SEARCH

SEARCH BY CITATION

Keywords:

  • cut-off low;
  • energetics;
  • ageostrophic flux convergence;
  • baroclinic conversion;
  • barotropic conversion

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and methodology
  5. 3. Results
  6. 4. Conclusions
  7. Acknowledgements
  8. References

The occurrence of cut-off lows (COLs) over South Pacific can be seen on different synoptic maps and on satellite imageries. However, the formation and maintenance mechanisms of the COLs are not well understood. An energetics analysis of a COL case through the eddy kinetic energy (EKE) equation is the aim of this study. The main terms analysed were ageostrophic flux convergence (AFC), baroclinic (BRC) and barotropic (BRT) conversions. This analysis shows that the formation mechanism of the COL was associated with BRC (COL western side) and AFC (COL eastern side). After the maximum intensity period, the BRC term was negative and the AFC positive.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and methodology
  5. 3. Results
  6. 4. Conclusions
  7. Acknowledgements
  8. References

One of the features in the subtropical upper troposphere of the Southern Hemisphere is the frequent existence of cut-off lows (COLs) (known also as upper tropospheric cyclonic vortex) throughout the year (Fuenzalida et al., 2005; Campetella and Possia, 2007). COLs are defined as cold lows that have grown out of a trough and become displaced out of the basic westerly current and lie equatorward of this current. These systems are more frequent during summer, autumn, and winter at 200, 300, and 500 hPa, respectively (Reboita et al., 2010) and generally are confined in the middle and upper troposphere (Campetella and Possia, 2007; Reboita et al., 2010), but they can sometimes extend to the surface with an equivalent barotropic structure (Funatsu et al., 2004). These systems are characterized by a cold core and a direct thermal circulation, with sinking of cold air in their center and rising of warm air on the periphery (Frank, 1970). They are generally smaller than extratropical cyclones during its mature stage (Nieto et al., 2005). The cloud pattern associated with the COL depends on the direct thermal circulation, the vortex location and its direction of movement (Kousky and Gan, 1981). The coldest temperatures are found at mid-levels of the troposphere, while temperatures near the surface are relatively unaffected by the presence of the vortex (Erickson, 1971; Kousky and Gan, 1981; Gan and Kousky, 1986). Some cumulonimbus activity may be present near the vortex center, especially over continental areas where this activity shows a marked diurnal variation (Kousky and Gan, 1981). When located over oceanic areas, these systems do not normally dissipate, but become absorbed into the mid-latitude westerlies by amplified troughs that pass by at higher latitudes (Ramage, 1962). However, Garreaud and Fuenzalida (2007) based on numerical results concluded that diabatic heating associated with latent heat release plays an important role on the COL dissipation.

The COLs can be classified as Palmén type when they are formed over subtropical latitudes and Palmer type when formed over tropical latitudes. As these systems are formed in different regions, they have some differences as the subtropical have formed throughout the year and the tropical from spring to autumn with more frequency during the summer months (Palmer, 1951; Kousky and Gan, 1981). The Palmén COLs are formed when large-scale amplified ridge in mid-latitudes with strong lower- and mid-tropospheric warm-air advection to the west of the ridge, and a short-wave trough jet streak system that intensified due to a downward intrusion of stratospheric air (Bell and Bosart, 1993). The formation of Parmer COLs is not yet explained; however, Kousky and Gan (1981) proposed a formation mechanism for Northeast of Brazil COL. These systems are formed when an upstream ridge associated with a cold frontal system penetrating to low latitudes intensifies associated with the strong warm advection, especially at low and mid-levels. The amplification of the upper level ridge downstream aids the amplification of the next downstream trough.

Mishra et al. (2001) observed for the Northeast Brazil COLs that the intensification of the Bolivian anticyclone and an associated ridge along with their eastward shift, east–west orientation of the South Atlantic trough and its intensification, and the presence of a trough in the equatorial westerlies contributed to the development of a narrow shear zone in the belt 17.58–7.58S during the prevortex period. The shear zone strongly satisfied the necessary condition for barotropic instability just before the vortex formation.

Most of the South Hemisphere Palmén COL type have a period of 2 or 3 days (Fuenzalida et al., 2005; Campetella and Possia, 2007). However, the Palmer COLs formed over Northeast of Brazil have a period around 7 days (Ramirez et al., 1999).

Some of the COLs formed over the eastern South Pacific can cross the Andes Cordillera and are responsible for the occurrence of unseasonal rainfall along the semiarid western slope of the Andes cordillera (Garreaud and Fuenzalida, 2007), convection in the central region of South America (Pizarro and Montecinos, 2000), frost events in southern Brazil (Fortune and Kousky, 1983) and the development of both Andes lee cyclones (Funatsu et al., 2004) and intense cyclones over Southern Brazil (Gan and Rao, 1996).

Although there is a conceptual model to explain the formation of COLs, what causes the formation and maintenance of these systems is not understood yet. In contrast, the formation of mid-latitude disturbances in the atmosphere has been explained by baroclinic, barotropic, non-modal instabilities and wave energy propagation, or downstream development (Nielsen-Gammon, 1995; Piva et al., 2010).

So, the aim of this study is to determine the relative importance of the terms of the eddy kinetic energy (EKE) equation for a case of COL that formed over the southeastern Pacific Ocean on 19 August 2006 at approximately 104°W, 33°S. This case was selected to be studied in details because it was over the ocean during most of its life and its lifetime of 7 days was one of the longest during the period analysed, i.e. from 1 March 2005 to 29 February 2008.

This article is organized as follows: in Section 'Data and methodology' the data and the methodology is discussed, in Section 'Results' the analysis of the energetics is provided and in Section 'Conclusions' the conclusions are presented.

2. Data and methodology

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and methodology
  5. 3. Results
  6. 4. Conclusions
  7. Acknowledgements
  8. References

In this study, we used GOES water vapour imageries and the 6-hourly gridded data from the National Centers for Environmental Prediction/Department of Energy (NCEP/DOE) reanalysis (Kanamitsu et al., 2002) for the period from 19 to 26 August 2006. The variable used are the zonal (u), meridional (v) and vertical (ω) wind components, geopotential high (z) and surface pressure (p).

The COL energetics are studied by the use of EKE (K′) equation developed by Orlanski and Katzfey (1991) and modified by (Chang, 2000):

  • display math(1)

In this equation, α is the specific volume, inline image, the 2D wind, inline image, the 2D eddy wind, inline image, the 2D eddy ageostrophic wind and ϕ′, the eddy geopotential heights. The subscript ‘3’ denotes three-dimensional vectors and the overbar a time mean. As Chang (2000) used the seasonal mean and our studied case was in August we used the average of the July, August and September, 2006.

The term on the left hand side of Equation (1) is the local tendency of K′. The first term on the right hand side represents the EKE flux convergence (KFC), the second the ageostrophic flux convergence (AFC) and the third term is the baroclinic conversion (BRC). The fourth and fifth terms are the Reynolds stresses and they are considered as barotropic conversion (BRT) terms. The sixth and seventh terms are the vertical flux convergence of energy. The term RES contains, among other things, mechanisms not explained by Equation (1), such as friction, diabatic effects and sub-grid flux, as well as errors introduced by numerical methods such as interpolation and finite differences.

The advantage of Equation (1) is that it lists the most important processes for the formation and dissipation of the disturbances, such as baroclinic instability, barotropic instability and downstream development. This equation, averaged over volume, results in the following equation (Chang, 2000):

  • display math(2)

where the subscript ‘v’ indicates the volume displacement velocity, the symbol ⟨ ⟩ represents the integral over the volume and the surface integral at the top (T) and the base (B).

The terms in Equation (2) have the same interpretation as those in Equation (1). The exception is the two terms representing the vertical flux convergence of energy, which in the latter equation were transformed into surface integrals. These can be interpreted as the vertical flux of energy through the lower boundary (B) to the top (T) of the atmosphere (sixth to ninth terms). The tenth term represents the energy flux due to the movement of the volume integration, while the last term is the change in the energy due to the variation of the mass in the volume. To apply Equation (2), we defined the cube (volume) centered on the vortex center with dimensions of 25° latitude × 30° longitude, because the cube must contain the two energy centers, one to the left and one to the right of the system axis (Chang, 2000). This cube was displaced step by step to follow the COL propagation (see black box in Figure 2) and always equatorward of the main polar jet stream. Vertical levels used were from the first isobaric level with real data to 100 hPa. Real data means that the isobaric level is above the topography height used in the reanalysis model. The surface pressure (ps) was the only variable used in surface level. However, in regions of high topography (over the Andes), the lowest level from which the integration was performed varied with the height of the topography used in the reanalysis. The tendency term was computed by two methods: the first one, a centered time difference between the previous and subsequent time steps and the other a summation of the all terms in Equation (2), except the RES term. The RES term was determined by the differences of these two tendencies.

3. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and methodology
  5. 3. Results
  6. 4. Conclusions
  7. Acknowledgements
  8. References

The COL that formed on 19 August 2006 at 1200 UTC over the Pacific Ocean at approximately 104°W, 30°S (Figure 1(a)) was studied because most of its life cycle occurred over the ocean. In this sense, we avoid other mechanisms associated with topography which could influence the formation, intensification or dissipation phases of this COL. This is important because the COL can interact with the Rossby topographic wave (Funatsu et al., 2004) and to induce an erroneous conclusion. In Figure 1(a), a large cloud band over the eastern South Pacific can be observed. Upstream of this band there is dry and moisture air side by side with cyclonic circulation characteristics (spiral format). Similar pattern was observed in the water vapour satellite channel by Bengtsson et al. (1982) for a COL over Northeast Brazil. Twenty-four hours later (Figure 1(b)), the vortex was sharply observed with dry and wet air in spiral form. On 21 August at 1200 UTC (Figure 1(c)), the vortex was completely cut off, as we can see from the dry air around the vortex. During the next 4 days (Figures 1(d)–(g)) the COL continued its slow eastward propagation. On 25 August (Figure 1(g)), it approached the Andes. On the 26 August, it was over the South American continent (Figure 1(h)); then the dry and wet air pattern changed and took on the characteristics of an extratropical cyclone in its development phase.

image

Figure 1. Satellite water vapour images at 1200 UTC (a) 19 August 2006, (b) 20 August, (c) 21 August, (d) 22 August, (e) 23 August, (f) 24 August, (g) 25 August and (h) 26 August.

Download figure to PowerPoint

In the 300 hPa geopotential field (Figure 2) for 19 August, a mid-latitude trough with high amplitude over the subtropical southeastern Pacific (Figure 2(a)) is seen. At this time, in this field the vortex was not closed yet, which is in agreement with the satellite image (Figure 1(a)). The vortex only appeared closed on 20 August at 0000 UTC (Figure not shown). The vertically averaged EKE (Figure 2) presents initially two EKE centers, one on the west side and the other on the east side of the trough axis. The EKE center on the west side increases faster than that on the east side, with a typical EKE behaviour associated with midlatitudes transient troughs (Orlanski and Sheldon, 1995). However, with the COL development, these two EKE centers merged and just one EKE maximum grew located just at southern sector of the COL. Another important point to highlight is that during 20 August, the EKE maximum associated with the COL was separated from the EKE maximum associated with the trough located at upstream, it began to pass at the south side of the COL, leaving the COL isolated from the main westerly flow (Figure 2(b)). The COL reached its maximum intensity during 20 and 21 August with values higher than 750 m2 s−2 (Figure 2(b) and (c), respectively). On 22 August at 1200 UTC, the kinetic energy maximum associated with COL was weaker (Figure 2(d)). This decay phase continued until 26 August at 1200 UTC when the geopotential field showed an open cyclonic circulation (Figure 2(h)). This evolution can be explained by interaction of an upper level cyclonic disturbance with a Rossby topographic wave ridge. This process was seen to happen in South America by Funatsu et al. (2004), in a study of Andes lee cyclogenesis associated with a COL.

image

Figure 2. Geopotential height isolines (300-hPa) (continuous lines) and vertically averaged EKE (m2 s−2) (values above 250 m2 s−2 are shaded) at 1200 UTC (a) 19 August 2006, (b) 20 August, (c) 21 August, (d) 22 August, (e) 23 August, (f) 24 August, (g) 25 August and (h) 26 August. The box represents the region where the EKE and the conversion terms were vertically integrated.

Download figure to PowerPoint

By examination of the vertical averaged spatial distribution of the terms in Equation (1) (Figures 3 and 4), we see positive values of the BRC term on the west side of the vortex and negative on the east side (Figure 3(a)) on 19 August at 1200 UTC. However, the positive values were higher than the negatives. After 20 August at 1200 UTC (Figure 3(b)–(f)), negative BRC conversions were observed around the vortex region. The magnitude and area coverage by the BRC term were higher in EKE center passing south of the COL than on the COL center.

image

Figure 3. Vertical-averaged kinetic energy (m2 s−2) (thick line) and BRC term (continuous (positive) and dashed thin (negative) lines, contour interval 2 × 102m2 s−2 day−1, zero omitted).

Download figure to PowerPoint

image

Figure 4. Vertical averaged kinetic energy (m2s−2) (thick line) and AFC term (continuous (positive) and dashed thin (negative) lines. Values plotted are 2, 7, 15 and 25 × 102 m2s−2.day−1). The vectors represent the ageostrophic flux.

Download figure to PowerPoint

Negative values of the BRT term can be observed in the area around the COL on 19 August at 1200 UTC (Figure not shown). After this time, negative and positive values were observed around the COL.

The AFC term (Figure 4) also shows positive and negative contributions in the COL region. Divergence of ageostrophic flux (negative values) was observed to the west of the trough and a convergence (positive values) region to the east on 19 August (Figure 4(a)). The ageostrophic fluxes (arrows in Figure 4(b)) indicated that the energy in the west side center was transported to the center on the east side. This AFC pattern continued through most of the time of analysis and it was typical of the energy cycle associated with mid-latitudes transient troughs (Orlanski and Katzfey, 1991; Orlanski and Sheldon, 1995; Orlanski and Gross, 2000; Decker and Martin, 2005). It is also important to note here how was the behaviour of the ageostrophic fluxes along of the southwestern and southeastern boundary of the box. On 19 and 20 October, the fluxes crossing the southwestern corner transporting EKE from the polar jet stream to the kinetic energy center on the west side, and along the southeastern boundary the fluxes was extracted EKE from the eastern energy center (Figure 4(a) and (b)) to out of the box. On 21 October, the ageostrophic fluxes on the southwestern boundary continued transporting EKE (Figure 4(c) and (d)) from the jet stream, but the ageostrophic fluxes on the southeastern boundary behaved differently. On this day, the ageostrophic fluxes transported EKE from the polar jet stream to the COL (Figure 4(c) and (d)). Interesting to note that the polar jet that allowed the cutting of the COL, now it provided energy to the COL. However, the EKE from the polar jet was mostly generated by baroclinic conversion as we can be seen in Figure 3. From day 24 onwards, the fluxes weakened in both corners of the box, but the fluxes crossing the southeast corner was extracting energy from the COL, while the fluxes in the southwest corner were practically nonexistent (Figure 4(f)–(h)). These results show that the kinetic energy to maintain the COL for more time is coming from dispersive radiative fluxes, i.e. from the polar jet stream.

Calculating the EKE and the conversion terms integrated through the volume represented by the black box in Figure 2, one can see an intensification phase from 19 August at 1200 UTC until 20 August at 1800 UTC, followed by a continuous decay phase (Figure 5). During the EKE intensification, the AFC and KFC were positive. As the KFC term only contribute to displace the EKE center because it is associated with advective flux (Chang, 2000), it was not important to intensify the COL. However, the AFC term was positive since 19 August at 0600 UTC and kept a strong positive tendency during the EKE intensification until 20 August at 0000 UTC. The maximum intensity of AFC occurred on 21 August at 0000 UTC when a strong reduction of the fluxes dominated until 22 August at 1200 UTC. After that the AFC became positive and small, until 2500 UTC at 0600 UTC when became negative. It is important to highlight that the formation of the COL occurred during the intensification phase and during this phase the intensity of the AFC term had a positive tendency. The BRC term was positive until 19 August at 1800 UTC, and after 20 August, this term was negative until 24 August at 1200 UTC, when it became positive again. However, the BRT was negative during almost of the time. This result is different from that found by Mishra et al. (2001) for COLs over Northeast of Brazil, which the BRT term was important during the formation COL phase. This disagreement is because the COL studied by Mishra et al. (2001) is from tropical region where the geostrophy theory is not valid, the temperature gradient is small and the barotropic instability of zonal flow is considered as the main dynamical mechanism for the origin of most synoptic disturbances formed in the tropics (Kuo, 1973).

image

Figure 5. Temporal evolution of volume-averaged EKE (solid line), KFC (open circles), BRC (closed circles), AFC (open squares), BRT (closed squares) and RES (dotted-dashed line) for the period from 19 at 0000 UTC to 26 August 2006 at 1200 UTC. Units are m2 s−2 day−1, except for the EKE unit which is m2 s−2. The symblol ‘x’ represent the formation time, and ‘ + ’ the dissipation time. The dotted lines intersect the date at 0000 UTC.

Download figure to PowerPoint

Thus, the BRC and AFC terms were important mechanisms for intensification and formation of the COLs. But when we analyse the spatial distribution (Figure 3), we see that the BRC conversion was important in intensifying the western energy center, whereas the eastern energy center was intensified via AFC coming from the western kinetic energy center. However the western EKE center was also intensified by kinetic energy coming from the polar jet stream.

It is important to highlight that the RES term was negative during the formation and intensification phases. Thus other mechanisms, such as friction can have some contribution during these phases.

While the COL maintained its maximum intensity on 20 and 21 August, AFC had the highest positive values and it remained positive until 25 August, so this suggests that AFC was the main mechanism for the maintenance of the COL, and the energy imported to into the western box was generated by baroclinic conversion in the jet stream area located southward of the COL.

4. Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and methodology
  5. 3. Results
  6. 4. Conclusions
  7. Acknowledgements
  8. References

The analysis for the COL case formed on 19 August 2006 at 1200 UTC over the Pacific Ocean at approximately 104°W, 30°S shows that the formation of this COL was associated with a mid-latitude trough with high amplitude over the subtropical southeastern Pacific. When this trough was closing, baroclinic conversion on west side, mainly during the first 18 h and AFC on east side of the COL coming from the jet stream located to southward of the COL were the main mechanisms.

After the period of maximum intensity, the baroclinic mechanism converted EKE to eddy potential energy (negative contribution), the barotropic term was negative and the AFC had a positive contribution. Since the RES was negative during this phase, we can conclude that the AFC term was an important mechanism in extending the lifetime of the active COL for a few more days. However, in decaying phases the BRC term became positive and AFC and BRT remained negative.

On the basis of these results, we can conclude that the lifecycle of the COL studied here presented a similar evolution to the 300-hPa cyclonic vortex formed over the South Atlantic Ocean studied by Chang (2000). In this case, the baroclinic conversion and AFC had an equivalent importance during the growing phase. In the decaying phase the decrease in energy was dominated by ageostrophic flux divergence. The difference with our case was that the baroclinic conversion was positive during throughout of the Chang's COL life.

The energetic presented in our case was different from the downstream baroclinic development mechanism proposed by Orlanski and Sheldon (1995) and observed during the development of a North Pacific cyclone by Danielson et al. (2006). In the downstream baroclinic development case, the formation of a new upper level disturbance was associated with AFC and this new disturbance induced a low-level circulation that could grow, stimulating baroclinic conversion. In our case, the COL formation was not only associated AFC but also with baroclinic conversion and none low-level disturbance was formed.

As we only presented results of one case that explain how a mid-latitude trough is segregated and the mechanism that maintain the COL for more time, we cannot generalize that the mechanism presented in this study is representative of more COLs formed in subtropical latitudes of South Pacific. A study with more case is necessary to confirm our results.

Acknowledgements

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and methodology
  5. 3. Results
  6. 4. Conclusions
  7. Acknowledgements
  8. References

This work was supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, Projects 490401/2006-6 and 302188/2007-0).

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Data and methodology
  5. 3. Results
  6. 4. Conclusions
  7. Acknowledgements
  8. References