Climatology of Extreme Events
International Centre for Applied Climate Sciences
University of Southern Queensland
Citation
Insert citation text here
Copyright
© 20XXinsert copyright statement here.
Disclaimer
The view expressed herein are not necessarily the views of the Commonwealth of Australia, and the Commonwealth
does not accept responsibility for any information or advice contained herein.
Contents
Executive Summary ............................................................................................................................................................ 3
1. Introduction .................................................................................................................................................................... 4
2. Data................................................................................................................................................................................. 6
3. Methodology .................................................................................................................................................................. 8
4. Changes In Extreme Temperature Indices .................................................................................................................... 11
Changes In Duration and Frequency Of Extreme Events Each Year.......................................................................... 11
Changes In Absolute Temperature Extremes And Frequencies ................................................................................ 12
5. Change in Indices of Extreme Precipitation .................................................................................................................. 20
6. Conclusions ................................................................................................................................................................... 25
7. Acknowledgments ........................................................................................................................................................ 27
8. Appendix A. Regional Climate Data .............................................................................................................................. 28
8.1.1 Toowoomba ......................................................................................................................................................... 28
8.1.2 Border Rivers Maranoa-Balonne (St George) ..................................................................................................... 30
8.1.3 Northern Tablelands (Glen Innes) ....................................................................................................................... 32
8.1.4 North West (Gunnedah) ...................................................................................................................................... 34
8.1.5 Central West (Dubbo) .......................................................................................................................................... 36
9. Appendix B Climate Change Indices.............................................................................................................................. 38
10. References .................................................................................................................................................................. 42
Climatology of Extreme Events
2
Executive Summary
This selection of data from the Central Slopes
the specific needs of stakeholders (as cautioned by
region details historical trends in extreme daily
other authors including Haylock and Nicholls 2000,
minimum and maximum temperatures and daily
Collins et al 2000).
precipitation. It is designed to enhance NRM
planning for climate change and is intended to
supplement other reports from the Regional
Natural
Resource
Management
Planning
for
Climate Change Fund that detail the region’s
climatology, climate drivers and climate change
projections.
The trends in extreme temperature indices are
dominated by a strong, spatially consistent
increasing trend in the temperatures of warm
nights. The temperatures of warm and cool days
are increasing: warm days faster than cool days.
These effects are stronger in winter than summer.
Also, the warm days and nights are warming at a
rate greater than the cool day and nights.
Generally, most of the minimum temperature
indices were changing at a greater rate than the
maximum temperature indices.
Compared with the indices of temperature
extremes, the changes in precipitation extremes
between 1960 and 2010 are not as spatially
coherent, nor are the trends as statistically
significant. This is a commonly occurring feature in
studies of precipitation extremes (Plummer et al.
1999, Suppiah and Hennessy 1998).
Caution needs to be exerted when comparing
results from different studies on extremes, as the
definition of indices frequently varies depending on
Climatology of Extreme Events
3
1. Introduction
NRM organisations across the Central Slopes (CS)
for Climatology (CCl)/World Climate Research
region of NSW require climate data and
Programme (WCRP) project on Climate Variability
information to enhance their decision making
and Predictability (CLIVAR) Expert Team on
processes regarding adaptation to future climate
Climate Change Detection, Monitoring and Indices
change.
(ETCCDMI). More information can be found at that
Projections of future climate change for the region
projects website:
are provided by CSIRO as part of the “Stream2 of
http://etccdi.pacificclimate.org/index.shtml.
the Regional Natural Resource Management
of such an internationally accepted collection of
Planning for Climate Change Fund” (Ekstrom et al
indices ensures an objective measurement and
2014). To enable appropriate interpretation of
characterization of climate variability and change.
these future scenarios, it is critical to understand
These indices have been used in studies of global
the historical climate of the region. The CSIRO
and Australian climate variability (Alexander and
regional projections report (Ekstrom et al 2014) for
Arblaster 2009; Alexander et al. 2006; Frich et al.
the
historical
2002) and some are also used on the Bureau of
climatology of the region with a focus on mean
Meteorology’s webpages that discuss extreme
annual and seasonal climate. This report aims to
events, such as
supplement that information by providing detail
http://www.bom.gov.au/climate/change/about/e
on the historical climatology of extreme events in
xtremes.shtml. These indices are also discussed in
the CS region. This information will enhance the
(Zhang et al. 2011).
CS region has detailed the
Use
interpretation of projections of changes in
The following sections provide detail on the data
extreme events in future climate scenarios.
and
sources, the indices analysed, and results for each
precipitation data for key locations in each of the
NRM authority. Each index will be plotted in a
NRM regions within greater CS region, a suite of
timeseries graph, however not all will be shown in
indices detailing different aspects of climate
this document. Note that this document is not
extremes are analysed. The climate indices
intended to provide discussion of the general
analysed here are those produced by the joint
climate of the CS, climate drivers, and trends in
World Meteorological Organization Commission
mean climate. These can be found elsewhere
Using
observed
daily
temperature
Climatology of Extreme Events
4
including (Gallant et al. 2012; Manton et al. 2001;
the latest State of the Climate 2014 report from
Collins et al. 2000; Marshall et al. 2013; Haylock
the CSIRO/Bureau of Meteorology.
and Nicholls 2000; Klingaman et al. 2013), and also
Climatology of Extreme Events
5
2. Data
The CMA/LLS groups selected locations in their
radiation,
management region for detailed analysis and
pressure data was obtained from 1960 to 2010.
simulation modelling as shown in Table 1 and
The SILO PPD data was the preferred data source
Figure 1. Historical climate data for these locations
for two main reasons: 1) climate change data
was obtained from the Queensland Government
(CMIP3 models) was also available at these
SILO database,
locations via the Consistent Climate Change
http://www.longpaddock.qld.gov.au/silo
rainfall,
evaporation,
and
vapour
(Jeffrey
Scenarios component of SILO ((Burgess et al.
et al. 2001) which is a repository of historical data
2012)), and 2) NRM modelling such as Agricultural
from the Bureau of Meteorology (in SILO this is
Production Systems SIMulator (APSIM, (Keating et
called “Patched Point Data”). Historical daily
al. 2003)) has been designed in parallel with SILO
maximum and minimum temperature, solar
in terms of data formatting ((Jeffrey et al. 2001)).
Location of Climate Stations
-25.0
MitchellRoma/Waverley Downs
Chinchilla
Dalby
Toowoomba
Pittsworth
-27.5
St George
Warwick
Goondiwindi
LLS
latitude
Border Rivers Maranoa-Balonne
Walgett
-30.0
Pallamallawa/Moree
Warrialda
Glen Innes
Attunga
Narrabri
Condamine
North West
Northern Tablelands
Gunnedah
Nyngan
Central West
Gilgandra
Dubbo
-32.5
146
148
150
152
longitude
1.Meteorological recording station locations.
Climatology of Extreme Events
6
NRM region
Border Rivers MaranoaBalonne
Condamine Alliance
Northern Tablelands
North West
Central West
NAME
NUMBER
LATITUDE (oS)
LONGITUDE (oE)
Goondiwindi
41521
-28.52
150.33
Mitchell
43020
-26.49
147.98
St George
43034
-28.04
148.58
Roma/Waverley Downs
43093
-26.61
148.54
Chinchilla
41017
-26.74
150.6
Dalby
41023
-27.18
151.26
Warwick
41176
-28.22
152.02
Toowoomba
41103
-27.58
151.93
Pittsworth
41082
-27.72
151.63
Attunga
55000
-30.01
150.86
Warialda
54029
-29.54
150.58
Glen Innes
56011
-29.74
151.74
Walgett
48036
-29.67
148.11
Gunnedah
55024
-31.03
150.27
Narrabri
53030
-30.34
149.76
Pallamallawa/Moree
53033
-29.47
150.14
Nyngan
51031
-31.64
147.32
Gilgandra
64024
-31.56
148.95
Dubbo
65012
-32.24
148.61
Table 1. Bureau of Meteorology recording station and their official number and location.
Climatology of Extreme Events
7
3. Methodology
The indices of climate extremes are derived from
The annual precipitation indices in this category
historical daily SILO data: daily precipitation,
represent the amount of rainfall falling above the
minimum and maximum temperatures.
As per
95th (R95p) and 99th (R99p) percentiles and
Alexander et al 2006 (Alexander et al. 2006) we
include, but are not be limited to, the most
have chosen a range of indices from the ETCCDMI
extreme precipitation events in a year. Specifically
ClimDex project that encompass aspects of climate
the indices are:
change of particular relevance to natural resource
a. Very wet days (R95p).
management:
b. Extremely wet days (R99p).
namely
changing
intensity,
frequency, and duration of temperature and
precipitation events.
2. Absolute indices (six), calculated monthly.
There are 5 general types of indices and we
These represent maximum or minimum values
present them here as per the summary of
within a season or year and include
Alexander et al (2006). More specific details can be
a.
found in Appendix A.
(TNx),
c. minimum daily maximum temperature
(10th and 90th decile) of daily minimum and
maximum temperatures. These are calculated
(TXn),
d. minimum daily minimum temperature
each month to avoid being seasonally biased.
Specifically the indices are:
(TNn),
e. maximum 1-day precipitation amount
a. The occurrence of cold nights (TN10p),
occurrence
maximum
b. maximum daily minimum temperature
The temperature percentile-
based indices are the upper and lower deciles
b. The
daily
temperature (TXx),
1. Percentile-based rainfall and temperature
indices (six).
maximum
of
warm
nights
(TN90p),
c. The occurrence of cold days (TX10p),
d. The occurrence of warm days (TX90p).
(RX1day),
f.
maximum 5-day precipitation amount
(RX5day).
3. Threshold indices (six), calculated annually.
These are defined as the number of days on
which a temperature or precipitation value
falls above or below a fixed threshold:
Climatology of Extreme Events
8
a.
Frost Days: the number of days on
c. Consecutive dry days (CDD) defined as
which minimum temperature falls
is the length of the longest dry spell in
below 0 oC (FD),
a year,
b. annual occurrence of tropical nights:
d. Consecutive wet days (CWD) defined
minimum temperature greater than
as is the length of the longest wet spell
20oC (TR),
in a year.
c. number of heavy precipitation days >
10 mm (R10) and
5. Other indices include indices of annual
d. number of very heavy precipitation
days > 20 mm (R20).
precipitation
total
temperature
range
(PRCPTOT),
(DTR),
diurnal
simple
daily
intensity index (SDII), and annual contribution
from very wet days (R95pT).
4.
Duration indices (four), calculated annually.
These define periods of excessive warmth,
Some of the indices have the same name and
cold, wetness or dryness. Specifically they are:
definition as those used in previous studies [e.g.,
a. A cold spell duration indicator (CSDI),
Frich et al., 2002; Klein Tank et al., 2002], but they
defined as the annual number of days
may differ slightly in the way they are computed.
with at least 6 consecutive days when
Of particular importance is a recent finding that
daily minimum temperature less than
inhomogeneities exist at the boundaries of the
10th percentile,
climatological base period used to compute the
b. A warm spell duration indicator
thresholds for percentile based temperature
(WSDI) defined as the annual number
indices, i.e., TN10p, TN90p, TX10p and TX90p,
of days with at least 6 consecutive
because of sampling uncertainty [Zhang et al.,
days
maximum
2005b]. A bootstrapping method proposed by
90th
Zhang et al. [2005b] has been implemented in
when
temperature
percentile,
daily
greater
than
RClimDex and is used to compute indices analysed
in this paper. The bootstrap procedure removes
the inhomogeneities and thus eliminates possible
Climatology of Extreme Events
9
bias in the trend estimation of the relevant
indices.
A full descriptive list of the indices can be obtained
from
http://ccma.seos.uvic.ca/ETCCDMI/list_27_indices
.html.
We utilised the RClimDex software to calculate the
indices
(RClimDexhttp://etccdi.pacificclimate.org/softwar
e.shtml).
Trends for all indices were calculated using a
simple least-squares linear regression. Statistical
significance was determined using the Kendall-tau
test. A trend is deemed “statistically significant” if
it has at least 90% significance. In addition, as per
the discussion of trends in extreme indices at
http://www.bom.gov.au/climate/change/about/e
xtremes.shtml, trends should only be interpreted
at locations at which the index series is non-zero
for more than 50% of years for temperature
indices, and more than 25% of years for rainfall
indices. Seasonal trends are calculated for winter
(JJA,
June-July-August)
and
summer
(DJF,
December-January-February).
Climatology of Extreme Events
10
4. Changes In Extreme Temperature Indices
Changes in Duration and Frequency
Of Extreme Events Each Year
WSDI: linear trend between 1960-2010
-25.0
The length of extreme warm spells each year
(WSDI) is increasing and the length of cold spells
trend for the WSDI is positive for all locations in
the CS region, although not all are statistically
-27.5
magnitude of trend
5
Latitude
(CDSI) is decreasing as shown in Figure 2. The
10
15
-30.0
direction of trend
increasing
significant. The changes in CSDI are not as spatially
consistent across the region as the WSDI, nor is
-32.5
the magnitude of the changes as strong. The
trends for both indices are strongest in the eastern
146
148
150
152
Longitude
part of the CS.
CSDI: linear trend between 1960-2010
Examining the changes in the annual frequency of
-25.0
the minimum temperature threshold indices
Tropical Nights (TN) and Frost Days (FD) as shown
in Figure 3, it is clear there has been a strong and
-27.5
direction of trend
increasing
change in the number of frost days per year is not
decreasing
Latitude
spatially coherent increase in the TR index. The
magnitude of trend
4
-30.0
8
12
as spatially consistent across the region. There
have been strong statistically significant decreases
16
-32.5
in the east, but also increases at a couple of
locations, namely Dubbo, Attunga, and Dalby.
146
148
150
152
Longitude
Figure 2. Trends for annual series of two duration
temperature indices: Warm Spell Duration Index
(WSDI), and Cold Spell Duration Index (CSDI) from 1960
to 2010. Increasing (decreasing) trends are indicated by
a blue (red) triangle which is shaded grey if the trend is
not statistically significant at the 0.1 level.
Climatology of Extreme Events
11
Changes In Absolute Temperature
Extremes And Frequencies
TR: linear trend between 1960-2010
-25.0
Changes in extreme night-time temperatures are
represented by changes in the indices of daily minimum
temperatures tnx (the warmest nights each year) and
-27.5
magnitude of trend
10
Latitude
20
30
40
-30.0
direction of trend
increasing
tnn (the coldest nights each year) as shown in Figure 4.
Generally, the trends of minimum temperatures show
that the changes are greatest in the maximum
minimum temperatures (tnx and tn90p). That is, the
warm tail of the distribution of minimum temperatures
-32.5
is warming more than the cold tail of the distribution
146
148
150
(tnn and tn10p). This was also seen in the comparison
152
Longitude
of TR (a function of Tnx) and FD (a function of Tnn): the
FD: linear trend between 1960-2010
trend in TR was more consistent than that of FD.
-25.0
In winter, the warm nights (tnx) are getting warmer
everywhere in the region, however there is no
consistent pattern with the extreme cold nights (tnn):
-27.5
direction of trend
increasing
there are more stations where the coldest of cold
Latitude
decreasing
magnitude of trend
10
-30.0
20
nights are getting colder. This is a similar pattern of
trends to summer.
30
40
The percentage of extreme warm nights (tn90p) in both
winter and summer is increasing steadily everywhere.
-32.5
The percentage of extreme cold nights (tn10p) is
decreasing nearly everywhere in summer but not so
146
148
150
152
Longitude
Figure 3. Trends for the two temperature indices
derived from daily minimum temperatures: number of
frost days per year (FD), and the number of tropical
nights per year (TR) from 1960 to 2010. Increasing
(decreasing) trends are indicated by a blue (red)
triangle which is shaded grey if the trend is not
statistically significant at the 0.1 level.
consistently in winter. This unexpected weakness and
spatial consistency of trend is also reflected in the
number of frost days shown in Figure 3 and the CSDI in
Figure 2 which are functions of Tnn and Tn10p
respectively.
Climatology of Extreme Events
12
In terms of daytime temperatures shown in Figure 5,
the actual temperature of the coldest days, the number
the highest daily maximum temperature (txx) has
of cold days has actually increased.
increased significantly across the central slopes in both
In summary, as was found with the minimum
winter and summer. The trend is slightly stronger in
temperatures, the magnitudes of the trends are greater
winter with the greatest changes occurring in the north
in the warm end of the distribution of maximum
of the region. A similar spatial pattern is seen in the
temperatures. That is, the warm days are warming at a
changes in tx90p, the percentage of days with a daily
rate greater than the cool days. This asymmetric
th
maximum greater than the 90 percentile. There has
change in lower-and upper-tail extremes causes
been an increase in the percentage of extreme days as
greater temperature variance. This differs to a
measured by tx90p in nearly all locations, with the
greatest increases in the northeast of the region, and
slightly stronger trends in winter.
Examination of changes in the lowest daily maximum
temperatures, that is changes in the coldest daytime
temperatures as measured by the indices txn and tx10p
provide more evidence of a warming CS region and
greater climatic extremes. While the trends are not as
strong as those of the highest daytime temperatures,
there are significant increases in the lowest daytime
temperatures recorded especially in winter. The trends
are much weaker in summer, and in the western region
of the CS the lowest daytime temperatures have
actually been decreasing, although these trends are not
statistically significant.
In terms of the percentage of days in the lowest 10th
percentile of temperatures, there has been a decrease
in the summer months throughout the region. That is,
there are fewer very cold days. This trend is strongest in
the eastern parts of the CS. In winter the picture is
more complex. While there is a consistent increase in
previous study that has shown that in general cold and
warm maximum temperatures (Txn, Txx) are warming
at the same rate ((Trewin 2001).
Generally, most of the minimum temperature
indices were changing at a greater rate than the
maximum temperature indices as can be seen by
comparing figures 4 and 5. This is consistent with
previous studies such as Collins et al. (2000).
One of the strongest signatures of global warming is
the decrease in diurnal temperature range (DTR) which
reflects a general trend of minimum temperatures
increasing
as
a
greater
rate
than
maximum
temperatures (Plummer et al. 1995; Braganza et al.
2004). This reflects the lowest minimum temperatures
(Tnn) warming faster than warm end of the distribution
(Tnx, Plummer, 1999 #92},(Collins et al. 2000),). As
shown in figure 6, in the CS the trend in DTR is not
strong or consistent in direction across the region. In
winter there is a decreasing trend in the north, however
in summer there are both increasing and decreasing
trends. Indeed, as previously discussed while there is a
Climatology of Extreme Events
13
strong spatially consistent increasing trend in the warm
in the maps of cold spell duration (CSDI, TR, and FD)
nights, the decreasing trend in cold nights is not
and in winter minimum temperature indices (Tnn and
consistent in space or direction (Figures 2 and 3). This
Tn10p in Figure 4).
mixed trend in cold minimum temperatures is also seen
Climatology of Extreme Events
14
tnn: linear trend between 1960-2010 in JJA
tnx: linear trend between 1960-2010 in JJA
-25.0
-25.0
-27.5
-27.5
magnitude of trend
1
magnitude of trend
2
3
4
Latitude
Latitude
2
3
4
-30.0
-30.0
direction of trend
direction of trend
increasing
increasing
decreasing
-32.5
-32.5
146
148
150
146
152
148
150
152
Longitude
Longitude
tn10p: linear trend between 1960-2010 in JJA
tn90p: linear trend between 1960-2010 in JJA
-25.0
-25.0
-27.5
-27.5
direction of trend
increasing
magnitude of trend
decreasing
Latitude
10
15
Latitude
5
magnitude of trend
5
-30.0
-30.0
direction of trend
10
increasing
15
decreasing
20
-32.5
-32.5
146
146
148
150
152
148
150
152
Longitude
Longitude
Figure 4(a). Trends in daily minimum temperature indices warm nights (tnx), cold nights (tnn), and the 10 th
(tn10p) and 90th (tn90p) percentiles of these temperatures in winter (JJA) from 1960 to 2010. The trend in
absolute temperatures is in oC, and the trend in the percentiles are %. Increasing (decreasing) trends are
indicated by a blue (red) triangle which is shaded grey if the trend is not statistically significant at the 0.1 level.
Climatology of Extreme Events
15
tnx: linear trend between 1960-2010 in DJF
tnn: linear trend between 1960-2010 in DJF
-25.0
-25.0
-27.5
-27.5
magnitude of trend
direction of trend
1
increasing
magnitude of trend
-30.0
1
2
Latitude
Latitude
decreasing
3
4
-30.0
direction of trend
2
increasing
3
decreasing
-32.5
-32.5
146
148
150
152
146
Longitude
148
150
152
Longitude
tn10p: linear trend between 1960-2010 in DJF
tn90p: linear trend between 1960-2010 in DJF
-25.0
-25.0
-27.5
direction of trend
-27.5
increasing
magnitude of trend
Latitude
10
15
decreasing
Latitude
5
magnitude of trend
5
-30.0
10
20
-30.0
15
direction of trend
20
increasing
-32.5
-32.5
146
148
150
152
Longitude
146
148
150
152
Longitude
Figure 4(b). Trends in daily minimum temperature indices warm nights (tnx), cold nights (tnn), and the 10 th
(tn10p) and 90th (tn90p) percentiles of these temperatures summer (DJF) from 1960 to 2010. The trend in
absolute temperatures is in oC, and the trend in the percentiles is in number of days. Increasing (decreasing)
trends are indicated by a blue (red) triangle which is shaded grey if the trend is not statistically significant at the
0.1 level.
Climatology of Extreme Events
16
.
txx: linear trend between 1960-2010 in JJA
txn: linear trend between 1960-2010 in JJA
-25.0
-25.0
-27.5
-27.5
direction of trend
increasing
magnitude of trend
2
3
-30.0
direction of trend
decreasing
Latitude
Latitude
1
magnitude of trend
0.5
-30.0
1.0
increasing
1.5
2.0
-32.5
-32.5
146
148
150
152
146
148
150
152
Longitude
Longitude
tx90p: linear trend between 1960-2010 in JJA
tx10p: linear trend between 1960-2010 in JJA
-25.0
-25.0
-27.5
-27.5
magnitude of trend
direction of trend
20
direction of trend
-30.0
increasing
increasing
Latitude
Latitude
10
magnitude of trend
-30.0
5
10
decreasing
-32.5
-32.5
146
148
150
Longitude
152
146
148
150
152
Longitude
Figure 5(a). Trends in the daily maximum temperature indices in winter (JJA) derived from daily maximum
temperatures: from 1960 to 2010. Increasing (decreasing) trends are indicated by a blue (red) triangle which is
shaded grey if the trend is not statistically significant at the 0.1 level.
Climatology of Extreme Events
17
txx: linear trend between 1960-2010 in DJF
txn: linear trend between 1960-2010 in DJF
-25.0
-25.0
-27.5
magnitude of trend
Latitude
1
-27.5
2
magnitude of trend
3
direction of trend
increasing
Latitude
1
-30.0
2
direction of trend
-30.0
increasing
decreasing
-32.5
-32.5
146
148
150
152
Longitude
146
tx90p: linear trend between 1960-2010 in DJF
148
150
152
Longitude
tx10p: linear trend between 1960-2010 in DJF
-25.0
-25.0
-27.5
magnitude of trend
5
-27.5
Latitude
10
direction of trend
15
direction of trend
increasing
decreasing
Latitude
increasing
-30.0
decreasing
magnitude of trend
-30.0
5
10
-32.5
-32.5
146
148
150
152
Longitude
146
148
150
152
Longitude
Figure 5(b). Trends in the daily maximum temperature indices in summer (DJF) derived from daily maximum
temperatures: from 1960 to 2010. Increasing (decreasing) trends are indicated by a blue (red) triangle which is
shaded grey if the trend is not statistically significant at the 0.1 level.
Climatology of Extreme Events
18
dtr: linear trend between 1960-2010 in JJA
dtr: linear trend between 1960-2010 in DJF
-25.0
-25.0
-27.5
-27.5
magnitude of trend
direction of trend
Latitude
10
increasing
20
increasing
decreasing
Latitude
direction of trend
-30.0
magnitude of trend
-30.0
10
decreasing
20
30
-32.5
-32.5
146
148
150
Longitude
152
146
148
150
152
Longitude
Figure 6. Trends for diurnal temperature range in winter (JJA, left) and summer (DJF,right) from 1960 to 2010.
Increasing (decreasing) trends are indicated by a blue (red) triangle which is shaded grey if the trend is not statistically
significant at the 0.1 level.
Climatology of Extreme Events
19
5. Change in Indices of Extreme Precipitation
Compared with the indices of temperature
trend in a different direction (eg there may be a
extremes, the changes in precipitation extremes
positive trend in 95th percentile and a negative
between 1960 and 2010 are not as spatially
trend in the 99th percentile).
coherent, nor are the trends as statistically
Examining these locations where the R95ptot is
significant. This has been observed in many other
increasing, we from the SDII in Figure 9 that the
studies including (Suppiah and Hennessy 1998,
rainfall intensity (the amount of rain falling on wet
Plummer et al. 1999; Alexander and Arblaster
days) is also increasing. The direction of the trend
2009).
at specific locations is almost identical to that of
R99ptot, especially for the statistically significant
The trend in annual rainfall as shown by the
trends.
Prcptot index in Figure 7 varies across the region,
Changes in the frequency of high rainfall days is
decreasing in the northeast and central zones, and
shown by the changes in the indices R20mm and
increasing in the NW, central and southwest. The
R10mm in Figure 10: that is, changes in the
amount of annual rainfall coming from days of
number of days with more than 20mm and 10mm.
extreme rainfall is measured by the indices
In the northern half of the CS, there is a decrease
R95ptot and R99ptot, and trends of this rainfall
in frequency in the east and an increase in
are shown in Figure 8. There are few statistically
frequency of these days in the west. There are no
significant trends although the consistent direction
statistically significant changes in the southern
and magnitude of the decreases at Walgett,
part of the CS.
Warwick and Gunnedah are significant. Also the
increases in the north west of the region are
Changes in duration as measured by CDD and CWD
consistent with some significance. These increases
are shown in Figure 11. This is the longest dry
mean that the amount of rainfall coming from
period (measured in days) and the longest wet
extreme daily events is increasing. As per Suppiah
period each year. Splitting the CS region into north
and Hennessy (1998) the variability in the
and south, in the north we see a decrease in the
consistency of trends of the percentile indices is to
length of consecutive wet day in the east and an
be expected. They highlight that the percentile
increase in the west. There is an increase
indices measure different non-linear features of
throughout most of the southern region. The trend
the same variable. Therefore, each index may
Climatology of Extreme Events
20
in consecutive dry days is also increasing across
greatest 1-day rainfall amount per year and the
most of the region except for Gunnedah.
greatest 5 day rainfall amount. The most
statistically significant trend appears to be the
Examining the seasonality of changes in extreme
increase in the maximum 5-day fall across the CS
daily rainfall intensity (Rx1day and Rx5day in figure
region in winter (June-July-August). There is no
12), the most striking features are firstly the lack
similar increase in the greatest 1-day amount in
of statistical significance in the trends, and
winter.
secondly the difference between the trend in the
Prcptot : linear trend between 1960-2010
-25.0
-27.5
direction of trend
increasing
Latitude
decreasing
magnitude of trend
-30.0
100
200
300
-32.5
146
148
150
152
Longitude
Figure 7. The trend in annual rainfall (Prcptot) between 1960-2010.
Climatology of Extreme Events
21
R99ptot: linear trend between 1960-2010
R95ptot: linear trend between 1960-2010
-25.0
-25.0
-27.5
-27.5
magnitude of trend
direction of trend
50
decreasing
Latitude
Latitude
increasing
magnitude of trend
-30.0
100
direction of trend
-30.0
100
increasing
200
decreasing
-32.5
-32.5
146
148
150
146
152
148
150
152
Longitude
Longitude
Figure 8. Trends for the two precipitation indices: total rainfall per year occurring on days with precipitation above the
95th percentile (R95ptot) and 99th percentile (R99ptot) from 1960 to 2010.
SDII: linear trend between 1960-2010
-25.0
-27.5
direction of trend
increasing
Latitude
decreasing
magnitude of trend
2.5
-30.0
5.0
7.5
10.0
-32.5
146
148
150
152
Longitude
Figure 9. Trends for Simple Daily Intensity Index (SDII) from 1960 to 2010.
Climatology of Extreme Events
22
R20mm: linear trend between 1960-2010
R10mm: linear trend between 1960-2010
-25.0
-25.0
-27.5
-27.5
magnitude of trend
2
direction of trend
decreasing
4
Latitude
Latitude
increasing
6
8
-30.0
direction of trend
magnitude of trend
-30.0
5
increasing
10
decreasing
-32.5
-32.5
146
148
150
152
Longitude
146
148
150
152
Longitude
Figure 10. Trends for the two precipitation indices: the number of days per year with rainfall greater than 10 mm
(R10mm) and 20mm (R20mm) from 1960 to 2010.
CDD: linear trend between 1960-2010
CWD: linear trend between 1960-2010
-25.0
-25.0
-27.5
-27.5
direction of trend
magnitude of trend
increasing
1
3
4
-30.0
decreasing
Latitude
Latitude
2
magnitude of trend
-30.0
10
20
direction of trend
30
increasing
decreasing
-32.5
-32.5
146
148
150
152
Longitude
146
148
150
152
Longitude
Figure 11. Trends for the two precipitation indices: cumulative wet days (CWD) and cumulative dry days (CDD) from
1960 to 2010.
Climatology of Extreme Events
23
Seasonal Precipitation Extremes
rx5day: linear trend between 1960-2010 in JJA
rx1day: linear trend between 1960-2010 in JJA
-25.0
-25.0
-27.5
-27.5
magnitude of trend
direction of trend
1
increasing
magnitude of trend
-30.0
2
Latitude
Latitude
decreasing
3
4
-30.0
25
direction of trend
50
increasing
75
decreasing
-32.5
-32.5
146
148
150
146
152
148
150
152
Longitude
Longitude
rx5day: linear trend between 1960-2010 in DJF
rx1day: linear trend between 1960-2010 in DJF
-25.0
-25.0
-27.5
-27.5
magnitude of trend
1
increasing
2
Latitude
decreasing
magnitude of trend
-30.0
25
Latitude
direction of trend
3
4
-30.0
direction of trend
50
increasing
75
decreasing
-32.5
-32.5
146
146
148
150
Longitude
152
148
150
152
Longitude
Figure 12. Trends for the two precipitation indices: the maximum 1-day rainfall amount in the season (Rx1day, top) and
the maximum 5-day amount (Rx5day, bottom) from 1960 to 2010 for JJA (left) and DJF (right).
Climatology of Extreme Events
24
6. Conclusions
This study has examined changes in extreme
days. This effect is stronger in winter than
climate at 18 locations across the Central
summer.
Slopes natural resource management region.

The frequency of warm days is increasing
Trends were computed of indices of extreme
in both winter and summer (stronger in
daily maximum and minimum temperature
winter).
and precipitation from 1960-2010 that capture

The magnitudes of the trends are greater
the essence of changes in duration, intensity,
in the warm end of the distribution of
and frequency of climatic extremes. This data
both
is designed to be used in conjunction with the
temperatures. That is, the warm days and
CSIRO report (Ekstrom 2014) detailing other
nights are warming at a rate greater than
aspects of the historical climate of the CS
the cool day and nights.
region and projections of future climate

change.
maximum
Generally,
most
and
of
minimum
the
minimum
temperature indices were changing at a
greater
rate
than
the
maximum
The following conclusions are made from the
temperature indices as can be seen by
data:
comparing figures 4 and 5.


There is a strong, spatially consistent
extremes, the changes in precipitation
increasing trend in the temperature of
extremes between 1960 and 2010 are not
warm nights.
as spatially coherent, nor are the trends as
o

The
The temperature of warm nights
statistically significant. This is a commonly
are increasing faster in winter
occurring
than summer.
precipitation extremes ((Plummer et al.
frequency
of
warm
nights
is
increasing.

Compared with the indices of temperature
feature
in
studies
of
1999, {Suppiah, 1998 #73)}.

There is a consistency in the trends in the
The temperature of warm and cool days is
northeast: annual rain is increasing and
increasing: warm days faster than cool
the proportion of rain from extreme
events is increasing.
Climatology of Extreme Events
25

Caution needs to be exerted when
comparing results from different studies
on extremes, as the definition of indices
frequently
specific
varies
needs
depending
of
on
the
stakeholders
(as
cautioned by other authors including
Haylock and Nicholls 2000, Collins et al
2000).
Climatology of Extreme Events
26
7. Acknowledgments
We acknowledge and are grateful for the following
data and software:

The following data sources and software.
The Consistent Climate Scenarios Data was
developed by the Queensland Department
of
Science,
Information
Technology,
Innovation and the Arts under funding
from the Department of Agriculture,
Fisheries and Forestry's
'Australia's
Farming Future - Climate Change Research
Program'
(http://www.daff.gov.au/climatechange/a
ustralias-farming-future/climate-changeand-productivity-research)

R Project for Statistical Computing

ggplot. (Wickham 2009)
Climatology of Extreme Events
27
8. Appendix A. Regional Climate Data
The maps of trend of extreme temperature and
we present the timeseries data for each index
precipitation indices presented in Sections 3 and 4
upon which the trends were calculated.
are expanded upon here. For several key locations
8.1.1 Toowoomba
Climatology of Extreme Events
28
Climatology of Extreme Events
29
8.1.2 Border Rivers Maranoa-Balonne (St George)
Climatology of Extreme Events
30
Climatology of Extreme Events
31
8.1.3 Northern Tablelands (Glen Innes)
Climatology of Extreme Events
32
Climatology of Extreme Events
33
8.1.4 North West (Gunnedah)
Climatology of Extreme Events
34
Climatology of Extreme Events
35
8.1.5 Central West (Dubbo)
Climatology of Extreme Events
36
Climatology of Extreme Events
37
9. Appendix B Climate Change Indices
Let TXx be the daily maximum temperatures in
month k, period j. The maximum daily
maximum temperature each month is then:
Definitions of the 27 core indices (source: the ETCCDI
indice webpage:
http://etccdi.pacificclimate.org/list_27_indices.shtml).
Not all of the indices were used in this project.
TXxkj=max(TXxkj)
1.
FD, Number of frost days: Annual count of days
when TN (daily minimum temperature) < 0 oC.
5.
Let TNijbe daily minimum temperature on day i
in year j. Count the number of days where:
Let TNx be the daily minimum temperatures in
month k, period j. The maximum daily
minimum temperature each month is then:
TNij < 0oC.
2.
TR, Number of tropical nights: Annual count of
days when TN (daily minimum temperature) >
20oC.
TNxkj=max(TNxkj)
6.
Let TNijbe daily minimum temperature on day i
in year j. Count the number of days where:
GSL, Growing season length: Annual (1st Jan to
31st Dec in Northern Hemisphere (NH), 1st July
to 30th June in Southern Hemisphere (SH))
count between first span of at least 6 days
with daily mean temperature TG>5oC and first
span after July 1st (Jan 1st in SH) of 6 days with
TG<5oC.
Let TGij be daily mean temperature on day i in
year j. Count the number of days between the
first occurrence of at least 6 consecutive days
with:
TGij > 5oC.
and the first occurrence after 1st July (1st Jan. in
SH) of at least 6 consecutive days with:
TGij < 5oC.
4.
TXx, Monthly maximum value of daily
maximum temperature:
TXn, Monthly minimum value of daily
maximum temperature:
Let TXn be the daily maximum temperatures in
month k, period j. The minimum daily
maximum temperature each month is then:
TNij > 20oC.
3.
TNx, Monthly maximum value of daily
minimum temperature:
TXnkj=min(TXnkj)
7.
TNn, Monthly minimum value of daily minimum
temperature:
Let TNn be the daily minimum temperatures in
month k, period j. The minimum daily
minimum temperature each month is then:
TNnkj=min(TNnkj)
8.
TN10p, Percentage of days when TN < 10th
percentile:
Let TNij be the daily minimum temperature on
day i in period j and let TNin10 be the calendar
day 10th percentile centred on a 5-day window
for the base period 1961-1990. The percentage
of time for the base period is determined
where:
TNij < TNin10
Climatology of Extreme Events
38
To avoid possible inhomogeneity across the inbase and out-base periods, the calculation for
the base period (1961-1990) requires the use
of a bootstrap process. Details are described in
Zhang et al. (2005) .
9.
TX10p, Percentage of days when TX < 10th
percentile:
Let TXij be the daily maximum temperature on
day i in period j and let TXin10 be the calendar
day 10th percentile centred on a 5-day window
for the base period 1961-1990. The percentage
of time for the base period is determined
where:
TXij < TXin10
To avoid possible inhomogeneity across the inbase and out-base periods, the calculation for
the base period (1961-1990) requires the use
of a bootstrap process. Details are described in
Zhang et al. (2005) .
10. TN90p, Percentage of days when TN > 90th
percentile:
Let TXij be the daily maximum temperature on
day i in period j and let TXin90 be the calendar
day 90th percentile centred on a 5-day window
for the base period 1961-1990. The percentage
of time for the base period is determined
where:
TXij > TXin90
To avoid possible inhomogeneity across the inbase and out-base periods, the calculation for
the base period (1961-1990) requires the use
of a bootstrap processes. Details are described
in Zhang et al. (2005) .
12. WSDI, Warm spell duration index: Annual
count of days with at least 6 consecutive days
when TX > 90th percentile
Let TXij be the daily maximum temperature on
day i in period j and let TXin90 be the calendar
day 90th percentile centred on a 5-day window
for the base period 1961-1990. Then the
number of days per period is summed where,
in intervals of at least 6 consecutive days:
TXij > TXin90
Let TNij be the daily minimum temperature on
day i in period j and let TNin90 be the calendar
day 90th percentile centred on a 5-day window
for the base period 1961-1990. The percentage
of time for the base period is determined
where:
TNij > TNin90
To avoid possible inhomogeneity across the inbase and out-base periods, the calculation for
the base period (1961-1990) requires the use
of a bootstrap process. Details are described in
Zhang et al. (2005) .
11. TX90p, Percentage of days when TX > 90th
percentile:
13. CSDI, Cold spell duration index: Annual count
of days with at least 6 consecutive days when
TN < 10th percentile
Let TNij be the daily maximum temperature on
day i in period j and let TNin10 be the calendar
day 10th percentile centred on a 5-day window
for the base period 1961-1990. Then the
number of days per period is summed where,
in intervals of at least 6 consecutive days:
TNij < TNin10
14. DTR, Daily temperature range: Monthly mean
difference between TX and TN
Climatology of Extreme Events
39
Let TXij and TNij be the daily maximum and
minimum temperature respectively on day i in
period j. If I represents the number of days in j,
then:
amount on day i in period j. Count the number
of days where:
RRij ≥ 10mm
19. R20mm Annual count of days when PRCP≥
20mm: Let RRij be the daily precipitation
amount on day i in period j. Count the number
of days where:
RRij ≥ 20mm
15. Rx1day, Monthly maximum 1-day
precipitation:
Let RRij be the daily precipitation amount on
day i in period j. The maximum 1-day value for
period j are:
Rx1dayj = max (RRij)
16. Rx5day, Monthly maximum consecutive 5-day
precipitation:
Let RRkj be the precipitation amount for the 5day interval ending k, period j. Then maximum
5-day values for period j are:
Rx5dayj = max (RRkj)
17. SDII Simple precipitation intensity index: Let
RRwj be the daily precipitation amount on wet
days, w (RR ≥ 1mm) in period j. If W represents
number of wet days in j, then:
18. R10mm Annual count of days when PRCP≥
10mm: Let RRij be the daily precipitation
20. Rnnmm Annual count of days when PRCP≥
nnmm, nn is a user defined threshold: Let RRij
be the daily precipitation amount on day i in
period j. Count the number of days where:
RRij ≥ nnmm
21. CDD. Maximum length of dry spell, maximum
number of consecutive days with RR < 1mm:
Let RRij be the daily precipitation amount on
day i in period j. Count the largest number of
consecutive days where:
RRij < 1mm
22. CWD. Maximum length of wet spell, maximum
number of consecutive days with RR ≥ 1mm:
Let RRij be the daily precipitation amount on
day i in period j. Count the largest number of
consecutive days where:
RRij ≥ 1mm
23. R95pTOT. Annual total PRCP when RR > 95p.
Let RRwj be the daily precipitation amount on a
wet day w (RR ≥ 1.0mm) in period i and let
RRwn95 be the 95th percentile of precipitation
on wet days in the 1961-1990 period. If W
represents the number of wet days in the
period, then:
Climatology of Extreme Events
40
24. R99pTOT. Annual total PRCP when RR > 99p:
Let RRwj be the daily precipitation amount on a
wet day w (RR ≥ 1.0mm) in period i and let
RRwn99 be the 99th percentile of precipitation
on wet days in the 1961-1990 period. If W
represents the number of wet days in the
period, then:
25. PRCPTOT. Annual total precipitation in wet
days: Let RRij be the daily precipitation amount
on day i in period j. If I represents the number
of days in j, then
Climatology of Extreme Events
41
10. References
Alexander LV, Arblaster JM (2009) Assessing trends
in observed and modelled climate
extremes over Australia in relation to
future projections. International Journal of
Climatology
29
(3):417-435.
doi:10.1002/joc.1730
Alexander LV, Zhang X, Peterson TC, Caesar J,
Gleason B, Klein Tank AMG, Haylock M,
Collins D, Trewin B, Rahimzadeh F,
Tagipour A, Rupa Kumar K, Revadekar J,
Griffiths G, Vincent L, Stephenson DB, Burn
J, Aguilar E, Brunet M, Taylor M, New M,
Zhai P, Rusticucci M, Vazquez-Aguirre JL
(2006) Global observed changes in daily
climate extremes of temperature and
precipitation. Journal of Geophysical
Research: Atmospheres 111 (D5):D05109.
doi:10.1029/2005JD006290
Braganza K, Karoly DJ, Arblaster J (2004) Diurnal
temperature range as an index of global
climate change during the twentieth
century. Geophysical Research Letters 31
(13)
Burgess S, Ricketts J, Panjkov A, Carter J, Day K
(2012) Consistent Climate Scenarios
Project User Guide: ‘Change factor’ and
‘Quantile-matching’
based
climate
projections data.
Collins D, Della-Marta P, Plummer N, Trewin B
(2000) Trends in annual frequencies of
extreme temperature events in Australia.
Australian Meteorological Magazine 49
(4):277-292
Frich P, Alexander L, Della-Marta P, Gleason B,
Haylock M, Klein Tank A, Peterson T (2002)
Observed coherent changes in climatic
extremes during the second half of the
twentieth century. Climate Research 19
(3):193-212
Gallant A, Kiem A, Verdon-Kidd D, Stone R, Karoly
D, de Michele C (2012) Understanding
hydroclimate processes in the MurrayDarling Basin for natural resources
management. Hydrology & Earth System
Sciences 16 (7)
Haylock M, Nicholls N (2000) Trends in extreme
rainfall indices for an updated high quality
data set for Australia, 1910-1998.
International Journal of Climatology 20
(13):1533-1541
Jeffrey SJ, Carter JO, Moodie KB, Beswick AR
(2001) Using spatial interpolation to
construct a comprehensive archive of
Australian climate data. Environmental
Modelling & Software 16 (4):309-330
Keating BA, Carberry PS, Hammer GL, Probert ME,
Robertson MJ, Holzworth D, Huth NI,
Hargreaves JN, Meinke H, Hochman Z
(2003) An overview of APSIM, a model
designed for farming systems simulation.
European Journal of Agronomy 18 (3):267288
Klingaman NP, Woolnough S, Syktus J (2013) On
the drivers of inter‐annual and decadal
rainfall variability in Queensland, Australia.
International Journal of Climatology 33
(10):2413-2430
Manton M, Della‐Marta P, Haylock M, Hennessy K,
Nicholls N, Chambers L, Collins D, Daw G,
Finet A, Gunawan D (2001) Trends in
extreme daily rainfall and temperature in
Southeast Asia and the South Pacific:
1961–1998. International Journal of
Climatology 21 (3):269-284
Marshall A, Hudson D, Wheeler M, Alves O,
Hendon H, Pook M, Risbey J (2013) Intraseasonal drivers of extreme heat over
Australia in observations and POAMA-2.
Climate Dynamics:1-23
Plummer N, Lin Z, Torok S (1995) Trends in the
diurnal temperature range over Australia
since 1951. Atmospheric Research 37
(1):79-86
Climatology of Extreme Events
42
Plummer N, Salinger MJ, Nicholls N, Suppiah R,
Hennessy KJ, Leighton RM, Trewin B, Page
CM, Lough JM (1999) Changes in climate
extremes over the Australian region and
New Zealand during the twentieth
century. In:
Weather and Climate
Extremes. Springer, pp 183-202
Trewin BC (2001) Extreme temperature events in
Australia.
Wickham H (2009) ggplot2: elegant graphics for
data analysis. Springer,
Zhang X, Alexander L, Hegerl GC, Jones P, Tank AK,
Peterson TC, Trewin B, Zwiers FW (2011)
Indices for monitoring changes in
extremes based on daily temperature and
precipitation data. Wiley Interdisciplinary
Reviews: Climate Change 2 (6):851-870
Climatology of Extreme Events
43
Contact Details
Contact person: Dr Allyson Williams
Telephone: +61 7 4631 5415
Email address: acsc@usq.edu.au
Web address: web: http://www.usq.edu.au/acsc
Climatology of Extreme Events
44
Climatology of Extreme Events
45
Climatology of Extreme Events
46