1. Introduction
Cutoff lows are cold-cored cyclonic circulations, which develop in the mid- to upper troposphere and become completely detached from the midlatitude westerly jet (e.g., Pinheiro et al. 2017). The presence of a pool of cold air indicates that the system has originated at a higher latitude, and subsequently moved toward the equator (Nieto et al. 2005; Risbey et al. 2009a, and others). Globally, the study of cutoff lows has been motivated by their association with a range of impacts, including their potential to deliver high volumes of precipitation (Qi et al. 1999; Nieto et al. 2005; Favre et al. 2012; and others), the transfer of stratospheric air and ozone into the troposphere (Fuenzalida et al. 2005; Ndarana and Waugh 2010), and the potential for explosive cyclogenesis at the surface (Qi et al. 1999; Fuenzalida et al. 2005; Browning and Goodwin 2013).
Over Australia, cutoff lows are a key source of rainfall for the southeast region, especially during the cool season (April–October inclusive; Pook et al. 2006, hereafter P06). Southeastern Australia (SEA) is one of the most agriculturally productive regions in the country (Risbey et al. 2009a), and its output depends strongly on both the amount and timing of rainfall received during the growing season. Studying the Mallee region (Fig. 1) of northwestern Victoria over the years 1970–2002, P06 found that cutoff lows deliver 51% of cool-season rainfall, and, perhaps more importantly, 80% of rainfall events of daily totals of 25 mm or more. Heavy rain events of this nature are especially important to grain growers, so much so that they name the first heavy rain event of autumn the “autumn break,” which enables them to sow their winter crops (McIntosh et al. 2007; Pook et al. 2009). Extending this analysis to the period 1956–2009, Pook et al. (2014) reported similar results.
Map of the Australian region showing cutoff low detection domain (blue rectangle) and approximate locations of the Mallee (orange rectangle) and Snowy Mountains (red rectangle) regions.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Map of the Australian region showing cutoff low detection domain (blue rectangle) and approximate locations of the Mallee (orange rectangle) and Snowy Mountains (red rectangle) regions.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Map of the Australian region showing cutoff low detection domain (blue rectangle) and approximate locations of the Mallee (orange rectangle) and Snowy Mountains (red rectangle) regions.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
For the separate study region of the Snowy Mountains, also in SEA (Fig. 1), Chubb et al. (2011) reported that cutoff lows deliver 56% of the precipitation between May and September. Addressing heavy precipitation events specifically, Theobald et al. (2015) found cutoff lows to be associated with 53% of days of a precipitation total of at least 10 mm, while Fiddes et al. (2015) found cutoff lows responsible for 44% of extreme precipitation (95th percentile) days. Precipitation in the Snowy Mountains is of additional importance as the region is a catchment of water for irrigated agriculture downstream as well as for local hydroelectricity generation, and for the winter ski industry (Chubb et al. 2011; Theobald et al. 2015; Fiddes et al. 2015).
While the average annual rainfall in SEA is sufficient for agriculture, interannual variability is known to be high. For instance, the Mallee region receives on average about 380 mm rainfall per year; however, the annual rainfall standard deviation is as large as 100 mm (Fig. 2). It is possible, too, for individual months to be exceptionally dry or wet within a year of relatively average total rainfall. Indeed, the variability of rainfall from cutoff lows has been identified as a large component of this overall variability of total rainfall (McIntosh et al. 2007). It seems clear, then, that understanding the variability of cutoff lows is vital to understanding rainfall variability in SEA.
Annual rainfall totals for Ouyen, in the Mallee region shown in Fig. 1. The solid line shows the mean annual rainfall over the period 1915–2019; dashed lines show mean ±1 standard deviation. Note that rainfall data for the years 1924 and 1926 were missing and were excluded from this analysis. Data were sourced from the Bureau of Meteorology Climate Data Online.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Annual rainfall totals for Ouyen, in the Mallee region shown in Fig. 1. The solid line shows the mean annual rainfall over the period 1915–2019; dashed lines show mean ±1 standard deviation. Note that rainfall data for the years 1924 and 1926 were missing and were excluded from this analysis. Data were sourced from the Bureau of Meteorology Climate Data Online.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Annual rainfall totals for Ouyen, in the Mallee region shown in Fig. 1. The solid line shows the mean annual rainfall over the period 1915–2019; dashed lines show mean ±1 standard deviation. Note that rainfall data for the years 1924 and 1926 were missing and were excluded from this analysis. Data were sourced from the Bureau of Meteorology Climate Data Online.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
The climate of SEA has long been known to be influenced by remote drivers, and many studies have sought to quantify the relationships between the drivers and seasonal or annual rainfall totals. These remote drivers include El Niño–Southern Oscillation (ENSO) (e.g., Gallant et al. 2007; Risbey et al. 2009b), the Indian Ocean dipole (IOD) (e.g., Ashok et al. 2003; Risbey et al. 2009b; Ummenhofer et al. 2009, 2011), and the southern annular mode (SAM) (e.g., Meneghini et al. 2007; Hendon et al. 2007; Risbey et al. 2009b; Nicholls 2010). El Niño events are associated with suppressed rainfall and La Niña events are associated with enhanced rainfall in SEA, with maximum correlations between the Southern Oscillation index and rainfall found in austral winter and spring (Risbey et al. 2009b). Similarly, positive IOD events are associated with reduced rainfall and negative IOD events are associated with increased rainfall in SEA, from June to November. Positive SAM events exhibit some association with reduced rainfall and negative SAM events are associated with increased rainfall in SEA during austral winter, while in austral spring the relationship reverses, with enhanced rainfall across eastern Australia observed in association with positive SAM events. These drivers are also known to interact with each other (especially ENSO and the IOD), and their relationships with Australian rainfall is also seen to vary over longer (decadal) time scales (Risbey et al. 2009b). We believe that by studying the possible relationships between the climate drivers and cutoff lows, we might gain new insights into how the remote climate drivers influence rainfall variability in SEA.
To our knowledge, a study presenting the correlations between cutoff low occurrence (number of systems and/or system days) in the SEA region and the remote drivers (ENSO, IOD, or SAM) has not been done. Risbey et al. (2009a) studied relationships between measures of cutoff low intensity and the remote drivers, but not cutoff low frequency as such. We are not aware of any other Australian-based study that addresses relationships between cutoff low occurrence and the climate drivers. On the other hand, studies of the relationships between cutoff low occurrence at the hemispheric scale and the remote drivers have reported contrasting results in some cases. Fuenzalida et al. (2005) find no relationship between annual numbers of cutoff low systems and ENSO. Similarly, Muñoz et al. (2020) find no significant correlations between numbers of systems and either ENSO or the SAM. In contrast, Favre et al. (2012) find higher numbers of systems during La Niña events than El Niño. Additionally, Favre et al. (2012) find a northward shift in the mean track of cutoff low systems over eastern Australia during La Niña (see Fig. 13 therein). It would be of some value to see if any of these relationships (i.e., greater system frequency during La Niña versus El Niño, and northward shift in system tracks during La Niña versus El Niño, or no relationship at all) are evident in the SEA region.
The synoptic rainfall studies discussed previously (i.e., P06, Chubb et al. 2011; Fiddes et al. 2015; Theobald et al. 2015, Pook et al. 2014) have largely employed the traditional synoptic decomposition methodology, namely obtaining a dataset of daily rainfall (and sometimes snowfall) amounts, and for each precipitation event, identifying the synoptic atmospheric system associated with the rainfall, according to some criteria. This implies that only weather systems that generated rainfall were recorded and studied. It is noted that P06 did include non-rain-bearing cutoff lows in their dataset, but these were only identified on the remaining days when no rain was recorded in their study region. It remains theoretically possible for a rain-bearing system of a different type (e.g., cold front) to be recorded for a given day, while a cutoff low system being present elsewhere in the domain. To our knowledge a climatology of all cool-season cutoff lows, regardless of the occurrence of precipitation, has not yet been done.
In this study, we attempt to conduct an identification of cool-season cutoff lows at the synoptic scale, by an automated method (i.e., tracking scheme). Aside from the efficiency of detection of weather systems, automated detection methods also carry the advantage of pinpointing the location of the system, as well as separating one system from the next. This enables the development of each system across its lifetime to be examined, and the duration of each individual system, movement speed, and the track of each system to be studied.
The aims of this study, then, are to
develop a tracking scheme to identify cutoff lows and to explore the climatology of cool-season cutoff low systems over SEA;
compare the climatological characteristics of the identified systems between two reanalyses (NCEP1 and ERA-Interim); and
study the variability of cutoff lows associated with the known remote drivers of SEA climate, namely the ENSO, the IOD and the SAM.
This paper is set out as follows: section 2 describes the methodology and datasets, including the reanalysis and rainfall datasets used in this study, cutoff low tracking scheme and identification criteria, and indices to represent the remote climate drivers; section 3 presents the results, including the general climatology of identified systems (average number and duration of systems, tracks of systems, and associated precipitation), comparisons between the NCEP1 and ERA-Interim reanalyses, and correlations between system frequency, duration and tracks, and the remote climate drivers; and section 4 provides a discussion and conclusion of the main findings.
2. Methods and data
a. Reanalysis and rainfall datasets
Cutoff lows are primarily upper-level, synoptic scale systems (e.g., Nieto et al. 2005), and as such we conduct our identification on 6-hourly geopotential height at the 500-hPa level. It is acknowledged that some studies of cutoff lows over SEA include systems at the sea level in their classification (e.g., P06; Chubb et al. 2011; Fiddes et al. 2015; Theobald et al. 2015); however, the majority of studies of cutoff lows at the hemispheric level identify cutoff lows as systems of upper-level origin (e.g., Fuenzalida et al. 2005; Reboita et al. 2010; Favre et al. 2012; Pinheiro et al. 2017; and others) and this is the convention we follow here. The thermal signature of cutoff lows is assessed using 6-hourly 500–1000-hPa thickness data. These data are obtained from the NCEP1 reanalysis (Kalnay et al. 1996) and the ERA-Interim reanalysis (Dee et al. 2011). It is acknowledged that the NCEP1 reanalysis has been found to be less suited to the study of midlatitude synoptic systems in the Southern Hemisphere than more modern reanalyses (Hines et al. 2000; Hartmann et al. 2013), so the more modern reanalysis (ERA-Interim) was also chosen for analysis and comparison. Data from both reanalyses are used at a horizontal grid spacing of 2.5° latitude–longitude, with the higher resolution (0.75° latitude–longitude) ERA-Interim data regridded to this spacing using a bilinear interpolation.
Precipitation associated with cutoff low systems is represented with data from the Bureau of Meteorology: Australian Water Availability Project (AWAP), a gridded dataset based on rainfall gauge observations from the Bureau of Meteorology (Jones et al. 2009; see Fig. 2a therein for the rainfall gauge network). Cutoff low precipitation is aggregated during the days on which a cutoff low was identified. A precipitation composite is then computed by calculating the proportion of the aggregated cutoff low precipitation compared to the corresponding values of AWAP precipitation across the cool season as a whole.
Cutoff low centers are identified within a spatial region bound by 27.5°S and 47.5°S, and 125°E and 152.5°E (see Fig. 1). The time period of identification covers the months April–October inclusive, for the years 1979–2018 inclusive. It is acknowledged that some studies of cutoff lows over SEA begin their climatology earlier (e.g., P06; Theobald et al. 2015); however, ERA-Interim reanalysis begins in 1979, so it is necessary to choose 1979 as the start of our analysis to compare results between ERA-Interim and NCEP1 reanalyses. This study period includes the full extent of the Millennium drought and subsequent record-breaking wet years of 2010 and 2011 (CSIRO 2012 and references therein).
b. Cutoff low identification
Since cutoff lows are closed cyclonic circulations with a cold core, we have two defining features to base our identification criteria on. Our tracking scheme is therefore divided into three stages:
Stage 1: Identification of closed low centers in individual 500-hPa geopotential height fields.
Stage 2: Joining of low centers into tracks across multiple time steps and filtering out insignificant systems.
Stage 3: Identification of a cold core or trough in the 500–1000-hPa thickness field during the low track.
Stage 1 detects closed low centers in 500-hPa geopotential height data by scanning each point and calculating the difference in geopotential height between the central point and the surrounding eight points. A point is identified as a closed low center if it has a lower geopotential height than each of the surrounding eight points one and two grid spaces away. A local minimum of geopotential found in this way indicates a closed contour of geopotential height around the center, and hence a cyclonic circulation. The mean value of the differences in geopotential height between the center and the eight points two grid spaces away was also computed (denoted test 2; to be utilized in postidentification filtering). Note that the diagonal differences of geopotential height were divided by a factor of
Configuration of detection tests for (a) closed lows identified in stage 1, (b) cold troughs identified in stage 3, and (c) the domain surrounding a closed low center within which a cold trough is searched for.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Configuration of detection tests for (a) closed lows identified in stage 1, (b) cold troughs identified in stage 3, and (c) the domain surrounding a closed low center within which a cold trough is searched for.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Configuration of detection tests for (a) closed lows identified in stage 1, (b) cold troughs identified in stage 3, and (c) the domain surrounding a closed low center within which a cold trough is searched for.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
In stage 2, weaker secondary lows around a deeper parent low are removed by iterating through each low center in a given field and removing all other low centers of a shallower depth within a distance of 10° latitude–longitude. In situations where two low centers in the same time step had equal central depth, the low center with weaker test 2 geopotential height gradient was eliminated. Low centers are then joined together into tracks by linking them to the track of the closest low center in the previous time step, if one exists within a spatial distance of 7.5° latitude–longitude. If no previous low centers are found within these bounds, a new event track is assigned. Events of duration less than 24 h, or that do not achieve a test 2 geopotential height gradient of 15 gpm during at least 1 time step, are then discarded as insignificant systems.
Stage 3 identifies a cold core or trough in the vicinity of the closed low centers. It is acknowledged that many studies require a complete cold core to identify a cutoff low (e.g., Fuenzalida et al. 2005; Favre et al. 2012; Pinheiro et al. 2017, and others); however, our study is motivated by a goal to understand rainfall variability in SEA, and it has been shown that systems with only a cold trough also make a meaningful contribution to cool-season precipitation in SEA (P06). Therefore, we do not distinguish between a cold core or a cold trough, simply identifying a cold trough where a point in the 500–1000-hPa thickness field meets both of the following conditions:
The central point has a lower thickness than each of the five points one grid space to the west, northwest, north, northeast, and east.
The central point has a lower thickness than each of the five points two grid spaces to the west, northwest, north, northeast, and east, by a minimum threshold of at least 8 gpm.
This feature is searched for within a domain 7.5° latitude to the north, 2.5° latitude to the south, and 10° longitude to the west of the closed low center. These tests are illustrated in Figs. 3b,c. Closed low tracks exhibiting this thermal feature during at least two consecutive points of their duration are finally recorded as a cutoff low.
c. Climate driver indices
The ENSO is represented by the oceanic Niño index (ONI), defined as the smoothed (3-month running mean) area averaged SST anomaly in the Niño-3.4 region (5°S–5°N, 170°–120°W) (Trenberth 1997). The ONI is computed with SST data from the Extended Reconstructed Sea Surface Temperature (ERSST) version 5 dataset (Huang et al. 2017).
The IOD is represented by the dipole mode index (DMI), defined as the anomaly of the SST gradient between the western equatorial Indian Ocean (50°–70°E and 10°S–10°N) and the south eastern equatorial Indian Ocean (90°–110°E and 10°S–0°N) (Saji et al. 1999). The DMI is also computed with ERSST version 5 data (Huang et al. 2017).
The SAM is represented with the Antarctic Oscillation (AAO) index, which is computed by projecting the 700-hPa geopotential height anomaly poleward of 20°S onto the leading empirical orthogonal function (EOF) of monthly mean 700-hPa height for the same region (Mo 2000).
3. Results
a. General climatology and comparison between reanalyses
Cool-season mean values of number of cutoff low systems and total system duration (system days), and their associated standard deviations, are presented in Table 1. For the years 1979–2018 we identify noticeably more cutoff low systems in the ERA-Interim reanalysis than in the NCEP1. In NCEP1 data we identify 640 systems spanning a total of 1495.5 days, while in ERA-Interim data we identify 866 systems spanning a total of 1963.5 days. Standard deviations of systems per year and systems days per year are approximately equal for both NCEP1 and ERA-Interim reanalyses, indicating a similar range of interannual variability of cutoff lows in both reanalyses.
Basic climatological statistics of cutoff lows identified in both NCEP1 and ERA-Interim reanalysis, over the whole cool season (April–October inclusive) for each year from 1979 to 2018.
The month-to-month variability of number of systems and system days are presented in Fig. 4. Both reanalyses exhibit a similar moderate monthly cycle of cutoff low frequency and duration over the cool season. Numbers of systems and system days are seen to rise strongly to a peak value in June/July, then decline to minimum values in September/October (though variability about the mean values for each month is fairly large). Additionally, though there are 226 more systems found in ERA-Interim than NCEP1 in total (Table 1), it can be seen in Fig. 4 that the there is a high degree of overlap in the variability of systems across all months of the cool season. This gives us confidence that both NCEP1 and ERA-Interim reanalyses provide a similar representation of cutoff low systems in these months.
Numbers of (a) cutoff low systems and (b) cutoff low system days per month over the cool season, from NCEP1 (blue) and ERA-Interim (red) reanalyses. Mean values shown by solid shapes, and error bars show ±1 standard deviation.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Numbers of (a) cutoff low systems and (b) cutoff low system days per month over the cool season, from NCEP1 (blue) and ERA-Interim (red) reanalyses. Mean values shown by solid shapes, and error bars show ±1 standard deviation.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Numbers of (a) cutoff low systems and (b) cutoff low system days per month over the cool season, from NCEP1 (blue) and ERA-Interim (red) reanalyses. Mean values shown by solid shapes, and error bars show ±1 standard deviation.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Mean tracks of the identified cutoff lows are presented in Fig. 5. These tracks are centered largely between −35° and −37.5° latitude, but again with substantial variability across each longitude. Tracks are located somewhat farther to the north over the landmass of SEA in the east of the search domain. Little variation in the mean tracks from month-to-month was evident in either reanalysis (Table 2). Also note that this describes the tracks of cutoff lows in aggregate individual systems may move from west to east following the progression of mid latitude weather systems in general, they may remain stationary for a period of their duration, or even move westward, as shown in Fig. 6.
Mean tracks of cutoff low systems in NCEP1 (blue) and ERA-Interim (red) reanalyses data. Solid lines denote mean track in 500-hPa geopotential height fields, with dots denoting the associated standard deviations.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Mean tracks of cutoff low systems in NCEP1 (blue) and ERA-Interim (red) reanalyses data. Solid lines denote mean track in 500-hPa geopotential height fields, with dots denoting the associated standard deviations.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Mean tracks of cutoff low systems in NCEP1 (blue) and ERA-Interim (red) reanalyses data. Solid lines denote mean track in 500-hPa geopotential height fields, with dots denoting the associated standard deviations.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Tracks of a random sample of 20 individual cutoff low systems in (a) NCEP1 and (b) ERA-Interim reanalyses data. Note the westward progression of systems around 35°S, 145°E in (a), and around 37.5°S, 137.5°E in (b).
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Tracks of a random sample of 20 individual cutoff low systems in (a) NCEP1 and (b) ERA-Interim reanalyses data. Note the westward progression of systems around 35°S, 145°E in (a), and around 37.5°S, 137.5°E in (b).
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Tracks of a random sample of 20 individual cutoff low systems in (a) NCEP1 and (b) ERA-Interim reanalyses data. Note the westward progression of systems around 35°S, 145°E in (a), and around 37.5°S, 137.5°E in (b).
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Mean and standard deviation (parentheses) of system latitude and longitude values, across each month, for each year from 1979 to 2018.
The distribution of system longitudes are presented in Fig. 7. A peak in the number of systems is found around 145°E, in both reanalyses, as shown in Fig. 7a. The number of system days occurring at each longitude shows a similar peak around this latitude, with more than twice as many system days occurring at 145°E than at the western limit of the study region at 125°E (Fig. 7b). Also shown is a slightly higher percentage of systems east of 145°E in NCEP1 data, and a slightly higher percentage of systems west of 145°E in ERA-Interim data.
Percentage of (a) cutoff low systems and (b) system days occurring at each longitude in NCEP1 (blue) and ERA-Interim (red) reanalyses.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Percentage of (a) cutoff low systems and (b) system days occurring at each longitude in NCEP1 (blue) and ERA-Interim (red) reanalyses.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Percentage of (a) cutoff low systems and (b) system days occurring at each longitude in NCEP1 (blue) and ERA-Interim (red) reanalyses.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
The distributions of individual system durations are presented in Fig. 8. It can be seen that the most common system duration is 1.25 days, or five reanalysis time steps, in both reanalyses. Also, the greatest difference in number of cutoff lows identified in the NCEP1 and ERA-Interim reanalyses occurs in systems of short to moderate duration, as seen in Fig. 8a. The relative proportions of system durations, though, remain more or less evenly distributed, as seen in Fig. 8b.
(a) Number and (b) percentage of cutoff low systems of a given duration in NCEP1 (blue) and ERA-Interim (red) reanalyses.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
(a) Number and (b) percentage of cutoff low systems of a given duration in NCEP1 (blue) and ERA-Interim (red) reanalyses.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
(a) Number and (b) percentage of cutoff low systems of a given duration in NCEP1 (blue) and ERA-Interim (red) reanalyses.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Composites of precipitation amounts associated with cutoff lows, and the proportion of total cool-season precipitation associated with cutoff lows, are presented in Fig. 9. Cutoff lows identified in the NCEP1 reanalysis coincide with around 35% of precipitation over most of SEA during the cool season (Fig. 9a), while proportions of cool-season rainfall coinciding with cutoff lows identified in the ERA-Interim reanalysis are roughly 5% higher across the region (Fig. 9b). Highest proportions of precipitation coinciding with cutoff lows are found in the northeast of the region.
Precipitation amount (colors) from cutoff lows, and proportion of cutoff low precipitation to total cool-season precipitation (contours), for cutoff lows in (a) NCEP1 and (b) ERA-Interim reanalyses. Rectangles show approximate locations of the Mallee (orange) and Snowy Mountains (red rectangle) regions. The coastline is marked in white. Precipitation data from AWAP dataset (Jones et al. 2009).
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Precipitation amount (colors) from cutoff lows, and proportion of cutoff low precipitation to total cool-season precipitation (contours), for cutoff lows in (a) NCEP1 and (b) ERA-Interim reanalyses. Rectangles show approximate locations of the Mallee (orange) and Snowy Mountains (red rectangle) regions. The coastline is marked in white. Precipitation data from AWAP dataset (Jones et al. 2009).
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Precipitation amount (colors) from cutoff lows, and proportion of cutoff low precipitation to total cool-season precipitation (contours), for cutoff lows in (a) NCEP1 and (b) ERA-Interim reanalyses. Rectangles show approximate locations of the Mallee (orange) and Snowy Mountains (red rectangle) regions. The coastline is marked in white. Precipitation data from AWAP dataset (Jones et al. 2009).
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
b. Associations with remote climate drivers
Using cutoff low databases from both reanalyses, cool-season totals of number of systems and system days, along with mean system latitude and longitude, are computed. Scatterplots of these quantities against mean cool-season values of the ONI, DMI and AAO are then created, along with Pearson’s linear correlation coefficients (r) and Student’s t-test probability (p) values (Figs. 10–13). Two correlations that are possibly significant at the 95% significance level are revealed in this way: a weak (r = 0.28) but significant (p = 0.04) correlation between the number of systems per cool season in the ERA-Interim reanalysis, and the AAO (Fig. 10f), and a weak (r = 0.28 also) but significant (p = 0.04 also) correlation between the mean system latitude in the ERA-Interim reanalysis, and the AAO (Fig. 12f). To investigate these correlations further this process is repeated for the individual months April–October. It is found that the correlation between number of cutoff low systems in the ERA-Interim reanalysis and the AAO are significant only during the month of August (r = 0.3, p = 0.03), while there are no individual months where the correlation between the mean system latitude and the AAO is found to be significant in this way (not shown). Some of the other monthly cutoff low measures also showed statistically significant correlation results with the driver indices in this way (i.e., p < 0.05), but again only during a single, isolated month (not shown). Further, the spread and shape of data on each of the monthly scatterplots was seen to vary markedly from one month to the next.
Scatterplots of cool-season total number of systems in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season total number of systems in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season total number of systems in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season number of system days in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season number of system days in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season number of system days in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season mean system latitude in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season mean system latitude in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season mean system latitude in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season mean system longitude in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season mean system longitude in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
Scatterplots of cool-season mean system longitude in (left) NCEP1 and (right) ERA-Interim reanalyses, and the (a),(b) ONI; (c),(d) DMI; and (e),(f) AAO. Linear regression lines are also shown, along with r and p values. Correlations significant at the 5% level are shown in bold.
Citation: Monthly Weather Review 149, 12; 10.1175/MWR-D-21-0142.1
4. Discussion and conclusions
The first objective was to develop an automated algorithm to identify cutoff lows, and to explore the climatology of cool-season cutoff lows over SEA. The average number of system days identified with our tracking scheme (37.4 in NCEP1 data, Table 1) is somewhat lower than the average number of system days during the cool season presented in other studies. P06 reports 45.9 cutoff low days per cool season; however, this includes systems in both 500-hPa and MSLP data. In an earlier study, Qi et al. (1999) reports 43.1 days per cool season (derived using the relative percentages for areas 2, 3 and 4, which is similar to our study region, across the months April–October). However, little information is given regarding the identification criteria and data utilized in Qi et al. (1999). Also note that the analysis period covered in each of these studies differs: P06 covers the years 1970–2002 and Qi et al. (1999) covers the years 1983–96, while our study covers the years 1979–2018. Finally, our identification is limited to systems of at least 24-h duration, and it can be seen in Fig. 8 that if our study included systems of shorter duration, a substantial increase in total system numbers might result. The seasonal cycle of cutoff low frequency, meanwhile, is similar, though not the same. P06 reports maximum cutoff low days in May, while the maximum of system days found in our study is in June (for NCEP1 data, Fig. 4) or July (for ERA-Interim data, Fig. 4). Qi et al. (1999) also find a maximum in June, though they also find similarly high numbers across the other cool-season months.
The precipitation associated with cutoff lows identified with our tracking scheme is notably lower than results reported in previous studies, particularly in the results from the NCEP1 reanalysis. Cutoff lows in NCEP1 data identified with our tracking scheme are associated with roughly 35% of cool-season precipitation in the Mallee region, and 40% of cool-season precipitation in the Snowy Mountains (Fig. 9). As mentioned in the introduction, previous studies of cool-season precipitation in the Mallee and Snowy Mountains have found 51% (P06) and 56% (Chubb et al. 2011) associated with cutoff lows, respectively. Again, though, both of these studies include systems occurring at the surface level as well as at the 500-hPa level, resulting in a higher number of systems identified and hence greater percentage of precipitation events covered. Alternatively, a reduction in the amount of precipitation per cutoff low event has been reported as a recent trend for both the Mallee (Risbey et al. 2012) and Snowy Mountains (Chubb et al. 2011) regions, and since our study extends to the year 2018, this trend could have also contributed to the lower precipitation percentages found here.
To gain some indication of the sensitivity of our climatological results to the choice of identification parameters, the identification was repeated over the same time period and both reanalyses, but with the required minimum geopotential height gradient reduced to 0 gpm (though a lower geopotential height at the system center compared to each of the test 1 and 2 points was still required) and also the required thickness gradient of the cold trough reduced to 0 gpm (though a lower thickness at the center of the trough compared to each of the test 1 and 2 points was still required). The resulting values of cool-season mean values of number of systems and system duration (system days), and their associated standard deviations, are presented in Table 3. From this it can be seen that a greater number of systems and system days are identified in both reanalyses using these reduced criteria, though the increase in numbers is not outside the range of numbers reported in other studies such as those discussed above. Thus it could be argued that it is unnecessary to apply minimum gradient thresholds for geopotential height and thickness, at least when identifying systems of minimum 24-h duration.
Basic climatological statistics of cutoff lows identified with reduced geopotential height and thickness thresholds for sensitivity analysis, in both NCEP1 and ERA-Interim reanalysis, over the whole cool season (April–October inclusive) for each year from 1979 to 2018.
The second objective of this study was to compare the databases of cutoff lows identified in the NCEP1 and ERA-Interim reanalyses. Even though ERA-Interim data were regridded to the same grid spacing as NCEP1, we have found that numbers of systems and system days are higher in the ERA-Interim reanalysis over the NCEP1 reanalysis, across all months of the cool season (Table 1 and Fig. 4), although with substantial overlap in the range of variability. Further, this increase in systems in the ERA-Interim reanalysis over the NCEP1 reanalysis appears to be evenly spread across the range of system durations (Fig. 8), while the spatial spread of the increase in systems shows a slight skew to the west in ERA-Interim compared to NCEP1 (Fig. 7).
Other studies of cutoff low systems in the Southern Hemisphere have also found differences between various reanalyses. Reboita et al. (2010) found a greater number of systems identified in the ERA-40 reanalysis than the NCEP1 reanalysis across the whole Southern Hemisphere, at each of the 500-, 300-, and 200-hPa levels. Comparing a suite of reanalysis datasets, Pinheiro et al. (2020) found greater numbers of cutoff low systems in each of the newer reanalyses (ERA-Interim, NCEP-CFSR, MERRA-2 and JRA-55) than the older JRA-25 reanalysis. Additionally, the newer reanalyses exhibited high similarity between themselves in the numbers and characteristics of these systems. Many reanalysis datasets are now available, including a new dataset designed specifically for the Australian region (BARRA; Su et al. 2019). Including data from the BARRA reanalysis in a future study could yield new insight into the suitability of the range of available reanalyses for studies of climate variability in the Australian region.
The final objective of this study was to analyze the variability of cutoff low frequency, duration and tracks with respect to the remote climate drivers. Scatterplots show little evidence of significant linear correlation between total number of systems, system days, mean system latitude and longitude, and indices representing each of the main known drivers of Australian climate variability (ENSO, IOD, and SAM), for either the cool season as a whole, or individual months (Figs. 10–13). Though some aspects of cutoff low occurrence show a correlation with one of the driver indices with a probability of significance greater than 95% during a single isolated month, we know that the climate drivers exert their influence on Australian climate at longer time scales, and as such we consider it unlikely that the apparent correlation in cutoff low occurrence is a result of those driver influences. On this basis we disregard the correlations spanning only a single month. We find this result somewhat surprising at first, as previous literature has suggested the possibility of influences on the frequency, duration, and position of cutoff lows over SEA by the remote drivers. Cool-season rainfall has been found to be enhanced by atmospheric blocking in the South Pacific region (Pook et al. 2013), and that the occurrence of cutoff lows is also associated with blocking (P06; Nieto et al. 2007; Risbey et al. 2009b; Muñoz et al. 2020; and others). Blocking in the South Pacific, in turn, was found to exhibit some association with ENSO, being more common during La Niña periods, and less common during El Niño (Risbey et al. 2009a), implying that a correlation (though perhaps weaker) between the frequency of cutoff lows and ENSO might be found here. Across the Southern Hemisphere, Favre et al. (2012) found that tracks of cutoff lows shift equatorward during La Niña periods, including in the vicinity of SEA (Fig. 13 in Favre et al. 2012). However, little evidence of either of these relationships (i.e., between cutoff low frequency and ENSO, or between cutoff low tracks and ENSO) was found in this study. Note that we have tested only linear correlations here; relationships between cutoff low occurrence and driver indices may exhibit asymmetries, or even more elaborate forms. Additionally, correlations between the system latitude and longitude and the driver indices were calculated with values of latitude and longitude averaged over the whole domain only, and it may be that the tracks of systems within a narrower latitude/longitude band may show a correlation with driver indices, which is concealed by the broader approach used here.
The lack of correlation found here between the frequency of cutoff lows and ENSO may be partly due to the relative strengths of the correlations between cutoff low frequency and blocking, and between blocking and ENSO. P06 reports a moderate correlation between cutoff low days and blocking at 140°E, with r values between 0.6 and 0.7 for the months June–August. Using SST anomalies in the Niño-3 region to represent ENSO, Risbey et al. (2009b) find a correlation between blocking at 140°E and ENSO of even weaker strength (r = −0.3). A greater correlation between blocking at 140°E and ENSO is found (r = 0.5) when using the Southern Oscillation index (SOI), which may be due to the fact that the SOI is more directly related to the atmospheric response to the ENSO phenomenon (Risbey et al. 2009b). It is perhaps unlikely that difference in criteria for identifying cutoff lows has played a role in the lack of correlation between cutoff low frequency and ENSO found here, since the association between cutoff low frequency and blocking is widely noted in the literature, and has been found for cutoff lows at the 200-hPa level identified with an alternative detection method (Nieto et al. 2007).
The contrast between the correlation between cutoff low tracks and ENSO found by Favre et al. (2012) and the lack of correlation found here, on the other hand, may be a direct result of the differences in cutoff low identification criteria used. Favre et al. (2012) require a complete cold core in their identification of a cutoff low, while our criteria require only a cold trough in the vicinity of the low center. Cyclonic systems exhibit a cold core when they have become completely detached from the midlatitude westerly flow, after which time they are free to move northward; while systems with only a cold trough are still attached to the westerly belt to some extent (sometimes called “tear-off lows,” e.g., Favre et al. 2012), and therefore may not traverse northward as far as completely detached systems might.
In most results discussed here (viz., average system duration, associated precipitation, and correlation between the mean system track and the ENSO) there is evidence to suggest that differences in the criteria for identifying cutoff lows could account for at least some of the differences between the climatological results found here, and elsewhere. Most climatological studies of cutoff lows require a local minimum in thickness/temperature at the center of the system, i.e., a cold core. The criteria of P06, and the criteria used in this study, require only a trough in the thickness field, and this feature may be located either through the center of the system, or along the system’s western flank. Nevertheless, systems with only a cold trough can contribute to rainfall in SEA, and as well as other parts of the world (Favre et al. 2012). Similarly, most climatological studies of cutoff lows detect systems at the 500-hPa level, or even higher, while P06 identifies systems at either the 500-hPa level or the surface. It has been suggested that systems originating in the lower troposphere are in fact a different type of system to those originating in the upper troposphere, and as such should be considered separately. Again, though, it has been found that systems at both levels make a contribution to cool-season rain SEA (P06). It has also been noted that systems with only a cold trough can deliver high precipitation volumes elsewhere (Favre et al. 2012) While systems of each of the variations described here contribute to cool-season rain in SEA, they may be the result of a mixture of different processes of system genesis and development, and hence it is little surprise that they might not show the expected climatology or patterns of variability. Indeed, differences between detection criteria have been found to result in notably different correlations between the variability of cyclones in general over eastern Australia and indices of the remote climate drivers (Pepler et al. 2015).
Last, the characteristics of cutoff low intensity have not been examined here, or any measure of the presence of water vapor within the system. Risbey et al. (2009a) studied composites of various atmospheric conditions such as 1000–500-hPa thickness gradient, Eady growth rate, relative vorticity, displacement from the jet stream, and vertical velocity on days of different rainfall amounts from cutoff lows, and found each of these properties to be related to cutoff low precipitation to some extent. Further, it was found that horizontal thickness gradients and baroclinicity in the SEA region show some relationship with ENSO, and also the IOD (Risbey et al. 2009a). Last, a number of works have identified that cutoff lows of the highest precipitation volumes derive moisture from tropical oceans to the northeast and northwest of Australia (McIntosh et al. 2007; Chubb et al. 2011; Theobald et al. 2015). The passage of humid air masses from the tropics to midlatitude cutoff lows may be influenced by the remote climate drivers, as has indeed been suggested by Chubb et al. (2011), who find some evidence to suggest the influence of the SAM on the transport of humid air masses into precipitating systems. Thus it may be possible that, while the frequency, duration or tracks of cool-season cutoff lows may not be influenced by the remote drivers, the atmospheric properties that intensify the systems or deliver them moisture might be, and these relationships might become apparent using an automated identification method for cutoff lows in future work.
In summary, an automated method has been developed to identify cutoff low systems over SEA during the cool season, and has been found to successfully identify most systems of meteorological importance by comparison with a previously developed manual database. A moderate cycle of cutoff low frequency is seen across the cool season, rising to a peak in June before declining to minimum values in September and October. Percentages of cool-season precipitation coinciding with cutoff lows are again found to be upward of 35% for most of SEA, with even higher values across northeastern NSW than the Mallee or Snowy Mountains regions studied previously. Comparing different reanalyses, more cutoff lows are identified in the ERA-Interim reanalysis than NCEP1, resulting in higher values of system frequency, total system days, and rainfall percentages attributed to these systems. Finally, little evidence of statistically significant relationships was found between the values of cutoff low frequency, duration or tracks, and the ENSO, the IOD or the SAM. It is concluded that future work on the variability of cutoff lows over SEA should employ more specific identification criteria to capture systems of only upper-level origin, and could also benefit from differentiating between systems with either a cold core or a cold trough.
Acknowledgments
We acknowledge support from the Australian Research Council (ARC) Centre of Excellence for Climate Extremes (CE170100023). A.S.T. is supported by the ARC through Grant FT160100495. S.M. was supported by the ARC through Grant FT160100162. A.S.T. and S.M. acknowledge support from the Australian Government’s National Environmental Science Program. N.H.G. was supported by a University of New South Wales Faculty of Science Dean’s honors relocation scholarship. We thank Stuart Browning for providing an example of code for the detection of cyclones, which aided the development of the tracking scheme used in this study. We also thank Mike Pook for providing a copy of the manual identification of cutoff lows presented in his study, and for subsequent correspondence regarding the methods of his work. James Risbey and two other anonymous reviewers are thanked for their comments on the manuscript. Finally, the concept for this study was developed following valued discussions with Kevin Walsh and Wenju Cai.
Data availability statement.
NCEP1 reanalysis data are available at https://psl.noaa.gov/data/gridded/data.ncep.reanalysis.html. ERA-Interim reanalysis data available at https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era-interim. AWAP precipitation data are available at http://www.bom.gov.au/metadata/catalogue/19115/ANZCW0503900567. BoM rain gauge data are available at http://www.bom.gov.au/climate/data/. Data for the ONI are retrieved from https://climatedataguide.ucar.edu/climate-data/nino-sst-indices-nino-12-3-34-4-oni-and-tni. Data for the DMI are retrieved from https://climexp.knmi.nl/getindices.cgi?WMO=NCDCData/dmi_ersst&STATION=DMI_ERSST&TYPE=i&id=someone@somewhere. Monthly AAO index values are retrieved from https://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/aao/aao.shtml.
REFERENCES
Ashok, K., Z. Guan, and T. Yamagata, 2003: Influence of the Indian Ocean dipole on Australian winter rainfall. Geophys. Res. Lett., 30, 1821, https://doi.org/10.1029/2003GL017926.
Browning, S. A., and I. D. Goodwin, 2013: Large-scale influences on the evolution of winter subtropical maritime cyclones affecting Australia’s East Coast. Mon. Wea. Rev., 141, 2416–2431, https://doi.org/10.1175/MWR-D-12-00312.1.
Chubb, T. H., S. T. Siems, and M. J. Manton, 2011: On the decline of wintertime precipitation in the snowy mountains of southeastern Australia. J. Hydrometeor., 12, 1483–1497, https://doi.org/10.1175/JHM-D-10-05021.1.
CSIRO, 2012: Climate and water availability in South-Eastern Australia—A synthesis of findings from Phase 2 of the South Eastern Australian Climate Initiative (SEACI). CSIRO, Australia, 41 pp.
Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553–597, https://doi.org/10.1002/qj.828.
Favre, A., B. Hewitson, M. Tadross, C. Lennard, and R. Cerezo-Mota, 2012: Relationships between cut-off lows and the semiannual and southern oscillations. Climate Dyn., 38, 1473–1487, https://doi.org/10.1007/s00382-011-1030-4.
Fiddes, S. L., A. B. Pezza, and V. Barras, 2015: Synoptic climatology of extreme precipitation in alpine Australia. Int. J. Climatol., 35, 172–188, https://doi.org/10.1002/joc.3970.
Fuenzalida, H. A., R. Sanchez, and R. D. Garreaud, 2005: A climatology of cutoff lows in the Southern Hemisphere. J. Geophys. Res., 110, D18101, https://doi.org/10.1029/2005JD005934.
Gallant, A. J. E., K. J. Hennessy, and J. Risbey, 2007: Trends in rainfall indices for six Australian regions: 1910–2005. Aust. Meteor. Mag., 56, 223–239.
Hartmann, D. L., and Coauthors, 2013: Observations: Atmosphere and surface. Climate Change 2013: The Physical Science Basis, T. F. Stocker et al., Eds., Cambridge University Press, 159–254.
Hendon, H. H., D. W. J. Thompson, and M. C. Wheeler, 2007: Australian rainfall and surface temperature variations associated with the Southern Hemisphere annular mode. J. Climate, 20, 2452–2467, https://doi.org/10.1175/JCLI4134.1.
Hines, K. M., D. H. Bromwich, and G. J. Marshall, 2000: Artificial surface pressure trends in the NCEP–NCAR reanalysis over the Southern Ocean and Antarctica. J. Climate, 13, 3940–3952, https://doi.org/10.1175/1520-0442(2000)013<3940:ASPTIT>2.0.CO;2.
Huang, B., and Coauthors, 2017: Extended Reconstructed Sea Surface Temperature version 5 (ERSSTv5): Upgrades, validations, and intercomparisons. J. Climate, 30, 8179–8205, https://doi.org/10.1175/JCLI-D-16-0836.1.
Jones, D., W. Wang, and R. Fawcett, 2009: High-quality spatial climate data-sets for Australia. Aust. Meteor. Oceanogr. J., 58, 233–248, https://doi.org/10.22499/2.5804.003.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437–471, https://doi.org/10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.
McIntosh, P. C., M. J. Pook, J. S. Risbey, S. N. Lisson, and M. Rebbek, 2007: Seasonal climate forecasts for agriculture: Towards better understanding and value. Field Crops Res., 104, 130–138, https://doi.org/10.1016/j.fcr.2007.03.019.
Meneghini, B., I. Simmonds, and I. Smith, 2007: Association between Australian rainfall and the Southern Annular Mode. Int. J. Climatol., 27, 109–121, https://doi.org/10.1002/joc.1370.
Mo, K. C., 2000: Relationships between low-frequency variability in the Southern Hemisphere and sea surface temperature anomalies. J. Climate, 13, 3599–3610, https://doi.org/10.1175/1520-0442(2000)013<3599:RBLFVI>2.0.CO;2.
Muñoz, C., D. Schultz, and G. Vaughan, 2020: A midlatitude climatology and interannual variability of 200- and 500-hPa cutoff lows. J. Climate, 33, 2201–2222, https://doi.org/10.1175/JCLI-D-19-0497.1.
Ndarana, T., and D. W. Waugh, 2010: The link between cut-off lows and Rossby wave breaking in the Southern Hemisphere. Quart. J. Roy. Meteor. Soc., 136, 869–885, https://doi.org/10.1002/qj.627.
Nicholls, N., 2010: Local and remote causes of the southern Australian autumn-winter rainfall decline, 1958–2007. Climate Dyn., 34, 835–845, https://doi.org/10.1007/s00382-009-0527-6.
Nieto, R., and Coauthors, 2005: Climatological features of cutoff low systems in the Northern Hemisphere. J. Climate, 18, 3085–3103, https://doi.org/10.1175/JCLI3386.1.
Nieto, R., and Coauthors, 2007: Interannual variability of cut-off low systems over the European sector: The role of blocking and the Northern Hemisphere circulation modes. Meteor. Atmos. Phys., 96, 85–101, https://doi.org/10.1007/s00703-006-0222-7.
Pepler, A. S., A. Di Luca, F. Ji, L. V. Alexander, J. P. Evans, and S. C. Sherwood, 2015: Impact of identification method on the inferred characteristics and variability of Australian East Coast lows. Mon. Wea. Rev., 143, 864–877, https://doi.org/10.1175/MWR-D-14-00188.1.
Pinheiro, H. R., K. I. Hodges, M. A. Gan, and N. J. Ferreira, 2017: A new perspective of the climatological features of upper-level cut-off lows in the Southern Hemisphere. Climate Dyn., 48, 541–559, https://doi.org/10.1007/s00382-016-3093-8.
Pinheiro, H. R., K. I. Hodges, and M. A. Gan, 2020: An intercomparison of subtropical cut-off lows in the Southern Hemisphere using recent reanalyses: ERA-Interim, NCEP-CFRS, MERRA-2, JRA-55, and JRA-25. Climate Dyn., 54, 777–792, https://doi.org/10.1007/s00382-019-05089-1.
Pook, M. J., P. C. McIntosh, and G. A. Meyers, 2006: The synoptic decomposition of cool-season rainfall in the southeastern Australian cropping region. J. Appl. Meteor. Climatol., 45, 1156–1170, https://doi.org/10.1175/JAM2394.1.
Pook, M. J., S. Lisson, J. Risbey, C. C. Ummenhofer, P. McIntosh, and M. Rebbeck, 2009: The autumn break for cropping in southeast Australia: Trends, synoptic influences and impacts on wheat yield. Int. J. Climatol., 29, 2012–2026, https://doi.org/10.1002/joc.1833.
Pook, M. J., J. S. Risbey, P. C. McIntosh, C. C. Ummenhoffer, A. G. Marshall, and G. A. Meyers, 2013: The seasonal cycle of blocking and associated physical mechanisms in the Australian region and relationship with rainfall. Mon. Wea. Rev., 141, 4534–4553, https://doi.org/10.1175/MWR-D-13-00040.1.
Pook, M. J., J. S. Risbey, and P. C. McIntosh, 2014: A comparative synoptic climatology of cool-season rainfall in major grain-growing regions of southern Australia. Theor. Appl. Climatol., 117, 521–533, https://doi.org/10.1007/s00704-013-1021-y.
Qi, L., L. M. Leslie, and S. X. Zhao, 1999: Cut-off low pressure systems over southern Australia—Climatology and case study. Int. J. Climatol., 19, 1633–1649, https://doi.org/10.1002/(SICI)1097-0088(199912)19:15<1633::AID-JOC445>3.0.CO;2-0.
Reboita, M. S., R. Nieto, L. Gimeno, R. P. da Rocha, T. Ambrizzi, R. Garreaud, and L. F. Krüger, 2010: Climatological features of cutoff low systems in the Southern Hemisphere. J. Geophys. Res., 115, D17104, https://doi.org/10.1029/2009JD013251.
Risbey, J. S., M. J. Pook, P. C. McIntosh, C. C. Ummenhofer, and G. Meyers, 2009a: Characteristics and variability of synoptic features associated with cool season rainfall in southeastern Australia. Int. J. Climatol., 29, 1595–1613, https://doi.org/10.1002/joc.1775.
Risbey, J. S., M. J. Pook, P. C. McIntosh, M. C. Wheeler, and H. H. Hendon, 2009b: On the remote drivers of rainfall variability in Australia. Mon. Wea. Rev., 137, 3233–3253, https://doi.org/10.1175/2009MWR2861.1.
Risbey, J. S., P. C. McIntosh, and M. J. Pook, 2012: Synoptic components of rainfall variability and trends in southeast Australia. Int. J. Climatol., 33, 2459–2472, https://doi.org/10.1002/joc.3597.
Saji, N., B. Goswami, P. Vinayachandran, and T. Yamagata, 1999: A dipole mode in the tropical Indian Ocean. Nature, 401, 360–363, https://doi.org/10.1038/43854.
Su, C.-H., and Coauthors, 2019: BARRA v1.0: The Bureau of Meteorology Atmospheric high-resolution regional reanalysis for Australia. Geosci. Model Dev., 12, 2049–2068, https://doi.org/10.5194/gmd-12-2049-2019.
Theobald, A., H. McGowan, J. Spiers, and N. Callow, 2015: A synoptic classification of inflow-generating precipitation in the Snowy Mountains, Australia. J. Appl. Meteor. Climatol., 54, 1713–1732, https://doi.org/10.1175/JAMC-D-14-0278.1.
Trenberth, K. E., 1997: The definition of El Niño. Bull. Amer. Meteor. Soc., 78, 2771–2777, https://doi.org/10.1175/1520-0477(1997)078<2771:TDOENO>2.0.CO;2.
Ummenhofer, C. C., M. H. England, P. C. McIntosh, G. A. Meyers, M. J. Pook, and J. S. Risbey, 2009: What causes southeast Australia’s worst droughts? Geophys. Res. Lett., 36, L04706, https://doi.org/10.1029/2008GL036801.
Ummenhofer, C. C., and Coauthors, 2011: Indian and Pacific Ocean influences on southeast Australian drought and soil moisture. J. Climate, 24, 1313–1336, https://doi.org/10.1175/2010JCLI3475.1.
