Climate Change, p.e694.
Walsh, K., et al. (2016). Natural hazards in Australia: storms, wind and hail. Climate Change,
139, 55–67, doi:10.1007/s10584-016-1737-7
Wang, C., Wang, X. & Khoo, Y. B. (2013). Extreme wind gust hazard in Australia and its
sensitivity to climate change. Nat. Hazards, 67, 549–567, DOI:10.1007/s11069-013-0582-5
Wang, G., & Cai, W. (2013). Climate-change impact on the 20th-century relationship between
the Southern Annular Mode and global mean temperature. Scientific Reports 3, 2039,
https://doi.org/10.1038/srep02039
Wang, B., Chen, G. & Liu, F. (2019). Diversity of the Madden-Julian Oscillation. Science
Advances, 5, eaax0220. https://doi.org/10.1126/sciadv.aax0220
Waugh, D. W., Garfinkel, C. I., & Polvani, L. M. (2015). Drivers of the recent tropical
expansion in the southern hemisphere: Changing SSTs or ozone depletion? Journal of
Climate, 28(16), 6581–6586, https://doi.org/10.1175/JCLI-D-15-0138.1
Weckwerth, T. M. (2000). The effect of small-scale moisture variability on thunderstorm
initiation. Monthly Weather Review, 128(12), 4017-4030.
Wehrli, K., Guillod, B. P., Hauser, M., Leclair, M., & Seneviratne, S. I. (2019). Identifying key
driving processes of recent heatwaves and droughts, 21, 13643,
https://doi.org/10.1029/2019JD030635
Williams, S., Nitschke, M., Sullivan, T., Tucker, G. R., Weinstein, P., Pisaniello, L., Parton, K.
A., & Bi, P. (2012). Heat and health in Adelaide, South Australia: Assessment of heat
thresholds and temperature relationships. Science of the Total Environment, 414, 126–133,
https://doi.org/10.1016/j.scitotenv.2011.11.038
Woollings, T., Barriopedro, D., Methven, J., Son, S. W., Martius, O., Harvey, B., et al. (2018).
Blocking and its Response to Climate Change. Current Climate Change Reports,
https://doi.org/10.1007/s40641-018-0108-z
Wouters, H., et al. (2017). Heat stress increase under climate change twice as large in cities as
in rural areas: A study for a densely populated midlatitude maritime region, Geophysical
Research Letters., 44, 8997– 9007, doi:10.1002/2017GL074889
Zhao, L., Oppenheimer, M., Zhu, Q., Baldwin, J. W., Ebi, K. L., Bou-Zeid, E., Guan, K., & Liu,
X. (2018). Interactions between urban heat islands and heat waves. Environmental Research
Letters, 13, 034003, https://doi.org/10.1088/1748-9326/aa9f73
57