Subtropical Cyclones over the Southwestern South Atlantic: Climatological Aspects and Case Study

Luiz Felippe Gozzo Department of Atmospheric Sciences, Universidade de São Paulo, São Paulo, Brazil

Search for other papers by Luiz Felippe Gozzo in
Current site
Google Scholar
PubMed
Close
,
Rosmeri P. da Rocha Department of Atmospheric Sciences, Universidade de São Paulo, São Paulo, Brazil

Search for other papers by Rosmeri P. da Rocha in
Current site
Google Scholar
PubMed
Close
,
Michelle S. Reboita Natural Resources Institute, Universidade de Itajuba, Minas Gerais, Brazil

Search for other papers by Michelle S. Reboita in
Current site
Google Scholar
PubMed
Close
, and
Shigetoshi Sugahara Instituto de Pesquisas Meteorologicas e Programa de Pos-Graduacao da Faculdade de Ciencias, UNESP, Bauru, Brazil

Search for other papers by Shigetoshi Sugahara in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Hurricane Catarina (2004) and subtropical storm Anita (2010) called attention to the development of subtropical cyclones (SCs) over the South Atlantic basin. Besides strong and organized storms, a large number of weaker, shallower cyclones with both extratropical and tropical characteristics form in the region, impacting the South American coast. The main focus of this study is to simulate a climatology of subtropical cyclones and their synoptic pattern over the South Atlantic, proposing a broader definition of these systems. In addition, a case study is presented to discuss the main characteristics of one weak SC. The Interim ECMWF Re-Analysis (ERA-Interim) and NCEP–NCAR reanalysis are used to construct the 33-yr (1979–2011) climatology, and a comparison between them is established. Both reanalyses show good agreement in the SCs’ intensity, geographical distribution, and seasonal variability, but the interannual variability is poorly correlated. Anomaly composites for austral summer show that subtropical cyclogenesis occurs under a dipole-blocking pattern in upper levels. Upward motion is enhanced by the vertical temperature gradient between a midtropospheric cold cutoff low/trough and the intense low-level warm air advection by the South Atlantic subtropical high. Turbulent fluxes in the cyclone region are not above average during cyclogenesis, but the subtropical high flow advects great amounts of moisture from distant regions to fuel the convective activity. Although most of the SCs develop during austral summer (December–February), it is in autumn (March–May) that the most “tropical” environment is found (stronger surface fluxes and weaker vertical wind shear), leading to the most intense episodes.

Corresponding author address: Luiz Felippe Gozzo, IAG-USP, Rua do Matao, 1226, Cidade Universitaria, São Paulo, SP, CEP 05508-090, Brazil. E-mail: luiz.gozzo@iag.usp.br

Abstract

Hurricane Catarina (2004) and subtropical storm Anita (2010) called attention to the development of subtropical cyclones (SCs) over the South Atlantic basin. Besides strong and organized storms, a large number of weaker, shallower cyclones with both extratropical and tropical characteristics form in the region, impacting the South American coast. The main focus of this study is to simulate a climatology of subtropical cyclones and their synoptic pattern over the South Atlantic, proposing a broader definition of these systems. In addition, a case study is presented to discuss the main characteristics of one weak SC. The Interim ECMWF Re-Analysis (ERA-Interim) and NCEP–NCAR reanalysis are used to construct the 33-yr (1979–2011) climatology, and a comparison between them is established. Both reanalyses show good agreement in the SCs’ intensity, geographical distribution, and seasonal variability, but the interannual variability is poorly correlated. Anomaly composites for austral summer show that subtropical cyclogenesis occurs under a dipole-blocking pattern in upper levels. Upward motion is enhanced by the vertical temperature gradient between a midtropospheric cold cutoff low/trough and the intense low-level warm air advection by the South Atlantic subtropical high. Turbulent fluxes in the cyclone region are not above average during cyclogenesis, but the subtropical high flow advects great amounts of moisture from distant regions to fuel the convective activity. Although most of the SCs develop during austral summer (December–February), it is in autumn (March–May) that the most “tropical” environment is found (stronger surface fluxes and weaker vertical wind shear), leading to the most intense episodes.

Corresponding author address: Luiz Felippe Gozzo, IAG-USP, Rua do Matao, 1226, Cidade Universitaria, São Paulo, SP, CEP 05508-090, Brazil. E-mail: luiz.gozzo@iag.usp.br

1. Introduction

A subtropical cyclone (SC) is a low pressure system that presents both extratropical and tropical structure during its development. These hybrid cyclones are nonfrontal, with a low tropospheric warm core and an upper-level cold core (Hart 2003). They can be formed with such characteristics and maintain them throughout the cyclone lifetime, or they may originate as an intermediary stage in a tropical or extratropical transition (Ritchie and Elsberry 2001; Hulme and Martin 2009).

A procedure to identify such cyclones is the cyclone phase space (CPS; Hart 2003), an algorithm that describes the three-dimensional thermal structure of the cyclone by its thermal wind profile and horizontal thermal symmetry. Employing this algorithm, Evans and Guishard (2009) studied 18 SCs over the North Atlantic basin, showing that these systems form by the intrusion of an upper-level trough over an unstable subtropical low troposphere [warm sea surface temperature (SST) and weak static stability]. The formation of a midtropospheric cutoff low reduces the trough scale and strengthens the interaction of upper-level features and the low-level cyclonic perturbation, leading to the development of a hybrid cyclone.

Guishard et al. (2009) proposed a first unambiguous set of criteria to define subtropical storms: they should be diagnosed as hybrid systems (by the CPS parameters) for more than one diurnal cycle and should attain sustained gale-force winds (>17 m s−1) during their life cycle. Climatology based on these criteria for the North Atlantic basin showed an annual mean of four subtropical storms, and the months of September and October as the most active. Their study reported subtropical storms developing in highly “unfavorable” conditions, in SST as cool as 16°C and vertical wind shears as strong as 40 m s−1.

Evans and Braun (2012) presented the first SC climatology for the South Atlantic basin, reporting a mean of 1.2 SCs per year and little monthly variability. In this study, Rossby wave breaking was pointed out as a genesis mechanism for just a few cyclones, while the lee-cyclogenesis downstream of the Andes and over the warm Brazilian Current accounted for the other cyclogenesis mechanisms.

Over the southwestern South Atlantic, the most organized and interesting system recorded was Hurricane Catarina, which began as an extratropical precursor and presented subtropical structure before the transition into a tropical cyclone, in March 2004 (McTaggart-Cowan et al. 2006). It did not develop over high SST, but cold air aloft, associated with a midtropospheric trough, reinforced the upward motion and convective activity, while a persistent upper-level Rex blocking pattern (Rex 1950) was responsible for weakening the vertical wind shear and the eastward steering of the cyclone (McTaggart-Cowan et al. 2006). In March 2010, the subtropical storm Anita developed in a very similar upper-level condition of dipole blocking, and although the interaction with another extratropical disturbance prevented the system from undergoing tropical transition (Dias Pinto et al. 2013), it acquired all the characteristics of a well-organized SC.

The purpose of this work is to present a climatology of SCs over the southwestern sector of South Atlantic basin, including weaker and shallower systems observed mostly in a lower-latitude cyclogenetic region along the South American coast, named RG1 by Reboita et al. (2010b). For this purpose, the definition of SC proposed is slightly different of that used by Guishard et al. (2009) and Evans and Braun (2012). While abundant research has been done to describe the representation of extratropical and tropical cyclone climatology in several reanalyses, a direct comparison of SC climatologies has not yet been evaluated. This will also be addressed in this paper, where the Interim European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA-Interim) and National Centers for Environmental Prediction (NCEP) reanalyses will be compared. Furthermore, we present composite analyses for the SCs over the most active region of the basin, in order to better understand the environment in which they are formed. The paper is organized as follows: section 2 describes the data and methodology, section 3 presents the results, and section 4 brings the conclusions.

2. Data and methodology

a. Data

This study uses data from ERA-Interim (Dee et al. 2011), referred to hereinafter as ERAInt. We have used data with horizontal resolution of 1.5° × 1.5° and 10 pressure levels (from 1000 to 200 hPa), at every 6 h (0000, 0600, 1200, and 1800 UTC). This work will focus on the period from 0000 UTC 1 January 1979 to 1800 UTC 31 December 2011. The ERAInt reanalysis presents a four-dimensional variational data assimilation system (4D-VAR) and a new moisture analysis, which reduces issues from the previous 40-yr ECMWF Re-Analysis (ERA-40; Hólm 2002).

A comparison between the climatology constructed with ERAInt and the NCEP–National Center for Atmospheric Research (NCAR) reanalysis (Kalnay et al. 1996), hereafter NCEP1, was carried out to investigate whether these reanalyses have a similar representation of SCs. This dataset was chosen for comparison because it was never used for SC climatology studies before and spans a large period (available from 1948 onward), being a potential source for studies about the low-frequency variability of such systems. NCEP1 data used have 2.5° × 2.5° horizontal resolution, with pressure levels and times identical to ERAInt.

SST and sensible and latent surface heat fluxes over the oceans were obtained from the Woods Hole Oceanographic Institute (WHOI) objective analysis. This dataset is a blend of various satellite data and three atmospheric reanalyses (Yu et al. 2008), with the daily fluxes computed from variables estimated by the Coupled Ocean–Atmosphere Response Experiment (COARE) bulk algorithm version 3.0, and available over the globe at 1.0° × 1.0° resolution, from 1985 to 2011.

The case study was carried out with atmospheric data from the ERAInt and oceanic data from the WHOI analysis. The precipitation data are from the Tropical Rainfall Measuring Mission (TRMM)-3B42 daily dataset (Huffman et al. 2007; available at http://mirador.gsfc.nasa.gov/). Geopotential height and temperature anomaly profile were calculated as deviation of a zonal mean from 55° to 35°W.

b. The cyclone tracking algorithm and the cyclone phase space

The algorithm for identification and tracking of cyclones employed in this study was first developed by Sugahara (2000), using a methodology similar to that of Sinclair (1994, 1995). It identifies the cyclone trajectory as a sequence of local minima (in the Southern Hemisphere) of relative vorticity in the gridded horizontal wind field. A detailed description of the tracking is given by Reboita et al. (2010b).

The tracking scanned an area encompassing most of the southwestern South Atlantic ocean and a portion of South America (this region is hereinafter referred to as SAO), as shown in Fig. 1. The cyclones are identified in 925-hPa horizontal wind fields as relative vorticity nuclei with less than −1.5 × 10−5 s−1 for more than 24 consecutive hours, regardless of its extratropical, subtropical, or tropical structure. Closed isobars were not a required condition, because even open isobar cyclones may cause significant weather events, especially near the coast (Sugahara 2000). The region delimited by the solid black box in Fig. 1, 30.5°–21°S and 49.5°–35.5°W, is the RG1, in which the cyclones will be analyzed in further detail (section 3d).

Fig. 1.
Fig. 1.

Effective cyclone tracking area (SAO, blue dashed rectangle) and cyclogenetic region 1 (RG1, black solid rectangle).

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

The algorithm provides the date, time, position (latitude and longitude), and sea level pressure of the cyclones’ centers during their lifetimes to the CPS code, which sorts out the SCs from the extratropical storms by their thermal features. The CPS describes the structure of cyclones based on three parameters: 1) the lower-tropospheric thermal symmetry (B), 2) the lower-tropospheric thermal wind (−VTL), and 3) the upper-tropospheric thermal wind (−VTU). These parameters are calculated from the three-dimensional geopotential field alone, and according to their values the cyclones can be classified as extratropical, subtropical, or tropical. There is no clear division between the classes, which allows the existence of cyclones with mixed characteristics and transformation between types (transitions). This approach evidences the idea of cyclones as a continuum and not strictly separated in classes (Beven 1997) and has been widely used in the study of SCs all over the globe; for the parameters, equations, and application to the Southern Hemisphere, see Dias Pinto and da Rocha (2011) and Dias Pinto et al. (2013).

CPS parameters are defined so that for an extratropical cyclone, B >> 0, −VTL < 0, and −VTU < 0 (the horizontal low-level temperature field is asymmetric, and the cyclone has a cold core extending throughout the troposphere), and for a tropical cyclone B ≈ 0, −VTL > 0, and −VTU > 0 (implying a thermally symmetric low-level field and a warm core from the surface up to the tropopause). The SCs are nonfrontal, with small values of B (around 10; Evans and Guishard 2009; Guishard et al. 2009), and present a low-level warm core and an upper-level cold core, implying −VTL > 0 and −VTU < 0.

c. Criteria for SCs identification

In this study, a cyclone is classified as subtropical if the following conditions are met:

  1. The SC forms between 20° and 40°S. In the region of study, this limitation is important to avoid the inclusion of polar lows or mesoscale cyclones that can occur throughout the year around the Antarctic continent (Carrasco et al. 2003), in the southern portion of the domain.

  2. It presents horizontal thermal symmetry and hybrid structure for more than 36 consecutive hours. Both conditions are diagnosed by using the CPS parameters, with B < 25 m, −VTU < −10 and −VTL > −50.

  3. It attains the required values of B, −VTU, and −VTL over the ocean, and within 24 h after its genesis if first tracked as an extratropical system.

These criteria largely follow Evans and Guishard (2009) and Evans and Braun (2012), but with three main differences. First, the −VTL threshold was relaxed, passing from −10 to −50. Because the CPS can sometimes give a very low or negative value of this parameter even for pure tropical cyclones (Braun 2009), this relaxation ensures that all potential SCs are maintained in the climatology. Extratropical systems eventually retained by using this new threshold were rejected upon visual inspection of the geopotential height anomaly profile (if they presented cold and westward tilting cores) and 925-hPa temperature fields (if they had associated frontal zones).

The two most fundamental differences here for previous definitions are the elimination of the gale-force wind threshold and no requirement of an upper closed low at 500 hPa. The reasoning behind this is the abundant occurrence of shallow cyclonic systems, especially near the southeastern Brazilian coast, that do not reach sustained winds of 17 m s−1 but may cause notable weather events (persistent rains, floods, etc.).

Thus, we present here a slightly different definition of SC, simply as a nonfrontal low pressure system with a warm core in lower levels and a cold core in upper levels. Although it will be clear in the results section that most SCs indeed reach gale-force winds at some stage of their development and/or extend their circulation up to 500 hPa, the absence of these limitations provided remarkable differences in climatology compared to that of Evans and Braun (2012).

3. Results

a. A shallow SC case study

A short case study of a shallow SC from 11 to 13 January 2008 is presented to illustrate the kind of SC that is included in the climatology by the new proposed definition of section 2c. This particular cyclone was selected to the study because it develops and stays semistationary near the southwestern Brazilian coast (a region of particular economic and societal interest) during summer (the most active season for SCs, as shown later). Also, we have chosen a cyclone that was not associated with the South Atlantic convergence zone (SACZ).

The cyclone is detected by the tracking algorithm at 0000 UTC 12 January 2008, lasting for more than 36 h (until 1800 UTC 13 January 2008). During all this time, the cyclone presents small values of B and positive (negative) −VTL (−VTU) values, placing it in the symmetric shallow warm core region of the CPS phase diagram (Fig. 2). It forms near the southeastern Brazilian coast, with no significant upper-level forcings (during all the development, the upper jet velocity is below 10 m s−1, and there are no potential vorticity anomalies; figures not shown). The SST was moderately high (25°–26°C) and the sum of surface latent and sensible heat fluxes reached a peak of 90 W m−2 on 13 January (not shown).

Fig. 2.
Fig. 2.

Cyclone phase diagram of a shallow hybrid cyclone from 0000 UTC 12 Jan to 1800 UTC 13 Jan 2008: (a) B vs −VTL and (b) −VTL vs −VTU.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

The Geostationary Operational Environmental Satellite (GOES)-10 infrared image at 0000 UTC 12 January 2008 shows that the surface cyclone formed in a region with few clouds (Fig. 3a). There are nuclei of strong convection over the continent, but the SACZ is not configured. At 1200 UTC 12 January and 0000 UTC 13 January (Figs. 3b,c) the cloud band associated with the cold front propagates from the south. In the cyclone region, there are mostly shallow clouds at these times. At 1200 UTC 13 January (Fig. 3d) a region of stronger convection is seen to the south of the cyclone center. By this time, the system is already weakening. This sequence of satellite images shows that the SC was not associated with deep cumulus convection.

Fig. 3.
Fig. 3.

GOES-10 IR images at (a) 0000 UTC 12 Jan, (b) 1200 UTC 12 Jan, (c) 0000 UTC 13 Jan, and (d) 1200 UTC 13 Jan 2008. The capital “L” indicates the low pressure center according to the tracking algorithm.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

On 11 January, one day before the cyclone begins to be tracked, there is already a precursor cyclonic vorticity nucleus at 25°S, 43°W (Fig. 4a). Accumulated rainfall in 24 h occurred over the continent and in a prefrontal region over the ocean. The region of the cyclonic precursor is subject to warm air advection and presents a trough in the sea level pressure field, near 25°S, 45°W (Fig. 4c). Farther south from this region, a cold front extends from the ocean to the north of Argentina, clearly seen in the cyclonic vorticity and horizontal wind fields. The wind field also shows a northwesterly flow from the Amazon region toward the South Atlantic, advecting great amounts of moist air (Figs. 4a,b). Moisture flux converges near the São Paulo state coast (Fig. 4b), in a region of warmer air, creating a suitable unstable environment for the cyclone to develop. The vertical cross section of geopotential height and temperature anomaly from the zonal mean through 24°S shows a wide region of warm air from the surface up to 700 hPa. The strongest positive temperature anomaly is located between 55° and 50°W. This originated from the warming of the continental surface. The warm air from the continent advances toward the ocean up to 40°W (Fig. 4d). During this day, the wind speed attains around 9 m s−1.

Fig. 4.
Fig. 4.

Synoptic fields at 1200 UTC 11 Jan 2008, from ERAInt. (a) The 925-hPa horizontal wind (vectors, in m s−1) and cyclonic relative vorticity (dashed lines, in s−1); daily precipitation (shaded, in mm). (b) Vertically integrated moisture flux (vectors, in kg m−1 s−1) and its divergence (shaded, in kg s−1). (c) Air temperature advection (shaded, in K day−1) and mean sea level pressure (contours, in hPa). (d) Longitudinal cross section of geopotential height (m) and air temperature (K) anomalies from the 24°S zonal mean.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

At 1200 UTC 12 January, the cold front displaces northeastward and is located near the vorticity nucleus (Fig. 5a). Moderate rainfall (40 mm in 24h) occurs along the São Paulo state seashore, likely reinforced by orographic uplifting of the southeasterly flow. The region of cyclogenesis still presents moisture flux convergence, and the vertically integrated moisture flux vectors indicate that, unlike the previous day, most of the moisture is coming from the oceanic area to the northeast of the development region (Fig. 5b). The warm advection is weaker in comparison to the day before, and it attains its minimum sea level pressure (Fig. 5c). By this time the wind reaches its maximum speed (13.4 m s−1), which is below the gale-force wind threshold. The warm air bubble from the surface up to 700 hPa detaches from the continental thermal low, and provides the hybrid structure for the cyclone centered on 45°W (Fig. 5d).

Fig. 5.
Fig. 5.

As in Fig. 4, but at 1200 UTC 12 Jan 2008.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

The cyclone, still coupled to the cold front, starts to weaken at 0000 UTC 13 January (Fig. 6a). Moderate to heavy rain persists along the São Paulo state coast and the cold front in the open ocean, regions where there is still moisture flux convergence. The cyclonic curvature of the moisture transport, seen in Fig. 6b, influences the coastline near the border of São Paulo and Rio de Janeiro states; although small, this perturbation may have an important impact on the weather pattern at the regional scale. The pressure of the SC rises to 1013 hPa and the maximum wind speed is 9.9 m s−1, while the warm air advection is negligible in the cyclone area (Fig. 6c). The low-level warm air core and geopotential height anomaly also weaken, while the middle levels are colder (Fig. 6d); this hybrid structure persists at least until 1200 UTC 13 January. By this time, the SC continues to weaken. The warm advection is similar to the previous time, but the moisture flux convergence is drastically reduced (not shown), indicating that the thermodynamic instability arising from this convergence may be an important mechanism to maintain the upward motions and the cyclonic circulation.

Fig. 6.
Fig. 6.

As in Fig. 4, but at 0000 UTC 13 Jan 2008.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

Although this SC fulfills the conditions 1–3 of section 2c, and thus can be classified as a SC, it would not be included in the Guishard et al. (2009) and Evans and Braun (2012) climatologies because of its weak maximum 925-hPa wind speed and shallow circulation (the geopotential height anomaly associated with the cyclone only extends from the surface up to 800 hPa; Figs. 4d, 5d, and 6d). However, as these systems are relatively common and may have significant localized impacts along the southeastern Brazilian coast, we relaxed the criteria and constructed a broader climatology of hybrid systems.

b. Interannual and seasonal variability

From January 1979 to December 2011, a total of 238 (233) subtropical cyclogeneses were detected in the ERAInt (NCEP1) database, corresponding to 3.7% (4.2%) of the total cyclogeneses tracked in the SAO domain (Fig. 1). The mean annual number of SCs and its standard deviation is nearly identical: 7.2 ± 2.8 and 7.1 ± 2.8 for ERAInt and NCEP1, respectively. The high number of cyclogenesis cases per year in comparison to results of Evans and Braun (2012) indicates that shallow SCs with moderate winds are relatively common in the region.

While the mean number of cyclones is similar in both reanalyses, their interannual variability is significantly different (Fig. 7a). The agreement between the annual number of events is poor for most years, resulting in a low Pearson correlation (+0.26). Similarly low correlations were found by Hanson et al. (2004), who also used ECMWF and NCEP1 reanalyses and used a similar tracking algorithm for total cyclones in the North Atlantic. They pointed out that this discrepancy is even larger for weak cyclones.

Fig. 7.
Fig. 7.

(a) Annual frequency and (b) seasonal mean of SC over SAO for ERAInt (blue) and NCEP1 (orange). In (b), the bar plot is the seasonal mean and the lines represent the seasonal percentage of SC in the total amount of cyclones.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

The mean seasonal distribution of SCs is almost identical for ERAInt and NCEP1 reanalyses. Summer is the season with highest cyclogenetic activity (Fig. 7b), January being the most active month of the year (3.2 cyclones in ERAInt, 2.8 cyclones for NCEP1 on average). There are various conditions favoring the occurrence of SCs at this time of the year such as the southward displacement of the subtropical upper-level jet, the sea level pressure decrease near the southeastern coast of Brazil (also influenced by the formation of the South Atlantic convergence zone), and the increased transport of heat and moisture to the region by the South American low-level jet (Reboita et al. 2012). This summer environment promotes more frequent shallow hybrid cyclones, by the transport of continental warm air to the ocean (as depicted by the case study of section 3a). Because of these systems, the number of cyclones in summer is higher than in autumn, the most active season according to the Evans and Braun (2012) methodology. In February the number of cyclones starts to decrease, but in March they increase again. Especially because of March, autumn is the second most active season of the year in this climatology. The smaller frequency of events occurs in the winter season, with about 0.3 cyclones for ERAInt and NCEP1. This frequency during winter is similar to the one documented by Evans and Braun (2012), but in that study winter was the second most active season.

c. Mean characteristics

On average, the SCs’ lifetime in ERAInt (NCEP1) is 4.1 (4.2) days, during which they travel a distance of 1397.0 (1172.4) km, with a velocity of 4.10 (3.17) m s−1. Thus, their lifetime is similar to extratropical cyclones (Simmonds and Keay 2000; Tilinina et al. 2013), but they are slower and travel a shorter distance than previously observed for all cyclones in SAO (Reboita et al. 2010b; Krüger et al. 2012). This is expected, since these systems develop usually associated with an upper-level blocking pattern that decreases the steering flow. This slow movement allows the SCs to interact more deeply with the environment, favoring organized convective activity, and poses a threat to the coastal regions of South America, as it causes prolonged adverse weather conditions. Mendes et al. (2010) shows that during summer and autumn, at the South American coast, there is a maximum number of cyclones traveling much shorter distances than the mean for the whole Southern Hemisphere; although this was not explored in that work, it is possible that SCs contribute to the reported maximum.

Differences in the histograms for lifetime, distance traveled, and mean velocity are small between the two analyzed datasets, as can be seen in Fig. 8. Most of the SCs travel 0–500 km in 2–3 days, with a mean velocity less than 2 m s−1, in both reanalyses. The main distinction in the distributions is the second maximum for 6–7 days and 2500–3000 km in ERAInt, which arises because many cyclones in this dataset form as subtropical and later turn into extratropical, with a longer lifetime and distance traveled. This is especially true for many near coastal shallow cyclones.

Fig. 8.
Fig. 8.

Histograms of the (a) lifetime (in days), (b) distance traveled (in km), and (c) mean velocity (in m s−1) for SC over SAO in ERAInt (light gray) and NCEP1 (dark gray).

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

Although some SCs undergo this extratropical transition, the peak intensity (minimum cyclonic vorticity and strongest low-level winds) was attained during the subtropical stage for most of the cyclones in both datasets. The cyclones that presented peak intensity during late extratropical stage (25.2% in ERAInt and 33.3% in NCEP1) were retained in the analysis because they formed as SCs and remained with hybrid structure for at least 36 consecutive hours, and thus must be accounted for in a subtropical cyclogenesis climatology. Details of the extratropical transition process are beyond the scope of this work.

To evaluate the SCs’ intensity and representation in different reanalyses, the frequency distribution of minimum 925-hPa relative vorticity and maximum value of −VTL attained by the SCs during its subtropical stage are compared. Data from cyclones that attained peak intensity in extratropical stage were not considered in this distribution. Relative vorticity is used here because it can show differences between the datasets more clearly than sea level pressure or wind speed, since this variable is more sensitive to changes in horizontal resolution. The −VTL parameter magnitude is used to assess the intensity of the low-level warm cores developed in the cyclones.

The minimum relative vorticity distribution for NCEP1 is strongly concentrated in weak values, while the ERAInt distribution is more spread out, with much more intense cyclones (Fig. 9a). This is a common pattern also documented by Hodges et al. (2011) for extratropical cyclones and Strachan et al. (2013) for tropical cyclones, for example. Most SCs reach intensities between −2 and −1.5 × 10−5 s−1, indicating that these hybrid systems are in general weaker than the extratropical cyclones in this area (Sinclair 1995; Reboita et al. 2010b).

Fig. 9.
Fig. 9.

ERAInt (light gray) and NCEP1 (dark gray) SAO SCs histograms of (a) minimum cyclonic vorticity (in 10−5 s−1), (b) maximum wind speed attained in the subtropical stage, (c) radius of maximum wind speed during the subtropical stage, and (d) maximum −VTL.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

Figure 9b presents the distribution of 925-hPa maximum wind speed during subtropical stage. About 20% (30%) of SCs in ERAInt (NCEP1) presented maximum sustained winds below gale force. In NCEP1 occurred a larger number of SCs with weaker winds (higher frequency between 14 and 16 m s−1) while ERAInt presented a near-normal distribution with a peak in the 18–20 m s−1 range.

According to Guishard et al. (2009), the maximum wind speed in SCs occurs at a distance of 100 miles (160 km) or more of the cyclone center. For most of the South Atlantic SCs, this maximum occurs at a distance of 300–450 km during subtropical stage (Fig. 9c). Smaller cyclones, with radius less than 150 km, were few for both reanalyses, but were retained in the SC climatology for this region if meeting the conditions presented in section 2c. These mesocyclones usually develop during summer along the oceanic portion of the South Atlantic convergence zone, in an environment of large-scale upward motion and moisture convergence (Quadro 2012).

Warm core intensity in SCs during subtropical stage is similar in both reanalyses, as shown in the maximum −VTL distribution in Fig. 9d. Most of them attain a maximum −VTL of 40–60 m s−1 and the more refined grid again allows the representation of more extreme systems: the ERAInt overestimates the NCEP1 number of cyclones for the weaker (below 20) and stronger (above 100) warm cores. The finer resolution of ERAInt allows, in the former case, the tracking of smaller vortices, and in the latter, the identification of smaller regions of high temperature that are not resolved by NCEP1.

d. SC spatial density and the importance of RG1 cyclogenetic area

The spatial distribution of subtropical cyclogenesis and cyclolysis over the SAO region is examined employing the spherical kernel method (Hodges 1996), which depicts the density map by superposing areas of influence of the cyclogenesis (cyclolysis) point for every cyclone. Further information on this technique can be found in Bombardi et al. (2014), who used this method to construct a density map of total cyclogenesis over the South Atlantic.

Most of the cyclogeneses occur in near-coastal SAO regions, as also pointed out by Evans and Braun (2012); here we show that the SCs occur mostly in a well-defined cyclogenetic area near the southeastern coast of Brazil, for both ERAInt and NCEP1 reanalyses (Figs. 10a,b). High density of cyclolysis occurs between 50° and 25°W, with the densest region coinciding with the genesis region, evidencing the semistationary feature of most SCs (Figs. 10c,d). A second cyclolysis maximum is seen near 45°S, 5°W, mainly because at this region the tracking is discontinued (the southeastern corner of the tracking domain is located there; Fig. 1). This second nucleus is much denser in ERAInt, due to a larger number of subtropical systems that turn into extratropical ones, then moving up to these regions and beyond. In NCEP1, more subtropical semistationary lows are present, resulting in the stronger cyclolysis nucleus near the South American coast and not far from the cyclogenesis region.

Fig. 10.
Fig. 10.

Subtropical cyclogenesis and cyclolysis (number of cyclones by square radian per day) for (a),(c) NCEP1 and (b),(d) ERAInt. (e) Seasonal mean (bar plot) and percentage of SC in the total amount of cyclones over the RG1 for ERAInt (blue) and NCEP1 (orange).

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

The denser nucleus of subtropical cyclogenetic activity in Fig. 10 occur over the RG1 region (Fig. 1, black box) defined in Reboita et al. (2010b) and also identified by Sinclair (1995), Hoskins and Hodges (2005), and Krüger et al. (2012). According to these works, RG1 is the genesis region of weaker cyclones that for the most part do not present a closed isobar at the sea level pressure field. That is why some climatologies based on tracking of cyclones in the sea level pressure field (e.g., Gan and Rao 1991; Mendes et al. 2010) do not report this cyclogenesis region.

Regarding all cyclogenesis, Reboita et al. (2010b) show summer (winter) as the most (least) active season in RG1, and it can also be seen in the seasonal distribution of SCs over this area (Fig. 10e). The correspondence of location and seasonal cycle leads us to believe that a significant portion of all cyclones that formed in RG1 is subtropical. Indeed, 34% (22%) of the tracked cyclones in ERAInt (NCEP1) over the RG1 in summer are subtropical. In spring and autumn the incidence of these systems is lower, but they still account for about 15% of all cyclones.

Based on the larger frequency of cyclogenesis over the RG1, this area is selected to construct storm-centered composites. The use of cyclones developed only in this region also contributes to a more precise description, as systems that formed farther south or in the northern flank of the subtropical high (as is the case of all winter cyclones of ERAInt) may have different characteristics.

e. Cyclone-centered composites during genesis

This section presents cyclone-centered composites around the genesis time, to describe the synoptic environment in which the South Atlantic SCs are formed. For this analysis we have used only the ERAInt dataset, as the composites derived from the NCEP1 presented similar patterns but with less detailed features, due to lower resolution. The center of the SCs corresponds to the location of the relative vorticity nucleus, whose coordinates are given by the tracking algorithm. The horizontal fields are displayed in an area of 40° × 40°, and the vertical profile of potential vorticity (PV) and potential temperature (θ) is shown to the latitude of the cyclone center spanning 70° of longitude. Composites are presented from 24 h prior to the cyclogenesis up to 24 h afterward.

A total of 126 SCs that formed over the RG1 from 1979 to 2011 were considered to construct the composites. We first present the results for summer (68 cases), and then the main differences between the summer SCs and the ones that developed in autumn and spring (39 and 19 cases, respectively).

1) Composites for summer cyclogenesis

The composites for sea level pressure field show that the summer SCs develop in a broad area of low pressure (by the geographical position of the RG1, it is possibly the thermal low over the South American continent) extending eastward, toward the South Atlantic Ocean, 24 h prior to the cyclogenesis (Fig. 11a). At this time, there is already a 1011-hPa closed isobar associated with the incipient cyclone. On the genesis day (t = 0), the cyclone maintains basically the same configuration, while the South Atlantic subtropical high (SASH) slightly strengthens (Fig. 11b). The cyclone attains 1010 hPa at t = +24 h, and the pressure gradient between the low and the SASH is intensified (Fig. 11c). The central closed isobar in the sea level field presents a diameter of 5° and the whole circulation associated with the cyclone has an approximate diameter of 15°. These features in the composite analysis indicate that most of the summer SCs present closed isobars and horizontal scale larger than mesocyclones, although as discussed above these criteria were not considered for the climatology construction. Figures 11d–f indicate that during the whole analyzed period the surface cyclone lies to the east of an equatorward extension of PV, in a region where the favored upward motions help the cyclone to intensify (Hoskins et al. 1985). It is interesting to notice that, on a climatological perspective, the summer hybrid cyclones in South America are not associated with a detached PV core that originated via Rossby wave break [as in the SC of Evans and Guishard (2009)], nor are they cradled by two PV regions as in the case study of Braun (2009). The composites of Fig. 11, with an upper-level cyclonic PV anomaly located westward of the cyclone, and high positive PV air to the east (associated with an outflow anticyclone), is qualitatively similar to the “favorable superposition” pattern for tropical cyclone intensification described in Hanley et al. (2001): the intensification arises from the reduction of vertical wind shear and the establishment of a divergent flow downshear from the cyclone. As shown by these authors, the upper-level cyclonic PV does not cross the center; instead, it is eroded by horizontal PV advection and interaction with the region of strongest diabatic heating (Fig. 11c).

Fig. 11.
Fig. 11.

Austral summer [December–February (DJF)] SC-relative composites of (a)–(c) sea level pressure (contours, in hPa) and PV at 250 hPa (shaded, in PVU); (d)–(f) vertical cross sections of PV (shaded, in PVU) and potential temperature (contours, in K), for (left) T = −24 h prior to the cyclogenesis, (middle) T = 0 (1200 UTC on the genesis day), and (right) T = +24 h.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

The upper-level PV anomaly can be seen in a cross section through the cyclone center (Figs. 11d–f), extending down to 300 hPa, with a magnitude of −0.6 to −0.9 PVU. At t = 0, two regions of negative PV develop in the central axis of the cyclone, induced by temperature anomalies: the one between 900 and 700 hPa is associated with the low-level warm core characteristic of hybrid systems, and the anomaly between 700 and 450 hPa arises as a response to cloud condensation (Fig. 11e). As the SC develops, these nuclei get stronger, but they do not form a single vorticity tube, resembling the composites of Evans and Guishard (2009). The weaker PV concentration near the cyclone center is one of the reasons why transitions to a tropical storm are less frequent in this area of the globe (Braun 2009).

The composite of 250-hPa horizontal flow difference between cyclogenesis days and the seasonal [December–February (DJF)] mean shows a cyclonic circulation linked to the PV anomaly, from 24 h prior to the genesis up to 24 h after, always located around 5° to the west of the surface low (Figs. 12a–c). Poleward of the SC, an anticyclonic circulation develops, and both structures form a Rex blocking pattern with slow zonal movement, decreasing the vertical shear and favoring organized convection.

Fig. 12.
Fig. 12.

Austral summer (DJF) SC-relative anomaly composites of (a)–(c) 250-hPa flow (streamlines) and 250-hPa magnitude of horizontal wind (shaded, in m s−1); (d)–(f) 500-hPa geopotential height (contours, in m) and sea level pressure (shaded, in hPa). Anomaly fields are the difference between days of cyclogenesis and the DJF mean. (left) T = −24 h, (middle) T = 0, (right) T = +24 h.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

The middle troposphere over the cyclone is cold, as indicated by the negative geopotential height anomaly in Figs. 12d–f. This anomaly represents a trough to the west of the region where the cyclone will develop (Fig. 12d), and in later stages a stronger trough or (in most cases) a cutoff low detached from the westerly flow (Figs. 12e,f). At t = +24 h, this feature is almost vertically stacked with the surface low (less than 5° to the west), decreasing the vertical wind shear over the cyclone (not shown) and favoring strong destabilization of the low troposphere, by quasigeostrophic and thermodynamic ascent. At the surface, the cyclone is intensified by the scale reduced upper-level low (Fig. 12f), and an anomalous high pressure develops downstream; this whole pattern resembles the midlevel and surface composites for SCs of the North Atlantic (Evans and Guishard 2009).

The thermodynamic ascent caused by anomalous cooling of the middle troposphere is enhanced by the intense low-level warm air advection that further increases the lapse rate. From t = −48 h the region where the cyclone will develop is subject to warm air advection (not shown), and 24 h before the genesis upward motion starts to occur (Fig. 13a). As the cyclone deepens, the pressure gradient with the SASH increases, strengthening the warm air advection eastward of the low due to more intense northerly winds (Figs. 13b,c). In the same region, the upward motion intensifies, reaching −0.4 Pa s−1, in part because of the enhanced thermodynamic destabilization and also by being located just below a region of upper-level diffluence in the eastern flank of the 250-hPa cyclonic circulation (Figs. 12b,c).

Fig. 13.
Fig. 13.

Austral summer (DJF) SC-relative composite of air temperature advection (shaded, in K day−1) and upward vertical velocity at 700 hPa (contours, in Pa s−1) at times (a) T = −24 h, (b) T = 0, and (c) T = +24 h.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

Evans and Guishard (2009) and Evans and Braun (2012) already mentioned the warm air advection as a fundamental mechanism to increase the atmospheric convective potential in hybrid cyclones that formed over relatively cold waters. This is true for the systems under study here that form over waters around 24°C and surface turbulent fluxes that do not exceed 90 W m−2 (Fig. 14a). This value does not differ much from the climatological mean for this region (Reboita et al. 2010a). At t = 0 and t = +24 h, the fluxes are even weaker (below 60 W m−2) due to cloudiness, precipitation, and the oceanic vertical mixing promoted by the cyclone (Figs. 14b,c). The summer SCs are not associated with very strong local surface heat fluxes, but there is a broad permanent region of intense heat fluxes (above 120 W m−2) and high SST to the northeast of the system, just over the northern flank of the SASH. By its position relative to the RG1, the circulation of the SASH advects not only warm but very moist air toward the region where SCs will develop. The inspection of moisture flux integrated in the troposphere and its divergence field (Figs. 14d–f) shows that the northeasterly flow transports great amounts of water vapor from this region toward the eastern side of the cyclone. It converges with a second northwesterly moisture flow from the continent, in a region of strong ascent, conducive to powerful convective development.

Fig. 14.
Fig. 14.

Austral summer (DJF) SC-relative composites of (a)–(c) SST (contours, in °C) and sensible + latent surface heat fluxes (shaded, in W m−2); (d)–(f) vertically integrated moisture flux (vectors, in kg m−1 s−1) and its divergence (shaded, in kg s−1). (left) T = −24 h, (middle) T = 0, (right) T = +24 h.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

2) Main differences between summer, spring, and autumn cyclogenesis environments

The main characteristics and the environment where the SCs are inserted in spring are almost the same as in summer. An incipient surface low is deepened by a westward PV anomaly in upper levels. Whereas in summer and autumn most SCs had a cutoff low at 500 hPa, the majority of spring cyclones counted on a shortwave trough forcing quasigeostrophic upward motions. Warm and moist air advection occur in a similar spatial pattern and magnitude compared to summer; the only difference is that in spring the moisture contribution almost entirely comes from the northeasterly SASH flow. The turbulent fluxes in the cyclone formation area are relatively weak again, but they are stronger in the northern flank of SASH (up to 150 W m−2).

The midtropospheric geopotential height field in autumn presents a stronger positive anomaly to the south of the cyclone, prior to and on the day of cyclogenesis (Figs. 15a,b). At the surface, the high pressure anomaly poleward of the autumn SCs arises due to transient anticyclones (in other seasons it is an extension of the SASH). As this anticyclone displaces to the south/southeast, the SCs also move farther south, placing the surface low under a pronounced easterly wind anomaly at 500 hPa (not shown). This results in a weaker vertical wind shear compared to summer or spring.

Fig. 15.
Fig. 15.

Austral autumn [March–May (MAM)] SC-relative anomaly composites of (a)–(c) 500-hPa geopotential height (contours, in m) and sea level pressure (shaded, in hPa); (d)–(f) SST (contours, in °C) and sensible + latent surface heat fluxes (shaded, in W m−2). (left) T = −24 h, (middle) T = 0, (right) for T = +24 h.

Citation: Journal of Climate 27, 22; 10.1175/JCLI-D-14-00149.1

Another remarkable difference of the autumn SCs is the configuration of the surface fluxes: they are more intense than in summer and spring over the whole South Atlantic basin, and the region of strongest fluxes is no longer northeastward of the SC, but southward of it (Figs. 15d,e). This area of stronger fluxes happens in response to the drier air and finer weather associated with high pressure systems. In this scenario, most of the turbulent latent and sensible heat fluxes fueling the cyclone may come from local contributions and from southern regions, by the southeasterly flow associated with the transient anticyclones.

The environment where SCs develop in autumn is the most conducive to tropical development, with strongest local turbulent fluxes and weakest vertical wind shear of all seasons. It is likely the reason why the most intense and organized SCs of the basin (Hurricane Catarina, and subtropical storms Anita and Arani) occurred during autumn.

4. Discussion and conclusions

A climatology of subtropical cyclones in the southwestern South Atlantic basin from the ERA-Interim and NCEP1 reanalyses is presented. The cyclone classification methodology follows mostly Evans and Guishard (2009) and Evans and Braun (2012), but here a relaxation in the maximum 925-hPa wind and vertical depth conditions is proposed to account for numerous shallow and weaker hybrid systems developing along the southeastern coast of South America.

A case study of such a shallow SC shows that the low-level warm core originated from low-level warm air masses advected from the continent to the sea. It does not develop over anomalously warm waters, but the horizontal moisture convergence seems to be the most important factor to sustain the low pressure region: once this convergence vanishes, the low weakens and disappears. This convergence may contribute to increase convection and latent heat release in the atmospheric column, deepening the surface cyclone. This mechanism described in numerous studies for extratropical cyclones (Gyakum 1983; Nuss and Anthes 1987, and others) is also fundamental for SC development. Sardie and Warner (1985) showed that the development of polar lows is more efficient when the latent heat release takes place in the lower troposphere, inducing greater convergence of warm and moist air near the surface. Satellite images show that the analyzed cyclone was not associated to deep convection, indicating that low tropospheric heating was a more important mechanism to maintain this SC than strong cumulus convection.

During the 1979–2011 period, the annual mean number of SCs is almost identical for ERAInt and NCEP1 (7.2 ± 2.8 and 7.1 ± 2.8 cyclones per year, respectively). This mean is much greater than presented by Evans and Braun (2012) for the same region, due to the inclusion of SCs that do not extend to the midtroposphere and systems that do not attain gale-force winds. However, each dataset presents distinct interannual variability, as indicated by the low correlation between the time series.

Regarding the complete lifetime of all tracked SCs, they have smaller traveled distance and slower velocity as compared to the extratropical ones, allowing them to interact more efficiently with the unstable environment in which they develop and to have an even greater impact on the South American coastal climate. Analysis of the complete lifetime indicates that extratropical transition takes place for some South Atlantic SCs.

The 925-hPa minimum relative vorticity and maximum strength of the low-level warm core attained during the SCs’ subtropical stage are similar in NCEP1 and ERAInt. The latter reanalysis is more skilled to detect the weakest and strongest cyclones, due to the finer resolution.

The difference between ERAInt and NCEP1 cyclogenesis annual cycle might be due to differences in the data horizontal resolution (as SCs have a smaller horizontal scale than extratropical ones, and the CPS values are sensitive to changes in resolution). However, the fact that ERAInt does not systematically present a higher number of annual cyclones indicates that this may not be the only source of discrepancy. The area under study has few conventional data, making reanalyses more dependent on their numerical models and data assimilation techniques. The difference in these features between ERAInt and NCEP1 may impact the SC formation throughout the years. It is important to notice, though, that once these systems are formed, their kinematic characteristics, strength, cyclogenesis and cyclolysis regions, and seasonal variability are consistently similar in both reanalyses.

The coastal region near southwestern Brazil (RG1) is the main subtropical cyclogenetic area in the South Atlantic basin. In fact, during the most active season (summer) more than ⅓ of the cyclones tracked over RG1 are of hybrid nature.

Composite analyses from a set of 126 SCs initiated over the RG1 showed that they form to the east of a PV minimum at 250 hPa, in a region of reduced vertical wind shear and upper-level mass divergence. For most of them, a Rex blocking pattern develops in the upper and middle troposphere, further reducing the vertical shear and also preventing the cyclones from being steered. This circulation anomaly in summer SC genesis days, with an anticyclone to the south/southwest of a cyclone placed over the surface low, is similar to the pattern presented in the composites of Evans and Braun (2012) for the same season.

Turbulent sensible and latent heat fluxes over the cyclogenesis region are close to their climatological mean value in summer [in agreement with the summer weak SST anomaly field of Evans and Braun (2012)] and spring. These fluxes tend to further decrease as the subtropical cyclone develops due to cloud coverage, precipitation, and vertical mixing of the upper ocean layers. To destabilize the environment and allow the SCs to develop, low-level warm and moist air advection takes place during all the cyclone lifetime (and even 24 h before genesis), mainly due to the SASH flow bringing a great amounts of moisture originating in the tropical South Atlantic (northeastward of the cyclone, around 5°S). During the summer months, an important secondary moisture source comes from the continent, transporting moisture from the Northern Hemisphere tropical Atlantic and from the Amazon region to the RG1. These transports favor an unstable stratification of the atmosphere over the RG1, leading to an enhanced convective potential for the SC. In autumn, more intense local turbulent fluxes and a weaker vertical wind shear promote the most “tropical” environment of all seasons.

This most favorable environment tends to produce stronger and more organized SCs in autumn than in summer, although most of the cyclones develop in the latter season. This raises the question of what season is then more likely to cause damages and monetary losses in the region. To quantify this information, we have used the accumulated cyclone energy (ACE) index. ACE is obtained by summing the square of 6-hourly maximum wind speed for all subtropical cyclones while they present intensity greater than tropical storm, merging information on the frequency, intensity, and lifetime of the systems (Bell et al. 2000; Camargo and Sobel 2005). Summer presented a higher ACE than autumn for both reanalyses (the seasonal values were 5.3 × 105 m2 s−2 and 2.2 × 105 m2 s−2, respectively, in ERAInt). These values indicate that summer is the most active season not only in frequency, but also in the combined cyclones’ intensity. Along with the fact that in this season the SCs present a more stationary development along the coastal southwestern Brazil (near the largest port of Latin America and over a concentration of many oil platforms), it is during summer that these systems can cause more financial and societal impacts.

The composites presented in this study were constructed for both deep and shallow SCs, because composites of shallow SCs (not shown) do not differ much from the deep ones. The overall structure of the cyclone, as well as the heat and moisture horizontal advections and turbulent surface fluxes patterns, is very similar in both cases. The main difference is observed in the magnitude of the fields. Notably, shallow cases present less pronounced upper-level PV anomalies and weaker horizontal moisture flux convergence. This is verified in all three analyzed seasons.

Results presented in this study add to the incipient investigation of SCs over the South Atlantic basin, assessing their seasonal variability, how two different datasets represent the cyclogenesis climatology, and the main synoptic features. It is particularly interesting how these systems develop and sustain themselves over a region of relatively weak turbulent heat fluxes. The moisture that is locally provided by the ocean to the development of hybrid cyclones in other regions of the globe seems to be reduced in the South Atlantic SCs. However, the moisture advection from distant regions may balance this weak local surface flux, offering the necessary moisture to fuel convection and warm the low troposphere by latent heat release. This mechanism maintains the SC. A more detailed study of these nonlocal moisture sources, applying a Lagrangian particle dispersion model, is ongoing research and will be presented in a future paper.

Acknowledgments

We thank Jean Peres and Livia Dutra for the help in automating the cyclone tracking and sorting algorithm. Thanks to Rodrigo Bombardi for providing assistance with the spherical kernel algorithm. Also we appreciate the helpful comments from the anonymous reviewers. This work was supported by CAPES/PROEX and the Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq) Grants 558121/2009-8, 307202/2011-9, 140839/2011-9, and 481942/2013-0.

REFERENCES

  • Bell, G. D., and Coauthors, 2000: Climate assessment for 1999. Bull. Amer. Meteor. Soc., 81, S1S50, doi:10.1175/1520-0477(2000)81[s1:CAF]2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Beven, J. L., 1997: A study of three “hybrid” storms. Preprints, 22nd Conf. on Hurricanes and Tropical Meteorology, Fort Collins, CO, Amer. Meteor. Soc., 645–646.

  • Bombardi, R. J., L. M. Carvalho, C. Jones, and M. S. Reboita, 2014: Precipitation over eastern South America and the South Atlantic sea surface temperature during neutral ENSO periods. Climate Dyn., 42, 1553–1568, doi:10.1007/s00382-013-1832-7.

    • Search Google Scholar
    • Export Citation
  • Braun, A. J., 2009: A comparison between South Atlantic and Tasman Sea subtropical storms. M.S. thesis, Department of Meteorology, The Pennsylvania State University, 150 pp.

  • Camargo, S. J., and A. H. Sobel, 2005: Western North Pacific tropical cyclone intensity and ENSO. J. Climate, 18, 29963006, doi:10.1175/JCLI3457.1.

    • Search Google Scholar
    • Export Citation
  • Carrasco, J. F., D. H. Bromwich, and A. J. Monaghan, 2003: Distribution and characteristics of mesoscale cyclones in the Antarctic: Ross Sea eastward to the Weddell Sea. Mon. Wea. Rev., 131, 289301, doi:10.1175/1520-0493(2003)131<0289:DACOMC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Dias Pinto, J. R., and R. P. da Rocha, 2011: The energy cycle and structural evolution of cyclones over southeastern South America in three case studies. J. Geophys. Res., 116, D14112, doi:10.1029/2011JD016217.

    • Search Google Scholar
    • Export Citation
  • Dias Pinto, J. R., M. S. Reboita, and R. P. da Rocha, 2013: Synoptic and dynamical analysis of subtropical cyclone Anita (2010) and its potential for tropical transition over the South Atlantic Ocean. J. Geophys. Res. Atmos., 118, 10 87010 883, doi:10.1002/jgrd.50830.

    • Search Google Scholar
    • Export Citation
  • Evans, J. L., and M. P. Guishard, 2009: Atlantic subtropical storms. Part I: Diagnostic criteria and composite analysis. Mon. Wea. Rev., 137, 20652080, doi:10.1175/2009MWR2468.1.

    • Search Google Scholar
    • Export Citation
  • Evans, J. L., and A. Braun, 2012: A climatology of subtropical cyclones in the South Atlantic. J. Climate, 25, 73287340, doi:10.1175/JCLI-D-11-00212.1.

    • Search Google Scholar
    • Export Citation
  • Gan, M. A., and V. B. Rao, 1991: Surface cyclogenesis over South America. Mon. Wea. Rev., 119, 12931302, doi:10.1175/1520-0493(1991)119<1293:SCOSA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Guishard, M. P., J. L. Evans, and R. E. Hart, 2009: Atlantic subtropical storms. Part II: Climatology. J. Climate, 22, 35743594, doi:10.1175/2008JCLI2346.1.

    • Search Google Scholar
    • Export Citation
  • Gyakum, J. R., 1983: On the evolution of the QE II storm: Dynamic and thermodynamic structure. Mon. Wea. Rev., 111, 11561173, doi:10.1175/1520-0493(1983)111<1156:OTEOTI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hanley, D., J. Molinari, and D. Keyser, 2001: A composite study of the interactions between tropical cyclones and upper-tropospheric troughs. Mon. Wea. Rev., 129, 25702584, doi:10.1175/1520-0493(2001)129<2570:ACSOTI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hanson, C. E., J. P. Palutikof, and T. D. Davies, 2004: Objective cyclone climatologies of the North Atlantic—A comparison between the ECMWF and NCEP reanalyses. Climate Dyn., 22, 757769, doi:10.1007/s00382-004-0415-z.

    • Search Google Scholar
    • Export Citation
  • Hart, R. E., 2003: A cyclone phase space derived from thermal wind and thermal asymmetry. Mon. Wea. Rev., 131, 585616, doi:10.1175/1520-0493(2003)131<0585:ACPSDF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hodges, K. I., 1996: Spherical nonparametric estimators applied to the UGAMP model integration for AMIP. Mon. Wea. Rev., 124, 29142932, doi:10.1175/1520-0493(1996)124<2914:SNEATT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hodges, K. I., R. W. Lee, and L. Bengtsson, 2011: A comparison of extratropical cyclones in recent reanalyses ERA-Interim, NASA MERRA, NCEP CFSR, and JRA-25. J. Climate, 24, 48884906, doi:10.1175/2011JCLI4097.1.

    • Search Google Scholar
    • Export Citation
  • Hólm, E. V., 2002: Revision of the ECMWF humidity analysis: Construction of a Gaussian control variable. Proc. ECMWF/GEWEX Workshop on Humidity Analysis, Reading, United Kingdom, ECMWF. [Available online at http://old.ecmwf.int/publications/library/ecpublications/_pdf/workshop/2002/Humidity/holm.pdf.]

  • Hoskins, B. J., and K. I. Hodges, 2005: A new perspective on Southern Hemisphere storm tracks. J. Climate, 18, 41084129, doi:10.1175/JCLI3570.1.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., M. E. McIntyre, and A. W. Robertson, 1985: On the use and significance of isentropic potential vorticity maps. Quart. J. Roy. Meteor. Soc., 111, 877946, doi:10.1002/qj.49711147002.

    • Search Google Scholar
    • Export Citation
  • Huffman, G. J., and Coauthors, 2007: The TRMM Multisatellite Precipitation Analysis (TMPA): Quasi-global, multiyear, combined-sensor precipitation estimates at fine scales. J. Hydrometeor., 8, 3855, doi:10.1175/JHM560.1.

    • Search Google Scholar
    • Export Citation
  • Hulme, A. L., and J. E. Martin, 2009: Synoptic- and frontal-scale influences on tropical transition events in the Atlantic basin. Part I: A six-case survey. Mon. Wea. Rev., 137, 36053625, doi:10.1175/2009MWR2802.1.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437471, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Krüger, L. F., R. P. da Rocha, M. S. Reboita, and T. Ambrizzi, 2012: RegCM3 nested in HadAM3 scenarios A2 and B2: Projected changes in extratropical cyclogenesis, temperature and precipitation over the South Atlantic Ocean. Climatic Change, 113, 599621, doi:10.1007/s10584-011-0374-4.

    • Search Google Scholar
    • Export Citation
  • McTaggart-Cowan, R., L. F. Bosart, C. A. Davis, E. H. Atallah, J. R. Gyakum, and K. A. Emanuel, 2006: Analysis of Hurricane Catarina (2004). Mon. Wea. Rev., 134, 30293053, doi:10.1175/MWR3330.1.

    • Search Google Scholar
    • Export Citation
  • Mendes, D., E. P. Souza, J. A. Marengo, and M. C. D. Mendes, 2010: Climatology of extratropical cyclones over the South American–southern oceans sector. Theor. Appl. Climatol., 100, 239250, doi:10.1007/s00704-009-0161-6.

    • Search Google Scholar
    • Export Citation
  • Nuss, W. A., and R. A. Anthes, 1987: A numerical investigation of low-level processes in rapid cyclogenesis. Mon. Wea. Rev., 115, 27282743, doi:10.1175/1520-0493(1987)115<2728:ANIOLL>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Quadro, M. F. L., 2012: Estudo de vortices ciclonicos de mesoescala associados a zona de convergencia do Atlantico Sul. Ph.D. thesis, Universidade de São Paulo, 158 pp.

  • Reboita, M. S., R. P. da Rocha, T. Ambrizzi, and E. Caetano, 2010a: An assessment of the latent and sensible heat flux on the simulated regional climate over southwestern South Atlantic Ocean. Climate Dyn., 34, 873889, doi:10.1007/s00382-009-0681-x.

    • Search Google Scholar
    • Export Citation
  • Reboita, M. S., R. P. da Rocha, T. Ambrizzi, and S. Sugahara, 2010b: South Atlantic ocean cyclogenesis climatology simulated by regional climate model (RegCM3). Climate Dyn., 35, 13311347, doi:10.1007/s00382-009-0668-7.

    • Search Google Scholar
    • Export Citation
  • Reboita, M. S., R. P. da Rocha, and T. Ambrizzi, 2012: Dynamic and climatological features of cyclonic developments over southwestern South Atlantic Ocean. Horizons in Earth Science Research, Vol. 6, Nova Science Publishers, 135–160.

  • Rex, D. F., 1950: Blocking action in the middle troposphere and its effect upon regional climate. I. An aerological study of blocking action. Tellus, 2, 196211, doi:10.1111/j.2153-3490.1950.tb00331.x.

    • Search Google Scholar
    • Export Citation
  • Ritchie, E. A., and R. L. Elsberry, 2001: Simulations of the transformation stage of the extratropical transition of tropical cyclones. Mon. Wea. Rev., 129, 14621480, doi:10.1175/1520-0493(2001)129<1462:SOTTSO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Sardie, J. M., and T. T. Warner, 1985: A numerical study of the development mechanisms of polar lows. Tellus, 37A, 460477, doi:10.1111/j.1600-0870.1985.tb00444.x.

    • Search Google Scholar
    • Export Citation
  • Simmonds, I., and K. Keay, 2000: Mean Southern Hemisphere extratropical cyclone behavior in the 40-year NCEP–NCAR reanalysis. J. Climate, 13, 873885, doi:10.1175/1520-0442(2000)013<0873:MSHECB>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Sinclair, M. R., 1994: An objective cyclone climatology for the Southern Hemisphere. Mon. Wea. Rev., 122, 22392256, doi:10.1175/1520-0493(1994)122<2239:AOCCFT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Sinclair, M. R., 1995: A climatology of cyclogenesis for the Southern Hemisphere. Mon. Wea. Rev., 123, 16011619, doi:10.1175/1520-0493(1995)123<1601:ACOCFT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Strachan, J., P. L. Vidale, K. Hodges, M. Roberts, and M.-E. Demory, 2013: Investigating global tropical cyclone activity with a hierarchy of AGCMs: The role of model resolution. J. Climate, 26, 133152, doi:10.1175/JCLI-D-12-00012.1.

    • Search Google Scholar
    • Export Citation
  • Sugahara, S., 2000: Variação anual da frequência de ciclones no Atlântico Sul. Proc. 11th Brazilian Congress of Meteorology, Vol. 11, Rio de Janeiro, Brazil, Brazilian Society of Meteorology (SBMET), 2607–2611.

  • Tilinina, N., S. K. Gulev, I. Rudeva, and P. Koltermann, 2013: Comparing cyclone life cycle characteristics and their interannual variability in different reanalyses. J. Climate, 26, 64196438, doi:10.1175/JCLI-D-12-00777.1.

    • Search Google Scholar
    • Export Citation
  • Yu, L., X. Jin, and R. A. Weller, 2008: Multidecade global flux datasets from the Objectively Analyzed Air-sea Fluxes (OAFlux) Project: Latent and sensible heat fluxes, ocean evaporation, and related surface meteorological variables. OAFlux Project Tech. Rep. OA-2008-01, 64 pp.

Save
  • Bell, G. D., and Coauthors, 2000: Climate assessment for 1999. Bull. Amer. Meteor. Soc., 81, S1S50, doi:10.1175/1520-0477(2000)81[s1:CAF]2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Beven, J. L., 1997: A study of three “hybrid” storms. Preprints, 22nd Conf. on Hurricanes and Tropical Meteorology, Fort Collins, CO, Amer. Meteor. Soc., 645–646.

  • Bombardi, R. J., L. M. Carvalho, C. Jones, and M. S. Reboita, 2014: Precipitation over eastern South America and the South Atlantic sea surface temperature during neutral ENSO periods. Climate Dyn., 42, 1553–1568, doi:10.1007/s00382-013-1832-7.

    • Search Google Scholar
    • Export Citation
  • Braun, A. J., 2009: A comparison between South Atlantic and Tasman Sea subtropical storms. M.S. thesis, Department of Meteorology, The Pennsylvania State University, 150 pp.

  • Camargo, S. J., and A. H. Sobel, 2005: Western North Pacific tropical cyclone intensity and ENSO. J. Climate, 18, 29963006, doi:10.1175/JCLI3457.1.

    • Search Google Scholar
    • Export Citation
  • Carrasco, J. F., D. H. Bromwich, and A. J. Monaghan, 2003: Distribution and characteristics of mesoscale cyclones in the Antarctic: Ross Sea eastward to the Weddell Sea. Mon. Wea. Rev., 131, 289301, doi:10.1175/1520-0493(2003)131<0289:DACOMC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Dias Pinto, J. R., and R. P. da Rocha, 2011: The energy cycle and structural evolution of cyclones over southeastern South America in three case studies. J. Geophys. Res., 116, D14112, doi:10.1029/2011JD016217.

    • Search Google Scholar
    • Export Citation
  • Dias Pinto, J. R., M. S. Reboita, and R. P. da Rocha, 2013: Synoptic and dynamical analysis of subtropical cyclone Anita (2010) and its potential for tropical transition over the South Atlantic Ocean. J. Geophys. Res. Atmos., 118, 10 87010 883, doi:10.1002/jgrd.50830.

    • Search Google Scholar
    • Export Citation
  • Evans, J. L., and M. P. Guishard, 2009: Atlantic subtropical storms. Part I: Diagnostic criteria and composite analysis. Mon. Wea. Rev., 137, 20652080, doi:10.1175/2009MWR2468.1.

    • Search Google Scholar
    • Export Citation
  • Evans, J. L., and A. Braun, 2012: A climatology of subtropical cyclones in the South Atlantic. J. Climate, 25, 73287340, doi:10.1175/JCLI-D-11-00212.1.

    • Search Google Scholar
    • Export Citation
  • Gan, M. A., and V. B. Rao, 1991: Surface cyclogenesis over South America. Mon. Wea. Rev., 119, 12931302, doi:10.1175/1520-0493(1991)119<1293:SCOSA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Guishard, M. P., J. L. Evans, and R. E. Hart, 2009: Atlantic subtropical storms. Part II: Climatology. J. Climate, 22, 35743594, doi:10.1175/2008JCLI2346.1.

    • Search Google Scholar
    • Export Citation
  • Gyakum, J. R., 1983: On the evolution of the QE II storm: Dynamic and thermodynamic structure. Mon. Wea. Rev., 111, 11561173, doi:10.1175/1520-0493(1983)111<1156:OTEOTI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hanley, D., J. Molinari, and D. Keyser, 2001: A composite study of the interactions between tropical cyclones and upper-tropospheric troughs. Mon. Wea. Rev., 129, 25702584, doi:10.1175/1520-0493(2001)129<2570:ACSOTI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hanson, C. E., J. P. Palutikof, and T. D. Davies, 2004: Objective cyclone climatologies of the North Atlantic—A comparison between the ECMWF and NCEP reanalyses. Climate Dyn., 22, 757769, doi:10.1007/s00382-004-0415-z.

    • Search Google Scholar
    • Export Citation
  • Hart, R. E., 2003: A cyclone phase space derived from thermal wind and thermal asymmetry. Mon. Wea. Rev., 131, 585616, doi:10.1175/1520-0493(2003)131<0585:ACPSDF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hodges, K. I., 1996: Spherical nonparametric estimators applied to the UGAMP model integration for AMIP. Mon. Wea. Rev., 124, 29142932, doi:10.1175/1520-0493(1996)124<2914:SNEATT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hodges, K. I., R. W. Lee, and L. Bengtsson, 2011: A comparison of extratropical cyclones in recent reanalyses ERA-Interim, NASA MERRA, NCEP CFSR, and JRA-25. J. Climate, 24, 48884906, doi:10.1175/2011JCLI4097.1.

    • Search Google Scholar
    • Export Citation
  • Hólm, E. V., 2002: Revision of the ECMWF humidity analysis: Construction of a Gaussian control variable. Proc. ECMWF/GEWEX Workshop on Humidity Analysis, Reading, United Kingdom, ECMWF. [Available online at http://old.ecmwf.int/publications/library/ecpublications/_pdf/workshop/2002/Humidity/holm.pdf.]

  • Hoskins, B. J., and K. I. Hodges, 2005: A new perspective on Southern Hemisphere storm tracks. J. Climate, 18, 41084129, doi:10.1175/JCLI3570.1.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., M. E. McIntyre, and A. W. Robertson, 1985: On the use and significance of isentropic potential vorticity maps. Quart. J. Roy. Meteor. Soc., 111, 877946, doi:10.1002/qj.49711147002.

    • Search Google Scholar
    • Export Citation
  • Huffman, G. J., and Coauthors, 2007: The TRMM Multisatellite Precipitation Analysis (TMPA): Quasi-global, multiyear, combined-sensor precipitation estimates at fine scales. J. Hydrometeor., 8, 3855, doi:10.1175/JHM560.1.

    • Search Google Scholar
    • Export Citation
  • Hulme, A. L., and J. E. Martin, 2009: Synoptic- and frontal-scale influences on tropical transition events in the Atlantic basin. Part I: A six-case survey. Mon. Wea. Rev., 137, 36053625, doi:10.1175/2009MWR2802.1.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437471, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Krüger, L. F., R. P. da Rocha, M. S. Reboita, and T. Ambrizzi, 2012: RegCM3 nested in HadAM3 scenarios A2 and B2: Projected changes in extratropical cyclogenesis, temperature and precipitation over the South Atlantic Ocean. Climatic Change, 113, 599621, doi:10.1007/s10584-011-0374-4.

    • Search Google Scholar
    • Export Citation
  • McTaggart-Cowan, R., L. F. Bosart, C. A. Davis, E. H. Atallah, J. R. Gyakum, and K. A. Emanuel, 2006: Analysis of Hurricane Catarina (2004). Mon. Wea. Rev., 134, 30293053, doi:10.1175/MWR3330.1.

    • Search Google Scholar
    • Export Citation
  • Mendes, D., E. P. Souza, J. A. Marengo, and M. C. D. Mendes, 2010: Climatology of extratropical cyclones over the South American–southern oceans sector. Theor. Appl. Climatol., 100, 239250, doi:10.1007/s00704-009-0161-6.

    • Search Google Scholar
    • Export Citation
  • Nuss, W. A., and R. A. Anthes, 1987: A numerical investigation of low-level processes in rapid cyclogenesis. Mon. Wea. Rev., 115, 27282743, doi:10.1175/1520-0493(1987)115<2728:ANIOLL>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Quadro, M. F. L., 2012: Estudo de vortices ciclonicos de mesoescala associados a zona de convergencia do Atlantico Sul. Ph.D. thesis, Universidade de São Paulo, 158 pp.

  • Reboita, M. S., R. P. da Rocha, T. Ambrizzi, and E. Caetano, 2010a: An assessment of the latent and sensible heat flux on the simulated regional climate over southwestern South Atlantic Ocean. Climate Dyn., 34, 873889, doi:10.1007/s00382-009-0681-x.

    • Search Google Scholar
    • Export Citation
  • Reboita, M. S., R. P. da Rocha, T. Ambrizzi, and S. Sugahara, 2010b: South Atlantic ocean cyclogenesis climatology simulated by regional climate model (RegCM3). Climate Dyn., 35, 13311347, doi:10.1007/s00382-009-0668-7.

    • Search Google Scholar
    • Export Citation
  • Reboita, M. S., R. P. da Rocha, and T. Ambrizzi, 2012: Dynamic and climatological features of cyclonic developments over southwestern South Atlantic Ocean. Horizons in Earth Science Research, Vol. 6, Nova Science Publishers, 135–160.

  • Rex, D. F., 1950: Blocking action in the middle troposphere and its effect upon regional climate. I. An aerological study of blocking action. Tellus, 2, 196211, doi:10.1111/j.2153-3490.1950.tb00331.x.

    • Search Google Scholar
    • Export Citation
  • Ritchie, E. A., and R. L. Elsberry, 2001: Simulations of the transformation stage of the extratropical transition of tropical cyclones. Mon. Wea. Rev., 129, 14621480, doi:10.1175/1520-0493(2001)129<1462:SOTTSO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Sardie, J. M., and T. T. Warner, 1985: A numerical study of the development mechanisms of polar lows. Tellus, 37A, 460477, doi:10.1111/j.1600-0870.1985.tb00444.x.

    • Search Google Scholar
    • Export Citation
  • Simmonds, I., and K. Keay, 2000: Mean Southern Hemisphere extratropical cyclone behavior in the 40-year NCEP–NCAR reanalysis. J. Climate, 13, 873885, doi:10.1175/1520-0442(2000)013<0873:MSHECB>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Sinclair, M. R., 1994: An objective cyclone climatology for the Southern Hemisphere. Mon. Wea. Rev., 122, 22392256, doi:10.1175/1520-0493(1994)122<2239:AOCCFT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Sinclair, M. R., 1995: A climatology of cyclogenesis for the Southern Hemisphere. Mon. Wea. Rev., 123, 16011619, doi:10.1175/1520-0493(1995)123<1601:ACOCFT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Strachan, J., P. L. Vidale, K. Hodges, M. Roberts, and M.-E. Demory, 2013: Investigating global tropical cyclone activity with a hierarchy of AGCMs: The role of model resolution. J. Climate, 26, 133152, doi:10.1175/JCLI-D-12-00012.1.

    • Search Google Scholar
    • Export Citation
  • Sugahara, S., 2000: Variação anual da frequência de ciclones no Atlântico Sul. Proc. 11th Brazilian Congress of Meteorology, Vol. 11, Rio de Janeiro, Brazil, Brazilian Society of Meteorology (SBMET), 2607–2611.

  • Tilinina, N., S. K. Gulev, I. Rudeva, and P. Koltermann, 2013: Comparing cyclone life cycle characteristics and their interannual variability in different reanalyses. J. Climate, 26, 64196438, doi:10.1175/JCLI-D-12-00777.1.

    • Search Google Scholar
    • Export Citation
  • Yu, L., X. Jin, and R. A. Weller, 2008: Multidecade global flux datasets from the Objectively Analyzed Air-sea Fluxes (OAFlux) Project: Latent and sensible heat fluxes, ocean evaporation, and related surface meteorological variables. OAFlux Project Tech. Rep. OA-2008-01, 64 pp.

  • Fig. 1.

    Effective cyclone tracking area (SAO, blue dashed rectangle) and cyclogenetic region 1 (RG1, black solid rectangle).

  • Fig. 2.

    Cyclone phase diagram of a shallow hybrid cyclone from 0000 UTC 12 Jan to 1800 UTC 13 Jan 2008: (a) B vs −VTL and (b) −VTL vs −VTU.

  • Fig. 3.

    GOES-10 IR images at (a) 0000 UTC 12 Jan, (b) 1200 UTC 12 Jan, (c) 0000 UTC 13 Jan, and (d) 1200 UTC 13 Jan 2008. The capital “L” indicates the low pressure center according to the tracking algorithm.

  • Fig. 4.

    Synoptic fields at 1200 UTC 11 Jan 2008, from ERAInt. (a) The 925-hPa horizontal wind (vectors, in m s−1) and cyclonic relative vorticity (dashed lines, in s−1); daily precipitation (shaded, in mm). (b) Vertically integrated moisture flux (vectors, in kg m−1 s−1) and its divergence (shaded, in kg s−1). (c) Air temperature advection (shaded, in K day−1) and mean sea level pressure (contours, in hPa). (d) Longitudinal cross section of geopotential height (m) and air temperature (K) anomalies from the 24°S zonal mean.

  • Fig. 5.

    As in Fig. 4, but at 1200 UTC 12 Jan 2008.

  • Fig. 6.

    As in Fig. 4, but at 0000 UTC 13 Jan 2008.

  • Fig. 7.

    (a) Annual frequency and (b) seasonal mean of SC over SAO for ERAInt (blue) and NCEP1 (orange). In (b), the bar plot is the seasonal mean and the lines represent the seasonal percentage of SC in the total amount of cyclones.

  • Fig. 8.

    Histograms of the (a) lifetime (in days), (b) distance traveled (in km), and (c) mean velocity (in m s−1) for SC over SAO in ERAInt (light gray) and NCEP1 (dark gray).

  • Fig. 9.

    ERAInt (light gray) and NCEP1 (dark gray) SAO SCs histograms of (a) minimum cyclonic vorticity (in 10−5 s−1), (b) maximum wind speed attained in the subtropical stage, (c) radius of maximum wind speed during the subtropical stage, and (d) maximum −VTL.

  • Fig. 10.

    Subtropical cyclogenesis and cyclolysis (number of cyclones by square radian per day) for (a),(c) NCEP1 and (b),(d) ERAInt. (e) Seasonal mean (bar plot) and percentage of SC in the total amount of cyclones over the RG1 for ERAInt (blue) and NCEP1 (orange).

  • Fig. 11.

    Austral summer [December–February (DJF)] SC-relative composites of (a)–(c) sea level pressure (contours, in hPa) and PV at 250 hPa (shaded, in PVU); (d)–(f) vertical cross sections of PV (shaded, in PVU) and potential temperature (contours, in K), for (left) T = −24 h prior to the cyclogenesis, (middle) T = 0 (1200 UTC on the genesis day), and (right) T = +24 h.

  • Fig. 12.

    Austral summer (DJF) SC-relative anomaly composites of (a)–(c) 250-hPa flow (streamlines) and 250-hPa magnitude of horizontal wind (shaded, in m s−1); (d)–(f) 500-hPa geopotential height (contours, in m) and sea level pressure (shaded, in hPa). Anomaly fields are the difference between days of cyclogenesis and the DJF mean. (left) T = −24 h, (middle) T = 0, (right) T = +24 h.

  • Fig. 13.

    Austral summer (DJF) SC-relative composite of air temperature advection (shaded, in K day−1) and upward vertical velocity at 700 hPa (contours, in Pa s−1) at times (a) T = −24 h, (b) T = 0, and (c) T = +24 h.

  • Fig. 14.

    Austral summer (DJF) SC-relative composites of (a)–(c) SST (contours, in °C) and sensible + latent surface heat fluxes (shaded, in W m−2); (d)–(f) vertically integrated moisture flux (vectors, in kg m−1 s−1) and its divergence (shaded, in kg s−1). (left) T = −24 h, (middle) T = 0, (right) T = +24 h.

  • Fig. 15.

    Austral autumn [March–May (MAM)] SC-relative anomaly composites of (a)–(c) 500-hPa geopotential height (contours, in m) and sea level pressure (shaded, in hPa); (d)–(f) SST (contours, in °C) and sensible + latent surface heat fluxes (shaded, in W m−2). (left) T = −24 h, (middle) T = 0, (right) for T = +24 h.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 884 333 21
PDF Downloads 727 294 25