• Agustí-Panareda, A., , C. D. Thorncroft, , G. C. Craig, , and S. L. Gray, 2004: The extratropical transition of Hurricane Irene (1999): A potential vorticity perspective. Quart. J. Roy. Meteor. Soc., 130 , 10471074.

    • Search Google Scholar
    • Export Citation
  • Agustí-Panareda, A., , S. L. Gray, , G. C. Craig, , and C. D. Thorncroft, 2005: The extratropical transition of tropical cyclone Lili (1996) and its crucial contribution to a moderate extratropical development. Mon. Wea. Rev., 133 , 15621573.

    • Search Google Scholar
    • Export Citation
  • Ahmadi-Givi, F., , G. C. Craig, , and R. S. Plant, 2004: The dynamics of a midlatitude cyclone with very strong latent-heat release. Quart. J. Roy. Meteor. Soc., 130 , 295323.

    • Search Google Scholar
    • Export Citation
  • Atallah, E. H., , and L. F. Bosart, 2003: The extratropical transition and precipitation distribution of Hurricane Floyd (1999). Mon. Wea. Rev., 131 , 10631081.

    • Search Google Scholar
    • Export Citation
  • Beare, R. J., , A. J. Thorpe, , and A. A. White, 2003: The predictability of extratropical cyclones: Nonlinear sensitivity to localized potential-vorticity perturbations. Quart. J. Roy. Meteor. Soc., 129 , 219237.

    • Search Google Scholar
    • Export Citation
  • Carroll, E. B., 1997: Use of dynamical concepts in weather forecasting. Meteor. Appl., 4 , 345352.

  • Chang, E. K. M., , S. Lee, , and K. Swanson, 2002: Storm track dynamics. J. Climate, 15 , 21632183.

  • Charney, J. G., 1955: The use of primitive equations of motion in numerical prediction. Tellus, 7 , 2226.

  • Clough, S. A., , C. S. A. Davitt, , and A. J. Thorpe, 1996: Attribution concepts applied to the omega equation. Quart. J. Roy. Meteor. Soc., 122 , 19431962.

    • Search Google Scholar
    • Export Citation
  • Cullen, M. J. P., 1993: The unified forecast/climate model. Meteor. Mag., 122 , 8194.

  • Darr, J. K., 2002: Quantitative measurements of extratropical transition in the Atlantic basin. Preprints, 19th Conf. on Weather Analysis and Forecasting, San Antonio, TX, Amer. Meteor. Soc., 186–189.

  • Davis, C. A., , and K. A. Emanuel, 1991: Potential vorticity diagnostics of cyclogenesis. Mon. Wea. Rev., 119 , 19291953.

  • Deveson, A. C. L., , K. A. Browning, , and T. D. Hewson, 2002: A classification of FASTEX cyclones using a height-attributable quasi-geostrophic vertical-motion diagnostic. Quart. J. Roy. Meteor. Soc., 128 , 93116.

    • Search Google Scholar
    • Export Citation
  • DiMego, G. J., , and L. F. Bosart, 1982: The transformation of tropical storm Agnes into an extratropical cyclone. Part I: The observed fields and vertical motion computations. Mon. Wea. Rev., 110 , 385411.

    • Search Google Scholar
    • Export Citation
  • Evans, J. L., , and B. E. Prater-Mayes, 2004: Factors affecting the posttransition intensification of Hurricane Irene (1999). Mon. Wea. Rev., 132 , 13551368.

    • Search Google Scholar
    • Export Citation
  • Gregory, D., , and P. R. Rowntree, 1990: A mass flux convection scheme with representation of cloud ensemble characteristics and stability-dependent closure. Mon. Wea. Rev., 118 , 14831506.

    • Search Google Scholar
    • Export Citation
  • Grotjahn, R., 2005: On type B cyclogenesis in a quasi-geostrophic model. Quart. J. Roy. Meteor. Soc., 131 , 109124.

  • Hakim, G. J., , D. Keyser, , and L. F. Bosart, 1996: The Ohio Valley wave-merger cyclogenesis event of 25–26 January 1978. Part II: Diagnosis using quasigeostrophic potential vorticity inversion. Mon. Wea. Rev., 124 , 21762205.

    • Search Google Scholar
    • Export Citation
  • Harr, P. A., , and R. L. Elsberry, 2000: Extratropical transition of tropical cyclones over the western North Pacific. Part I: Evolution of structural characteristics during the transition process. Mon. Wea. Rev., 128 , 26132633.

    • Search Google Scholar
    • Export Citation
  • Heming, J. T., , J. C. L. Chan, , and A. M. Radford, 1995: A new scheme for the initialisation of tropical cyclones in the UK Meteorological Office global model. Meteor. Appl., 2 , 171184.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., , and F. P. Bretherton, 1972: Atmospheric frontogenesis models: Mathematical formulation and solution. J. Atmos. Sci., 29 , 1137.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., , and M. A. Pedder, 1980: The diagnosis of middle-latitude synoptic development. Quart. J. Roy. Meteor. Soc., 106 , 707719.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., , I. Draghici, , and H. C. Davis, 1978: A new look at the ω equation. Quart. J. Roy. Meteor. Soc., 104 , 3138.

  • 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.

    • Search Google Scholar
    • Export Citation
  • Jones, S. C., and Coauthors, 2003: The extratropical transition of tropical cyclones: Forecast challanges, current understanding, and future directions. Wea. Forecasting, 18 , 10521092.

    • Search Google Scholar
    • Export Citation
  • Klein, P. M., , P. A. Harr, , and R. L. Elsberry, 2000: Extratropical transition of western North Pacific tropical cyclones: An overview and conceptual model of the transformation stage. Wea. Forecasting, 15 , 373395.

    • Search Google Scholar
    • Export Citation
  • Klein, P. M., , P. A. Harr, , and R. L. Elsberry, 2002: Extratropical transition of western North Pacific tropical cyclones: Midlatitude and tropical cyclone contributions to reintensification. Mon. Wea. Rev., 130 , 22402259.

    • Search Google Scholar
    • Export Citation
  • Lawrence, M. B., , L. A. Avila, , J. L. Beven, , J. L. Franklin, , J. L. Guiney, , and R. J. Pasch, 2001: Atlantic hurricane season of 1999. Mon. Wea. Rev., 129 , 30573084.

    • Search Google Scholar
    • Export Citation
  • Matano, H., , and M. Sekioka, 1971: Some aspects of extratropical transformation of a tropical cyclone. J. Meteor. Soc. Japan, 49 , 736743.

    • Search Google Scholar
    • Export Citation
  • McTaggart-Cowan, R., , J. R. Gyakum, , and M. K. Yau, 2001: Sensitivity testing of extratropical transitions using potential vorticity inversions to modify initial conditions: Hurricane Earl case study. Mon. Wea. Rev., 129 , 16171636.

    • Search Google Scholar
    • Export Citation
  • Petterssen, S., , and S. J. Smebye, 1971: On the development of extratropical cyclones. Quart. J. Roy. Meteor. Soc., 97 , 457482.

  • Simpson, R. H., 1974: The hurricane disaster potential scale. Weatherwise, 27 , 169186.

  • View in gallery

    Geostationary Operational Environmental Satellite (GOES)-8 (GOES-East) water vapor image. (Courtesy of the National Climatic Data Center.) Here, L denotes the center of the upstream extratropical low pressure system, wf1 is the warm front associated with the upstream baroclinic wave, T denotes the location of a large-scale upper-level trough, and D is the stratospheric dry intrusion associated with an upper-level potential vorticity anomaly. A belt of low pressure extending from L toward the north is marked with a trough line (dashed); wf2 is the warm front that developed as Gert approached the colder extratropical environment, and O is the outflow of Tropical Cyclone Gert.

  • View in gallery

    ECMWF analysis fields at 0000 UTC 23 Sep 1999: (a) humidity mixing ratio (g kg−1) at 850 hPa [gray shading: below 7.5 in white, 7.5–10.0 in light gray, 10.0–15.0 in gray, and larger than 15.0 in dark gray; also shown are PV contours equal to 2 PVU at 850 hPa (thick, solid line) and geopotential height contours at 1000 hPa (thin solid lines, 40-m interval)] and (b) isentropic PV on a 330-K surface [gray shading: black is PV from −1 to 0.2 PVU, dark gray is 2–4 PVU, and light gray is 4–14 PVU; also shown are potential temperature contours at 950 hPa (solid lines, 2-K interval)]. Symbols are as in Fig. 1.

  • View in gallery

    As in Fig. 2, but at 1200 UTC 23 Sep 1999.

  • View in gallery

    As in Fig. 2, but at 0000 UTC 24 Sep 1999.

  • View in gallery

    As in Fig. 2, but at 1200 UTC 24 Sep 1999.

  • View in gallery

    North–south cross sections of ECMWF analysis fields across the center of the tropical cyclone PV tower during the extratropical transition of Gert at (a) 0000 and (b) 1200 UTC 23 Sep 1999. Gray shading represents PV with values from −1 to 0 (black), 2–4 (dark gray), and 4–14 PVU (light gray). Thick solid lines are potential vorticity contours (1-PVU interval), thin solid lines are equivalent potential temperature contours (5-K interval), and dashed lines are potential temperature contours (5-K interval).

  • View in gallery

    North–south cross sections of ECMWF analysis fields across the center of the tropical cyclone PV tower during the extratropical transition of Gert at (a) 0000 and (b) 1200 UTC 24 Sep 1999. Same contours intervals as in Fig. 6.

  • View in gallery

    ECMWF analysis fields: humidity mixing ratio (g kg−1) at 500 hPa (gray shading: white is below 1.0, light gray is 1.0–2.5, gray is 2.5–5.0, and dark gray is larger than 5.0). Also shown are PV contours equal to 2 PVU at 500 hPa (thick, solid line) and potential temperature contours at 500 hPa (thin, solid line; 2-K interval). Symbols are as in Fig. 1.

  • View in gallery

    As in Fig. 8, but for 24 Sep 1999.

  • View in gallery

    Horizontal section of PV at 250 hPa with gray shading (black: from −1 to 0.3 PVU, dark gray: 2–4 PVU, and light gray: 4–14 PVU) covering the limited-area model (LAM) domain. The PV inversion domain is shown with the black box centered on Tropical Cyclone Gert (“G” depicts the location of its PV tower). The negative PV anomaly associated with the upper-level outflow of Gert is marked as O and that associated with the outflow of the frontal ascent across the warm front wf1 is marked as Owf1.

  • View in gallery

    (a) Tracks followed by Gert and the upper-level forced low upstream. Overlaid contours are mean SST for September 1999 from the Reynolds dataset (courtesy of Climate Diagnostics Center). (b) Minimum MSLP at the center of Gert during its extratropical transition and the upper-level forced low pressure system upstream. In both (a) and (b), ECMWF analysis, Met Office analysis, and the forecast initialized from ECMWF analysis fields have been plotted every 12 h. The control forecast initialized from the balanced fields (see section 3a) has been plotted every 12 h in (a) and every hour in (b). Both forecasts were started at 0000 UTC 23 Sep 1999.

  • View in gallery

    (a) Tracks of the center of the transitioning Gert and the upstream low for the control forecast and the forecast without Gert in the initial conditions every 1 h, and (b) time series of hourly minimum MSLP at the center of transitioning Gert and the upstream low for the CNTRL and NOTC.

  • View in gallery

    Horizontal sections of humidity mixing ratio (g kg−1) at 850 hPa [gray shading: white is below 7.5, light gray is 7.5–10.0, gray is 10.0–15.0, and dark gray is larger than 15.0; also shown are PV contours equal to 2 PVU at 850 hPa (thick, solid line), and geopotential height contours at 1000 hPa (thin, solid line; 40-m interval)] from (a) CNTRL and (b) NOTC. Symbols are as in Fig. 1.

  • View in gallery

    As in Fig. 13, but for 0000 UTC 24 Sep 1999.

  • View in gallery

    As in Fig. 13, but for 1200 UTC 24 Sep 1999.

  • View in gallery

    Horizontal cross sections of isentropic PV on a 330-K surface with values from −1 to 0.2 PVU (black), from 2 to 4 PVU (dark gray), and from 4 to 14 PVU (light gray) [also shown are potential temperature contours (thin, solid line; 2-K interval) at 950 hPa] for (a) CNTRL and (b) NOTC. Symbols are as in Fig. 1.

  • View in gallery

    As in Fig. 16, but for 0000 UTC 24 Sep 1999.

  • View in gallery

    As in Fig. 16, but for 1200 UTC 24 Sep 1999.

  • View in gallery

    Horizontal sections of humidity mixing ratio (g kg−1) at 500 hPa [gray shading: white is below 1.0, light gray is 1.0–2.5, gray is 2.5–5.0, and dark gray is larger than 5.0; also shown are PV contours equal to 2 PVU at 500 hPa (thick, solid line) and potential temperature contours at 500 hPa (thin, solid line; 2-K interval)] from (a) CNTRL and (b) NOTC. Symbols are as in Fig. 1.

  • View in gallery

    As in Fig. 19, but for 0000 UTC 24 Sep 1999.

  • View in gallery

    As in Fig. 19, but for 1200 UTC 24 Sep 1999.

  • View in gallery

    Horizontal cross sections of isentropic PV on a 345-K surface with values from −1 to 0.3 (black), from 2 to 4 (dark gray), and from 4 to 14 (light gray) PVU and potential temperature contours (thin, solid line; 2-K interval) at 950 hPa from (a) CNTRL and (b) NOTC. Symbols are as in Fig. 1.

  • View in gallery

    As in Fig. 22, but for 0000 UTC 24 Sep 1999.

  • View in gallery

    Quasigeostrophic vertical velocity (ωQG) at 700 hPa forced from lower levels (1050–750 hPa) with negative values as thick dashed contours and positive values as thick solid contours (0.008 m s−1 interval) and geopotential height at 1000 hPa (thin contours, 40-m interval) from (a) CNTRL and (b) NOTC. Note that ωQG is positive for upward motion and negative for downward motion.

  • View in gallery

    As in Fig. 24, but for 0000 UTC 24 Sep 1999.

  • View in gallery

    As in Fig. 24, but for 1200 UTC 24 Sep 1999.

  • View in gallery

    Quasigeostrophic vertical velocity (ωQG) at 700 hPa forced from upper levels (650–50 hPa) with negative values as thick dashed contours and positive values as thick solid contours (0.008 m s−1 interval) and geopotential height at 300 hPa (thin contours, 80-m interval) from (a) CNTRL and (b) NOTC. Note that ωQG is positive for upward motion and negative for downward motion.

  • View in gallery

    As in Fig. 27, but for 0000 UTC 24 Sep 1999.

  • View in gallery

    As in Fig. 27, but for 1200 UTC 24 Sep 1999.

  • View in gallery

    (a) Maximum positive quasigeostrophic vertical velocity (ωQG, m s−1) at 700 hPa forced from lower (PL) and upper (PU) levels for the CNTRL and NOTC forecasts, and (b) ratio of the (ωQG) dipole strength forced from lower and upper levels (U/L ratio). The dipole strength at low levels (DL) is calculated by summing the magnitudes of the maximum ascent [see (a)] and maximum descent (not shown) at 700 hPa forced from lower levels. The strength of the upper-level forced dipole (DU) at 700 hPa is calculated in a similar fashion. The U/L ratio is DU/DL. Note that positive vertical velocity is upward and negative vertical velocity is downward. Low-level forcing is from 1000 to 750 hPa, and upper-level forcing is from 650 to 100 hPa. See section 3c for more details.

  • View in gallery

    This schematic shows the interactions involved during an extratropical transition classified as a “compound system” (Matano and Sekioka 1971) based on concepts from the QG theory. (a) Interaction between a vortex associated with the remnants of a TC PV tower (depicted by arrows with a hurricane symbol at the center) and a zonal baroclinic zone (dotted line) during the transformation stage of the TC. The result is cold/warm advection to the west/east of the vortex and an associated QG vertical velocity (ωQG) dipole (solid/dashed closed contours with positive/negative signs denoting ascent/descent, respectively). The dipole is forced from both upper and lower levels. (b) Upper- and lower-level forcing in the development of an extratropical cyclone associated with an upper-level positive PV anomaly collocated with an upper-level trough T (see dark gray shading) and surface thermal anomalies depicted by the cold/warm fronts. An ascending/descending vertical motion dipole in gray shading and hatch shading, respectively, is centered on the trough axis T. The ascending region is the area of favorable development for the surface low pressure center L. At low levels the ωQG dipole is centered around L (thick solid/dashed closed contours with positive/negative signs denoting ascent/descent, respectively). All of the ωQG dipoles are shown at the steering level of the extratropical cyclone. The transformed TC can interact with (c) the warm front or (d) the cold front of a preexisting baroclinic cyclone. See text for a more detailed explanation.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 25 25 7
PDF Downloads 8 8 0

The Contribution of Ex–Tropical Cyclone Gert (1999) toward the Weakening of a Midlatitude Cyclogenesis Event

View More View Less
  • 1 Department of Meteorology, University of Reading, Reading, United Kingdom
© Get Permissions
Full access

Abstract

Tropical Cyclone Gert (1999) experienced an extratropical transition while it merged with an extratropical cyclone upstream. The upstream extratropical cyclone had started to intensify before it merged with the transitioning tropical cyclone, and it continued intensifying afterward (12 hPa in 12 h, according to the Met Office analysis). The question addressed in this paper is the following: what was the impact of the transitioning tropical cyclone on this intensification of the upstream extratropical cyclone? Until now, in the literature, tropical cyclones that experience extratropical transition have been found to have either no impact or a positive impact on the development of extratropical cyclogenesis events. The positive impact involves either a triggering of the development of the extratropical cyclone or simply a contribution to its deepening. However, the case studied here proves to have a negative impact on the developing extratropical cyclone upstream by diminishing its intensification. Forecasts are performed with and without the tropical cyclone in the initial conditions. They show that when Gert is not present in the initial conditions, the peak intensity of the cyclone upstream occurs 9 h earlier and it is 10 hPa deeper than when Gert is present. Thus, Gert acts to weaken the development by contributing to the filling of the extratropical surface low upstream. Quasigeostropic (QG) diagnostics show that the negative impact on the extratropical development is linked to the fact that the transitioning tropical cyclone interacts with a warm front inducing a negative QG vertical velocity over the developing extratropical low upstream. This interpretation is consistent with other contrasting cases in which the transitioning tropical cyclone interacts with a cold front and induces a positive QG vertical velocity over the developing low upstream, thus enhancing its development. The results are also in agreement with idealized experiments in the literature that are aimed at studying the predictability of extratropical storms. These idealized experiments yielded similar results using synoptic-scale and mesoscale vortices as perturbations on warm and cold fronts.

Corresponding author address: A. Agustí-Panareda, European Centre for Medium-Range Weather Forecasts, Shinfield Park, Reading RG2 9AX, United Kingdom. Email: anna.agusti-panareda@ecmwf.int

Abstract

Tropical Cyclone Gert (1999) experienced an extratropical transition while it merged with an extratropical cyclone upstream. The upstream extratropical cyclone had started to intensify before it merged with the transitioning tropical cyclone, and it continued intensifying afterward (12 hPa in 12 h, according to the Met Office analysis). The question addressed in this paper is the following: what was the impact of the transitioning tropical cyclone on this intensification of the upstream extratropical cyclone? Until now, in the literature, tropical cyclones that experience extratropical transition have been found to have either no impact or a positive impact on the development of extratropical cyclogenesis events. The positive impact involves either a triggering of the development of the extratropical cyclone or simply a contribution to its deepening. However, the case studied here proves to have a negative impact on the developing extratropical cyclone upstream by diminishing its intensification. Forecasts are performed with and without the tropical cyclone in the initial conditions. They show that when Gert is not present in the initial conditions, the peak intensity of the cyclone upstream occurs 9 h earlier and it is 10 hPa deeper than when Gert is present. Thus, Gert acts to weaken the development by contributing to the filling of the extratropical surface low upstream. Quasigeostropic (QG) diagnostics show that the negative impact on the extratropical development is linked to the fact that the transitioning tropical cyclone interacts with a warm front inducing a negative QG vertical velocity over the developing extratropical low upstream. This interpretation is consistent with other contrasting cases in which the transitioning tropical cyclone interacts with a cold front and induces a positive QG vertical velocity over the developing low upstream, thus enhancing its development. The results are also in agreement with idealized experiments in the literature that are aimed at studying the predictability of extratropical storms. These idealized experiments yielded similar results using synoptic-scale and mesoscale vortices as perturbations on warm and cold fronts.

Corresponding author address: A. Agustí-Panareda, European Centre for Medium-Range Weather Forecasts, Shinfield Park, Reading RG2 9AX, United Kingdom. Email: anna.agusti-panareda@ecmwf.int

1. Introduction

Tropical cyclones that move into midlatitudes experience an extratropical transition (Jones et al. 2003). During this transition the tropical cyclone interacts with a baroclinic zone and experiences a gradual transformation resulting in the development of a frontal structure (Klein et al. 2000). This transformation can occur as the tropical cyclone or its remnants interact with a baroclinic zone without a preexisting extratropical cyclone or with a frontal zone from a preexisting cyclone. Matano and Sekioka (1971) classified the first case as a complex system and the second as a compound system. The resulting extratropical cyclone can intensify if the remnants of the tropical cyclone are steered to the east of an upstream upper-level trough (Harr and Elsberry 2000), as in the Petterssen and Smebye (1971) type-B cyclogenesis (Klein et al. 2002). Type-B cyclones form when an upper-level trough moves over a region of warm advection (Petterssen and Smebye 1971). This conceptual model was developed as part of a synoptic classification scheme by Petterssen and Smebye (1971), and it has been widely used in case studies of specific events, climatologies, and idealized theoretical models (e.g., Deveson et al. 2002; Grotjahn 2005). A type-B cyclogenesis in the context of an extratropical transition (ET) implies that the region of warm advection induced by the tropical cyclone remnants is in phase with the extratropical forcing induced by the upper-level trough, such that a favorable interaction between the two may occur. Klein et al. (2002) showed that this interaction could be increased or decreased by displacing the initial vortex associated with the remnants of the tropical cyclone. In all the cases the contribution of the tropical cyclone to the development of an extratropical cyclone was either zero or positive. In this paper we present a case study where the extratropical transition of Tropical Cyclone Gert (1999) had a negative contribution to the development of an upstream extratropical cyclone, because the remnants of Tropical Cyclone Gert (1999) contributed to a reduction in the deepening of the extratropical cyclone.1 Thus, Gert (1999) provides an interesting case that contrasts with other cases in the literature in which the transitioning tropical cyclone contributed to an enhancement in the deepening of the resulting extratropical cyclone (e.g., Klein et al. 2002; Agustí-Panareda et al. 2004). The tropical cyclone can even have a crucial role by triggering the development of the extratropical cyclone [e.g., Tropical Cyclone Ginger (1997) in Klein et al. (2002) and Tropical Cyclone Lili (1996) in Agustí-Panareda et al. (2005)]. The crucial role of the tropical cyclone was linked to the unfavorable or neutral midlatitude contributions in Klein et al. (2002) and to the baroclinic life cycle associated with environmental anticyclonic shear in Agustí-Panareda et al. (2005).

The aim of this paper is to investigate the impact of the extratropical transition of Tropical Cyclone Gert in diminishing the intensification of an extratropical low upstream by performing numerical experiments. There are many processes involved in an ET [see Jones et al. (2003) for an extensive review], and an exhaustive investigation of all these processes is not in the scope of this paper. The focus of herein is on the synoptic development triggered by the vortex associated with the remnants of the tropical cyclone, in particular the development linked to synoptic-scale advection. The numerical experiments are designed to compare the simulation of the case study of Gert during its extratropical transition, with and without the tropical cyclone in the initial conditions, one day before Gert is classified as extratropical by the National Hurricane Center (NHC). They follow the approach of Agustí-Panareda et al. (2004, 2005), which was based on the method of Davis and Emanuel (1991). The potential vorticity (PV) tower associated with the tropical cyclone is removed from the initial conditions of the forecast. The impact of the tropical cyclone on the extratropical development at synoptic scales is assessed using quasigeostrophic (QG) diagnostics developed by Deveson et al. (2002) on the results of the numerical experiments. The diagnostic is based on the calculation of the quasigeostropic vertical motion at the steering level attributed to the forcing from the lower (1050–750 hPa) and upper (650–50 hPa) levels, separately. The results are found to be consistent with the work on predictability in the extratropics by Beare et al. (2003). They used PV perturbations associated with mesoscale and synoptic-scale vortices, which could be linked to the remnants of a tropical cyclone, to investigate their impact on the growth of an extratropical cyclone. Their results show that the positive PV perturbations located at the steering level of the extratropical cyclone in the region of the warm front, like the remnants of Gert, produce an optimum filling of the cyclone (Beare et al. 2003), and the PV perturbations located in the region of the cold front, like the remnants of Irene (1999; see Agustí-Panareda et al. 2004), produce an optimum deepening. This emphasizes how important extratropical transitions can be in the context of predictability and forecasting in the extratropics.

A description of the extratropical transition event is presented in section 2 using analysis data and satellite images. The method and results of the numerical experiments are described in sections 3 and 4, respectively. In section 5 the results are summarized. A discussion of their implications on predictability and case-to-case variability of extratropical transitions is included.

2. Synoptic overview

Hurricane Gert (1999) originated from a tropical wave over the Atlantic Ocean off the coast of Africa on 10 September 1999. It became a hurricane on 13 September as it moved across the central North Atlantic, where it finally reached category-4 (Simpson 1974) strength [see Lawrence et al. (2001) for more details on its tropical development]. At 0000 UTC 23 September, Gert had weakened to a tropical storm with peak winds near 30 m s−1. Synoptic features and analyses are shown in Figs. 1 and 2. The tropical storm had moved over the warm sector of a developing baroclinic wave with a low pressure center “L” upstream. This developing baroclinic wave was associated with an upper-level positive PV anomaly and stratospheric dry intrusion “D” upstream. The upper-level PV anomaly is defined by the 2-PVU contour or dynamic tropopause (Hoskins et al. 1985). The low pressure area associated with the developing baroclinic wave extended north of L along a low pressure belt. To the north of this low pressure belt there was a warm front “wf1”. An upper-level negative PV anomaly, defined with respect to a background PV obtained by area averaging, was located around 55°N, 60°W in the region of the extensive cloud associated with wf1. This negative PV anomaly was part of the outflow associated with the large-scale ascent within the cloud of the warm front wf1, as indicated by back trajectories (not shown). The warm sector of the extratropical cyclone upstream L was delimited by warm front wf1, a broad baroclinic zone, and a cold front off the eastern coast of North America (Figs. 1a and 2b). The broad baroclinic zone was located to the northeast of the upstream extratropical cyclone and to the north of Gert. At the center of Gert there was a PV tower associated with deep convection extending from approximately 900 to 300 hPa (see Fig. 6a). A moist anomaly in Fig. 2a was collocated with the PV tower at lower levels and there was constant equivalent potential temperature from the surface throughout the depth of the PV tower (Fig. 6a). This is consistent with deep convective mixing in the tropical cyclone core. However, the first signs of asymmetries developing in the vicinity of the tropical cyclone core began to appear as shown by a warm/dry anomaly at 500 hPa to the southwest of Gert’s PV tower (Fig. 8a), in association with descending motion (not shown).

Twelve hours later, at 1200 UTC 23 September, the NHC classified Gert as extratropical. The evolution can be seen in the 4 panels of Fig. 1 and also in the set Figs. 2, 3, 4, 5 the set Figs. 6, 7 and the set Figs. 8, 9. The storm had moved over much cooler sea surface temperatures of around 8°C (to be seen in Fig. 11a). The PV tower became broader and the high equivalent potential temperature, collocated with the PV tower, became decoupled from the surface as Gert moved over the baroclinic zone (Figs. 6b and 3a,b).

The circulation associated with the transitioning tropical cyclone had began to interact with the broad baroclinic zone, advecting colder and drier air toward its northwestern sector within a deep layer expanding from the surface to 500 hPa in Fig. 8b and also Fig. 3a. Figures 3a,b show the tongue of advected drier/colder air at low levels to the northeast of the developing upstream low pressure center L, as well as warmer/moister air to the northeast of Gert. This resulted in the creation of a new warm front “wf2,” with an outflow cloud in the region of warm advection (see Figs. 1b and 3b). Meanwhile, the low pressure center of the upstream baroclinic wave L continued to deepen and increase in size (Fig. 3a). This is consistent with the creation of more low-valued PV in the region of the warm front wf1 and an increase in the gradient of potential temperature at the surface (Fig. 3b).

By 0000 UTC 24 September, the transformed tropical cyclone (ex-Gert) started merging with the upstream frontal low pressure system L (Figs. 1c and 4a,b). The PV anomaly and low pressure center associated with ex-Gert became part of the warm front wf2 (Figs. 4 and 7). After the merging of the two cyclones, the resulting cyclone had the following two warm fronts associated with it: wf1 and wf2 (Figs. 1c and 4b). The upper-level PV anomaly and dry intrusion associated with the upstream baroclinic cyclone started wrapping up cyclonically around the center of the resulting low pressure system (see D in Figs. 1c and 4b). This intrusion of dry air reached the center of the developing cyclone at 500 hPa (Fig. 9). The low pressure system deepened about 12 hPa in 12 h, reaching peak intensity (963 hPa) at around 1200 UTC 24 September (Figs. 1d and 5a,b).

3. Methodology

The aim of the numerical experiments is to assess the role of the transitioning tropical cyclone vortex in the intensification of an extratropical cyclone upstream. Forecasts starting from initial conditions with and without the tropical cyclone are compared. The forecasts are initialized at 0000 UTC 23 September 1999. At this time, Gert had weakened to tropical storm force, but it still had a symmetric warm-core structure. When removing the tropical cyclone from the initial conditions, it is always a concern that this might have interacted with the extratropical environment previously, and therefore the experiment does not guarantee an account of all of the possible impacts of the tropical cyclone on the extratropical environment. There is a trade-off between removing the cyclone before it can have any impact on the extratropical environment and initializing the forecast close to the onset of the extratropical transition in order to get an accurate forecast of the extratropical transition event (Agustí-Panareda et al. 2004, 2005). In this case, experiments performed by removing Gert 2 days earlier from the initial conditions (i.e., before it recurved poleward) were not a viable solution, because the forecast tracks of the extratropical cyclone upstream and Gert diverged, and as a result the two cyclones did not merge. Thus, any impact that could have occurred before 23 October, when Gert was a tropical cyclone, cannot be accounted for in this experiment. One of the possible impacts of Gert is the enhancement of the upper-level ridge adjacent to the upper-level trough by the tropical cyclone outflow. The upper-level ridge and trough are shown as positive and negative PV anomalies in Fig. 10. The origin of the negative PV anomaly in the region of the upper-level ridge has been investigated using back trajectories in order to assess the possibility of any crucial interaction between Gert’s outflow and the midlatitude environment 1 day before the initialization of the experiment. The back trajectories have been computed using 6-hourly European Centre for Medium-Range Weather Forecasts (ECMWF) analyses from 0000 UTC 23 September to 0000 UTC 22 September (not shown). These confirm that the air within the negative PV anomaly on the upper-level ridge does not originate from the region of the tropical cyclone. Instead, it originates from levels between 710 and 500 hPa over the region of the upstream development (see section 2). Thus, the impact of the transitioning tropical cyclone on its immediate environment upstream can still be assessed by the experiments presented in this paper, bearing in mind that any other possible impact before 0000 UTC 23 September will not be included.

The method used to remove the tropical cyclone from the initial conditions is described in 3a. The forecasting model used for the numerical experiments is described in 3b. The effect of Gert on the development of the upstream low pressure system is analyzed using quasigeostrophic diagnostics on the numerical experiments. The quasigeostrophic diagnostics are described in 3c.

a. Removing the tropical cyclone

The cyclonic warm-cored vortex associated with the tropical cyclone is removed from the initial conditions by applying PV surgery, as in Agustí-Panareda et al. (2004, 2005). This method consists of replacing the PV tower (Fig. 6a) associated with the tropical cyclone by the background PV at each vertical level using the spatially averaged method in Ahmadi-Givi et al. (2004).

The negative PV anomaly associated with the anticyclonic outflow is also removed from the initial conditions. This is done by replacing the negative PV anomaly at upper levels (between 350 and 250 hPa) within the PV inversion domain (“O” in Fig. 10) with the value of background PV at each vertical level, following the same procedure as in the removal of the positive PV associated with the cyclonic vortex. Note that the upper-level negative PV anomaly associated with the tropical cyclone has the same “lambda” shape as the outflow cloud from the water vapor satellite image (Fig. 1a). This can be distinguished from the elongated negative PV anomaly associated with the outflow from the ascent across the warm front wf1 to the north (see “O” and “Owf1” in Fig. 10), as indicated by back trajectories (see text above).

The PV inversion technique based on Davis and Emanuel (1991) is then applied in order to obtain the wind, pressure, and temperature fields. The fields satisfy the Charney (1955) nonlinear balance. This implies that the divergent wind is small as compared with the rotational wind. The boundary conditions that are required to perform PV inversion assume that (i) the derivatives of geopotential height and streamfunction are prescribed using the hydrostatic balance by fixing the potential temperature on the horizontal (bottom and top) boundaries, and (ii) the boundary conditions across the lateral boundaries assume that the total horizontal divergence is zero, prescribing the geopotential height and streamfunction. The domain of the PV inversion (Fig. 10) is chosen to be centered on the tropical cyclone without including the developing upstream extratropical cyclone.

The moisture anomaly collocated with the tropical cyclone core is removed by replacing it with a background moisture value outside the tropical cyclone core. The background value is estimated from the average moisture around the tropical cyclone center. As the average moisture decreases radially away from the cyclone center, the background value is taken to be the value of moisture that remained constant with the radial increase. This method of removing the moisture collocated with the tropical cyclone core has also been used in Agustí-Panareda et al. (2004, 2005).

The resulting fields, after removing the cyclonic warm vortex and moisture collocated with it, are used to initialize the numerical model forecast without the tropical cyclone (NOTC). For consistency, the control forecast with the tropical cyclone present in the initial conditions (CNTRL) is also initialized from balanced fields obtained by applying PV inversion to the unmodified PV field.

b. Forecasting model

The numerical model that is used is the Met Office Unified Model, described in Cullen (1993). It is a hydrostatic primitive equation model and it was used operationally until August 2002. The model (version 4.5) is used on a limited area rotated grid with a horizontal resolution of approximately 40 km (Δx = 0.4425°, Δy = 0.4425°) and 38 vertical levels (14 in the boundary layer). The domain of the limited area model covers the North Atlantic, including the eastern coast of North America and western Europe. The boundary conditions for the limited area model are obtained from the Met Office global Unified Model forecast. The model configuration used for the numerical experiments includes the Gregory and Rowntree (1990) convection scheme with a convective available potential energy closure and a mixed-phased precipitation scheme. The model is initialized with analysis data from the ECMWF (1° horizontal resolution).2 Note that the ECMWF analysis data do not include a tropical cyclone initialization scheme.

c. Quasigeostrophic diagnostics

A diagnostic based on the height-attributable QG omega equation is used to analyze the numerical experiments (section 4). The term “height attributable” means we can attribute the dipole patterns of QG vertical velocity in baroclinic cyclones to forcing (i.e., differential vorticity advection and thermal advection) from a particular layer within a range of specific heights in the atmosphere. This concept is discussed extensively in Clough et al. (1996). The diagnostic was developed by Deveson et al. (2002) to objectively classify type-A and type-B extratropical cyclogenesis events (Petterssen and Smebye 1971). A type-A cyclone forms on a front without a preexisting upper-level trough, whereas a type-B cyclone forms when a preexisting upper-level trough moves over a region of warm advection. When ETs result in intensification, they can be classified as type-B cyclogenesis events (Klein et al. 2002). Grotjahn (2005) used a quasigeostrophic model to elucidate some aspects associated with the type-B cyclogenesis, in particular the influence of the lower-level warm anomaly on the upper-level trough forcing.

Although the influence of the transitioning tropical cyclone is not limited to the quasigeostrophic component of the vertical motion, quasigeostropic theory is well suited for the investigation of the mechanisms underlying the synoptic development in the extratropical environment. Indeed, it has been widely used in extratropical transition studies (e.g., Klein et al. 2002; Darr 2002; Atallah and Bosart 2003; Evans and Prater-Mayes 2004). It assumes that the Rossby number (U/fL) is much smaller than 1. Here, the Rossby number is estimated to be between 0.2 and 0.8, because characteristic velocities can range from U = 20 to 40 m s−1; the horizontal spatial scale L ranges between 5 × 105 and 106 m as the radius of maximum winds expands during extratropical transition (Jones et al. 2003); and the Coriolis parameter f = 1 × 10−4 s−1. Thus, the application of quasigeostrophic theory is in the limit of being formally valid. However, even if quasigeostrophic theory is not formally valid, it can still be used to gain insight into cyclogenesis events (Hakim et al. 1996). A description of the Deveson et al. (2002) diagnostic is presented below.

The full quasigeostrophic omega equation can be written as
i1520-0493-136-6-2091-e1
where z is the heightlike pressure coordinate of Hoskins and Bretherton (1972), ω is the quasigeostrophic vertical velocity dz/dt, N is a reference Brunt–Väisälä frequency dependent on height only, f0 is the Coriolis parameter, Q is the Q vector defined by Hoskins et al. (1978), h is the horizontal gradient on a constant-z surface, and ρs is the reference density profile with scale height Hρ, with ρ0 being the reference density at the surface:
i1520-0493-136-6-2091-e2

A height-attributable quasigeostrophic ω can be obtained from Eq. (1) by partitioning the source or forcing of vertical velocity [right-hand side of Eq. (1)] into two layers representing upper (from 650 to 50 hPa) and lower (from 1050 to 750 hPa) levels, as described in Clough et al. (1996). The relative contribution of upper- and lower-level forcing can be found by comparing the quasigeostropic omega in a midtropospheric level (700 hPa) forced by lower and upper levels. This midtropospheric level commonly corresponds to the steering level in baroclinic disturbances (Carroll 1997; Chang et al. 2002), and it is the level used by Hoskins and Pedder (1980) and Deveson et al. (2002). Details on the method used to solve the height-attributable quasigeostropic omega equation can be found in Clough et al. (1996) and Deveson et al. (2002).

The effect of the tropical cyclone on the midlatitude circulation is considered by investigating how the tropical cyclone affects the upper- and lower-level forcing associated with the development of the preexisting midlatitude low upstream.

4. Numerical experiments

The forecast from the Unified Model is first evaluated by comparing the evolution of the track and central mean sea level pressure (MSLP) of Gert and the upstream extratropical cyclone from the forecast initialized with ECMWF analysis with that from Met Office and ECMWF analyses data, as seen in Fig. 11. Initially, at 0000 UTC 23 September, the MSLP for the Met Office analysis is 17 hPa deeper than that of the forecast. There are two reasons for this: First, the forecast is initialized with ECMWF analysis data (1° resolution), which contain a weaker tropical storm than the Met Office analysis data (∼0.5° resolution). Second, the Met Office analysis data have a tropical cyclone initialization scheme (Heming et al. 1995), which uses information from advisory messages issued by the NHC. Thus, the MSLP of the Met Office analysis is only 4 hPa weaker than the NHC MSLP estimate (972 hPa).

Twelve hours later the analysis does not include a synthetic vortex for Gert because the storm is not classified as being tropical by the NHC anymore. At 1200 UTC 23 September, the transitioning Gert is about 5 hPa deeper in the analysis than in the forecast. The upper-level forced low pressure system upstream of Gert has the same MSLP for both the analysis (ECMWF and Met Office) and the forecast initially, because it was forced by a large-scale trough. However, after merging with Gert, there is a difference of around 5 hPa between the analysis and the forecast. This is probably due to the underestimation of Gert’s central MSLP in the low-resolution analysis used to initialize the forecast as well as model error. The tracks followed by Gert and the upstream low are similar for the analyses and forecast, (Fig. 11a). The use of the initial balanced fields in the region of Gert for the control experiment does not have any significant effect on the forecast, because this is nearly identical to the forecast initialized from analysis fields (not shown). The control forecast is slightly deeper (by about 1–6 hPa) than the forecast initialized from the analysis (see Fig. 11b). In summary, the control forecast is accurate for the purposes of the numerical experiments on this case study.

The impact of the tropical cyclone can be initially assessed by comparing the MSLP and tracks from the forecasts with and without the tropical cyclone vortex in the initial conditions (CNTRL and NOTC, respectively, in Fig. 12). The track of the resulting extratropical cyclone after ex-Gert merges with the upstream surface low is approximately 3° south of that when Gert is not present in the initial conditions (Fig. 12a). The surface low pressure system upstream is approximately 10 hPa weaker, and the time of its peak intensity is delayed by 9 h when Gert is present (Fig. 12b). These differences between the two experiments have been confirmed by using other measures of intensity, such as the integrated relative vorticity within 500-km radius of the cyclone’s center (not shown).

In viewing the sets Figs. 13, 14, 15 and Figs 16, 17, 18 further temporal changes can be seen. During its extratropical transition, the PV tower associated with Gert interacts with a broad baroclinic zone advecting colder/drier air from the north toward the upstream low pressure center, as observed in the control forecast and analysis data (Figs. 16a and 3b). The tongue of drier/colder air between Gert’s low pressure center and the low pressure system L upstream (see region around 44°N, 55°W in Figs. 13a and 16a) is not present in the NOTC forecast (Figs. 13b and 16b). This advection by the northerly winds on the western quadrant of ex-Gert is consistent with the southward shift in the track of the upstream low pressure system. As well as advecting dry/cold air toward southwest of the Gert’s PV tower remnants, the transitioning tropical cyclone advects moist/warm air toward its northeast. This produces the warm front wf2 along the broad baroclinic zone.

In the forecast without Gert, the low-level moisture and potential temperature gradient on the broad baroclinic zone (in the region of wf2) becomes meridionally oriented without any positive and negative PV anomaly at lower and upper levels, respectively, along the gradient (Figs. 13b, 14b and 23b). Instead, the warm front wf1 becomes the dominant warm front, with large-scale ascent and diabatically generated positive and negative PV anomalies alongside at lower and upper levels, respectively. At upper levels this is even more evident, because the broad baroclinic zone does not show any signature in the NOTC forecast as compared with the strong gradient in temperature and moisture in the CNTRL forecast in Figs. 19, 20 and 21. All of this indicates that the warm front wf2 is generated by the ex–tropical cyclone circulation by advecting warm and moist air on its eastern flank. The low-level PV structure of the final extratropical cyclone upstream, when it is close to reaching peak intensity, also shows differences between the two experiments resulting from the presence/absence of the second warm front wf2 (see Fig. 15).

A horizontal cross section of the positive PV anomaly associated with the PV tower remnant of ex-Gert can be seen at lower levels in Fig. 13a and upper levels in Fig. 19a. Although the low pressure center of ex-Gert is absorbed by the larger extratropical low pressure system upstream by 0000 UTC 24 September, ex-Gert maintains a separate identity as the remnant of the tropical cyclone PV tower. This PV tower remnant is located in the region of the warm front wf2 (at around 51°N, 43°W in Figs. 14a, 17a and 20a). It is embedded within another elongated PV anomaly associated with latent heat released during the broad ascent across wf2 (Fig. 13a). At 0900 UTC 24 September (T + 33) the PV tower remnant erodes completely (not shown) and ex-Gert loses its identity. Then, the cold-air advection toward the upstream low pressure systems comes to an end.

The negative PV anomaly associated with the initial tropical cyclone outflow (O in Fig. 22a) becomes more filamented as Gert experiences the extratropical transition and its outflow merges with the outflow of the frontal ascent across wf2 (Fig. 23a). The impact of the transitioning tropical cyclone Gert on the extratropical MSLP pattern downstream is minimal (not shown). Note that this does not exclude an impact during its tropical phase (i.e., before 0000 UTC 23 September).

As mentioned in the synoptic overview, the low pressure system upstream is associated with an upper-level positive PV anomaly upstream (see D in Figs. 2b, 3b, 4b and 5b). This upper-level PV anomaly that wraps up cyclonically around the center of the low pressure system upstream is also thinner and further south in the CNTRL forecast than in the forecast without Gert (see D in Figs. 17 and 18).

From the results of the numerical experiments, it is clear that the extratropical transition of Gert has a significant impact on the development of a baroclinic cyclone upstream. This baroclinic development is linked to the coupling between an upper-level PV anomaly upstream and a surface thermal anomaly. To explain how Gert affected the cyclogenesis upstream, we use a method developed by Deveson et al. (2002) (see section 3c). The vertical motion at 700 hPa is attributed to the quasigeostrophic thermodynamic structure at lower levels (from 1050 to 750 hPa, see Figs. 24, 25, 26) and upper levels (from 650 to 50 hPa, see Figs. 27, 28, 29). A dipole of positive (i.e., upward) and negative (i.e., downward) QG vertical velocity (ωQG) can be observed around both ex-Gert and the upstream low pressure system for the CNTRL and NOTC forecasts. This dipole can be associated with positive and negative thermal and differential vorticity advection (Deveson et al. 2002).

In the case of ex-Gert, thermal and vorticity advection associated with the PV tower remnant are due to (i) the advection of the thermal and vorticity anomalies associated with the PV tower by the large-scale environmental flow and (ii) the advection due to the interaction of the PV tower cyclonic vortex with the vorticity and thermal gradients in the environment. Away from the PV tower remnant, the advection of vorticity or PV and temperature will be dominated by the interaction of the vortex with the environmental gradient of PV and temperature. In the case of ex-Gert, the vortex interacts with a large-scale thermal gradient in the midlatitude environment, but not with a large-scale vorticity gradient. Note that the dynamic tropopause was located more than 1500 km away from the center of the PV tower remnant (see Fig. 16a). Thus, away from the PV tower remnant, the thermal advection resulting from the vortex circulation impinging on the large-scale thermal gradient is dominant. However, the differential vorticity advection associated with the transitioning tropical cyclone should not be ignored in the vicinity of the remnant PV tower (DiMego and Bosart 1982). The QG omega equation used here takes into account both forcings. It also allows us to concentrate on the superposition/interaction of the QG forcing coming from the different layers and different features in the upstream development and extratropical transition. The development of a vertical velocity dipole is a characteristic of the transformation stage of the tropical cyclone (Klein et al. 2000). The area of downward vertical velocity associated with ex-Gert is constraining the upward vertical velocity associated with the low pressure system upstream at both upper and lower levels. The magnitude of the minimum negative vertical velocity associated with ex-Gert at lower levels is larger than that at upper levels (−2.0 × 10−2 m s−1 in Fig. 24a, and −1.2 × 10−2 m s−1 in Fig. 27a). The effect on the positive (upward) vertical velocity associated with the low pressure system upstream can be seen more clearly in Fig. 30a. From T + 12 to T + 33 (i.e., 1200 UTC 23 September–0600 UTC 24 September) lower-level forced and upper-level forced positive maximum ωQG are smaller for the CNTRL forecast than the NOTC forecast (see PL and PU in Fig. 30a). This is the time period when Gert is in the warm sector as a PV tower remnant. Figure 25a shows the low-level forcing associated with ex-Gert’s PV tower remnant at 0000 UTC 24 September, even though at that time its low pressure system had already been absorbed by the larger upstream low.

There is a peak in PU at T + 36 (1200 UTC 24 September) in the CNTRL forecast rather than T + 27 (0300 UTC 24 September) as in the NOTC forecast, that is, 9 h later. This is consistent with the evolution of the MSLP, which also shows that the minimum in the MSLP of the resulting extratropical cyclone occurs 9 h later in the CNTRL forecast than in the NOTC forecast (Fig. 12b). This 9-h delay can be explained by the delay in the vertical coupling of the warm anomaly at the surface in the warm sector of the extratropical cyclone and the upper-level positive PV anomaly associated with the upper-level trough. This coupling is characteristic of the baroclinic growth of extratropical cyclones (Hoskins et al. 1985). It is not until ex-Gert has lost its identity and the two systems (i.e., ex-Gert and the upstream low) have merged completely that the resulting extratropical cyclone can reach its maximum intensity. Another consequence of the strong low-level negative forcing over the extratropical low pressure system upstream by the remnants of Gert is the overall reduction of the low-level QG forcing in the cyclogenesis event upstream. As a result, the ratio between upper- and lower-level forcing is higher for the control forecast than for the forecast without the remnants of Gert (see Fig. 30b).

5. Summary and discussion

The extratropical transition of Gert (1999) involved (i) the transformation of the tropical cyclone with the loss of its deep convective core and the development of a frontal structure, and (ii) a merger with an extratropical low pressure system upstream. This is often described in operational forecasting environments as a transitioning tropical cyclone being “absorbed” by a preexisting extratropical low. It is worth noting that this terminology might be appropriate when the extratropical low is a large, mature system because the merging does not involve any significant interaction between the two systems. However, this was not the case for the extratropical transition of Gert. The upstream extratropical low was just beginning to develop as it merged with the transitioning tropical cyclone. Thus, the merging process was not passive, but it involved an interaction between both systems. This interaction resulted in the reduction of the deepening of the resulting low pressure system by 10 hPa and a delay of 9 h to reach maximum intensity, as shown by the forecasts with and without Tropical Cyclone Gert in the initial conditions.

Figure 31 summarizes the main interactions involved during the extratropical transition of Gert using concepts from the quasigeostrophic theory of cyclogenesis and potential vorticity. Once the tropical cyclone enters the extratropical environment, there is an interaction between the cyclonic circulation associated with the tropical cyclone PV tower and the baroclinic zone. This interaction results in positive and negative QG forcing on the right and left sector of the PV tower with respect to its direction of translation, respectively, as well as the creation of a vertical velocity dipole centered on the PV tower (Fig. 31a). This dipole is present at both the lower and upper levels and is crucial in understanding the interaction between the transitioning tropical cyclone and a developing extratropical cyclone upstream (Fig. 31b). In the case of the extratropical transition of Gert, the transitioning cyclone moved toward the warm front of the developing low pressure system upstream. Thus, the region of negative vertical velocity overlapped the region of favorable development (and positive vertical velocity) associated with the extratropical low pressure system upstream (Fig. 31c). The result is a weakening in the development of the upstream low pressure system. The extratropical cyclone cannot achieve its peak intensity until the negative impact disappears after the two systems have merged. Therefore, the role of the transitioning tropical cyclone interacting with the warm front of an extratropical cyclone upstream is to weaken and delay its development.

An alternative scenario would be if a transitioning tropical cyclone interacts with a trailing cold front. Then, there is an overlap between the region of positive vertical velocity in the dipole associated with warm advection and the region of development associated with the extratropical cyclone upstream (Fig. 31d). The result thereof is in an enhancement in the growth of the resulting extratropical cyclone. There are many case studies in the literature that have shown this enhancement (e.g., Klein et al. 2002; McTaggart-Cowan et al. 2001; Agustí-Panareda et al. 2004). However, this is the first case study of an extratropical transition where the tropical cyclone vortex is shown to have a negative impact on the development of an extratropical cyclone.

According to the classification of Matano and Sekioka (1971), the interaction of a transitioning cyclone with a front from a preexisting extratropical cyclone is classified as a compound system, whereas the interaction of a transitioning cyclone with a baroclinic zone without a preexisting extratropical cyclone is classified as a complex system. Figure 31 emphasizes the importance of differentiating compound systems that involve cold fronts from those that involve warm fronts.

The results from this case study also have implications for the predictability of extratropical cyclones, in agreement with the results from the idealized modeling experiments of Beare et al. (2003). They studied the effect of localized vortices associated with potential vorticity perturbations on the development of extratropical cyclones using a quasigeostrophic model. Their results show that the most sensitive regions are in the warm and cold fronts at the steering level. They found that there is an optimum deepening of the cyclone (of 5 hPa for a vortex with 10 m s−1 maximum winds) when the vortex is over the cold front and an optimum filling when it is over the warm front. In this case study, the vortex is transitioning Tropical Cyclone Gert (1999) and the filling of the upstream cyclone is 10 hPa. Beare et al. (2003) differentiate between the dynamic contribution of the perturbation (here a transitioning tropical cyclone) and the instantaneous pressure contribution (i.e., the merging of the vortex or transitioning cyclone when this moves over the extratropical cyclone). The dynamic contribution refers to the thermal advection by the perturbation vortex on the developing extratropical cyclone. The instantaneous pressure contribution is zero here because the transitioning tropical cyclone does not move over the low pressure center of the cyclone upstream. Instead, the transitioning tropical cyclone remains to the east of the low pressure center upstream in the area of the warm front.

In summary, the impact of the transitioning tropical cyclone on the development of an extratropical cyclone does not only depend on its phasing with a preexisting upper-level trough (or positive PV anomaly) upstream (e.g., Klein et al. 2002) and the characteristics of the trough (e.g., Klein et al. 2002; Agustí-Panareda et al. 2005); it also depends on whether the transitioning cyclone interacts with the cold or the warm front of the developing extratropical cyclone, before the two systems merge. Therefore, it is crucial to get an accurate forecast of the relative positions of the transitioning cyclone and the developing extratropical cyclone in order to capture either the positive or negative impact on the surface pressure of the resulting extratropical cyclone.

Acknowledgments

Thanks are given to the Met Office for providing the analysis data and the Unified Model. The Universities Weather Research Network support is also acknowledged. In particular, Dr. Peter Panagi, Dr. Chang Wang, and Dr. Lois Steenman-Clark are acknowledged for providing the software to calculate and plot some of the diagnostics, as well as providing support on running the Unified Model. This work has benefited from numerous discussions with Dr. Suzanne Gray and Sally Furness. The comments by Dr. Suzanne Gray on the first draft of the manuscript are greatly appreciated. The help and encouragement from Prof. Sarah Jones in the revision stage of the paper is also greatly acknowledged. Also, thanks are given to the anonymous reviewers for their positive and constructive comments, which helped to improve the manuscript.

REFERENCES

  • Agustí-Panareda, A., , C. D. Thorncroft, , G. C. Craig, , and S. L. Gray, 2004: The extratropical transition of Hurricane Irene (1999): A potential vorticity perspective. Quart. J. Roy. Meteor. Soc., 130 , 10471074.

    • Search Google Scholar
    • Export Citation
  • Agustí-Panareda, A., , S. L. Gray, , G. C. Craig, , and C. D. Thorncroft, 2005: The extratropical transition of tropical cyclone Lili (1996) and its crucial contribution to a moderate extratropical development. Mon. Wea. Rev., 133 , 15621573.

    • Search Google Scholar
    • Export Citation
  • Ahmadi-Givi, F., , G. C. Craig, , and R. S. Plant, 2004: The dynamics of a midlatitude cyclone with very strong latent-heat release. Quart. J. Roy. Meteor. Soc., 130 , 295323.

    • Search Google Scholar
    • Export Citation
  • Atallah, E. H., , and L. F. Bosart, 2003: The extratropical transition and precipitation distribution of Hurricane Floyd (1999). Mon. Wea. Rev., 131 , 10631081.

    • Search Google Scholar
    • Export Citation
  • Beare, R. J., , A. J. Thorpe, , and A. A. White, 2003: The predictability of extratropical cyclones: Nonlinear sensitivity to localized potential-vorticity perturbations. Quart. J. Roy. Meteor. Soc., 129 , 219237.

    • Search Google Scholar
    • Export Citation
  • Carroll, E. B., 1997: Use of dynamical concepts in weather forecasting. Meteor. Appl., 4 , 345352.

  • Chang, E. K. M., , S. Lee, , and K. Swanson, 2002: Storm track dynamics. J. Climate, 15 , 21632183.

  • Charney, J. G., 1955: The use of primitive equations of motion in numerical prediction. Tellus, 7 , 2226.

  • Clough, S. A., , C. S. A. Davitt, , and A. J. Thorpe, 1996: Attribution concepts applied to the omega equation. Quart. J. Roy. Meteor. Soc., 122 , 19431962.

    • Search Google Scholar
    • Export Citation
  • Cullen, M. J. P., 1993: The unified forecast/climate model. Meteor. Mag., 122 , 8194.

  • Darr, J. K., 2002: Quantitative measurements of extratropical transition in the Atlantic basin. Preprints, 19th Conf. on Weather Analysis and Forecasting, San Antonio, TX, Amer. Meteor. Soc., 186–189.

  • Davis, C. A., , and K. A. Emanuel, 1991: Potential vorticity diagnostics of cyclogenesis. Mon. Wea. Rev., 119 , 19291953.

  • Deveson, A. C. L., , K. A. Browning, , and T. D. Hewson, 2002: A classification of FASTEX cyclones using a height-attributable quasi-geostrophic vertical-motion diagnostic. Quart. J. Roy. Meteor. Soc., 128 , 93116.

    • Search Google Scholar
    • Export Citation
  • DiMego, G. J., , and L. F. Bosart, 1982: The transformation of tropical storm Agnes into an extratropical cyclone. Part I: The observed fields and vertical motion computations. Mon. Wea. Rev., 110 , 385411.

    • Search Google Scholar
    • Export Citation
  • Evans, J. L., , and B. E. Prater-Mayes, 2004: Factors affecting the posttransition intensification of Hurricane Irene (1999). Mon. Wea. Rev., 132 , 13551368.

    • Search Google Scholar
    • Export Citation
  • Gregory, D., , and P. R. Rowntree, 1990: A mass flux convection scheme with representation of cloud ensemble characteristics and stability-dependent closure. Mon. Wea. Rev., 118 , 14831506.

    • Search Google Scholar
    • Export Citation
  • Grotjahn, R., 2005: On type B cyclogenesis in a quasi-geostrophic model. Quart. J. Roy. Meteor. Soc., 131 , 109124.

  • Hakim, G. J., , D. Keyser, , and L. F. Bosart, 1996: The Ohio Valley wave-merger cyclogenesis event of 25–26 January 1978. Part II: Diagnosis using quasigeostrophic potential vorticity inversion. Mon. Wea. Rev., 124 , 21762205.

    • Search Google Scholar
    • Export Citation
  • Harr, P. A., , and R. L. Elsberry, 2000: Extratropical transition of tropical cyclones over the western North Pacific. Part I: Evolution of structural characteristics during the transition process. Mon. Wea. Rev., 128 , 26132633.

    • Search Google Scholar
    • Export Citation
  • Heming, J. T., , J. C. L. Chan, , and A. M. Radford, 1995: A new scheme for the initialisation of tropical cyclones in the UK Meteorological Office global model. Meteor. Appl., 2 , 171184.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., , and F. P. Bretherton, 1972: Atmospheric frontogenesis models: Mathematical formulation and solution. J. Atmos. Sci., 29 , 1137.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., , and M. A. Pedder, 1980: The diagnosis of middle-latitude synoptic development. Quart. J. Roy. Meteor. Soc., 106 , 707719.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., , I. Draghici, , and H. C. Davis, 1978: A new look at the ω equation. Quart. J. Roy. Meteor. Soc., 104 , 3138.

  • 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.

    • Search Google Scholar
    • Export Citation
  • Jones, S. C., and Coauthors, 2003: The extratropical transition of tropical cyclones: Forecast challanges, current understanding, and future directions. Wea. Forecasting, 18 , 10521092.

    • Search Google Scholar
    • Export Citation
  • Klein, P. M., , P. A. Harr, , and R. L. Elsberry, 2000: Extratropical transition of western North Pacific tropical cyclones: An overview and conceptual model of the transformation stage. Wea. Forecasting, 15 , 373395.

    • Search Google Scholar
    • Export Citation
  • Klein, P. M., , P. A. Harr, , and R. L. Elsberry, 2002: Extratropical transition of western North Pacific tropical cyclones: Midlatitude and tropical cyclone contributions to reintensification. Mon. Wea. Rev., 130 , 22402259.

    • Search Google Scholar
    • Export Citation
  • Lawrence, M. B., , L. A. Avila, , J. L. Beven, , J. L. Franklin, , J. L. Guiney, , and R. J. Pasch, 2001: Atlantic hurricane season of 1999. Mon. Wea. Rev., 129 , 30573084.

    • Search Google Scholar
    • Export Citation
  • Matano, H., , and M. Sekioka, 1971: Some aspects of extratropical transformation of a tropical cyclone. J. Meteor. Soc. Japan, 49 , 736743.

    • Search Google Scholar
    • Export Citation
  • McTaggart-Cowan, R., , J. R. Gyakum, , and M. K. Yau, 2001: Sensitivity testing of extratropical transitions using potential vorticity inversions to modify initial conditions: Hurricane Earl case study. Mon. Wea. Rev., 129 , 16171636.

    • Search Google Scholar
    • Export Citation
  • Petterssen, S., , and S. J. Smebye, 1971: On the development of extratropical cyclones. Quart. J. Roy. Meteor. Soc., 97 , 457482.

  • Simpson, R. H., 1974: The hurricane disaster potential scale. Weatherwise, 27 , 169186.

Fig. 1.
Fig. 1.

Geostationary Operational Environmental Satellite (GOES)-8 (GOES-East) water vapor image. (Courtesy of the National Climatic Data Center.) Here, L denotes the center of the upstream extratropical low pressure system, wf1 is the warm front associated with the upstream baroclinic wave, T denotes the location of a large-scale upper-level trough, and D is the stratospheric dry intrusion associated with an upper-level potential vorticity anomaly. A belt of low pressure extending from L toward the north is marked with a trough line (dashed); wf2 is the warm front that developed as Gert approached the colder extratropical environment, and O is the outflow of Tropical Cyclone Gert.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 2.
Fig. 2.

ECMWF analysis fields at 0000 UTC 23 Sep 1999: (a) humidity mixing ratio (g kg−1) at 850 hPa [gray shading: below 7.5 in white, 7.5–10.0 in light gray, 10.0–15.0 in gray, and larger than 15.0 in dark gray; also shown are PV contours equal to 2 PVU at 850 hPa (thick, solid line) and geopotential height contours at 1000 hPa (thin solid lines, 40-m interval)] and (b) isentropic PV on a 330-K surface [gray shading: black is PV from −1 to 0.2 PVU, dark gray is 2–4 PVU, and light gray is 4–14 PVU; also shown are potential temperature contours at 950 hPa (solid lines, 2-K interval)]. Symbols are as in Fig. 1.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 3.
Fig. 3.

As in Fig. 2, but at 1200 UTC 23 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 4.
Fig. 4.

As in Fig. 2, but at 0000 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 5.
Fig. 5.

As in Fig. 2, but at 1200 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 6.
Fig. 6.

North–south cross sections of ECMWF analysis fields across the center of the tropical cyclone PV tower during the extratropical transition of Gert at (a) 0000 and (b) 1200 UTC 23 Sep 1999. Gray shading represents PV with values from −1 to 0 (black), 2–4 (dark gray), and 4–14 PVU (light gray). Thick solid lines are potential vorticity contours (1-PVU interval), thin solid lines are equivalent potential temperature contours (5-K interval), and dashed lines are potential temperature contours (5-K interval).

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 7.
Fig. 7.

North–south cross sections of ECMWF analysis fields across the center of the tropical cyclone PV tower during the extratropical transition of Gert at (a) 0000 and (b) 1200 UTC 24 Sep 1999. Same contours intervals as in Fig. 6.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 8.
Fig. 8.

ECMWF analysis fields: humidity mixing ratio (g kg−1) at 500 hPa (gray shading: white is below 1.0, light gray is 1.0–2.5, gray is 2.5–5.0, and dark gray is larger than 5.0). Also shown are PV contours equal to 2 PVU at 500 hPa (thick, solid line) and potential temperature contours at 500 hPa (thin, solid line; 2-K interval). Symbols are as in Fig. 1.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 9.
Fig. 9.

As in Fig. 8, but for 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 10.
Fig. 10.

Horizontal section of PV at 250 hPa with gray shading (black: from −1 to 0.3 PVU, dark gray: 2–4 PVU, and light gray: 4–14 PVU) covering the limited-area model (LAM) domain. The PV inversion domain is shown with the black box centered on Tropical Cyclone Gert (“G” depicts the location of its PV tower). The negative PV anomaly associated with the upper-level outflow of Gert is marked as O and that associated with the outflow of the frontal ascent across the warm front wf1 is marked as Owf1.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 11.
Fig. 11.

(a) Tracks followed by Gert and the upper-level forced low upstream. Overlaid contours are mean SST for September 1999 from the Reynolds dataset (courtesy of Climate Diagnostics Center). (b) Minimum MSLP at the center of Gert during its extratropical transition and the upper-level forced low pressure system upstream. In both (a) and (b), ECMWF analysis, Met Office analysis, and the forecast initialized from ECMWF analysis fields have been plotted every 12 h. The control forecast initialized from the balanced fields (see section 3a) has been plotted every 12 h in (a) and every hour in (b). Both forecasts were started at 0000 UTC 23 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 12.
Fig. 12.

(a) Tracks of the center of the transitioning Gert and the upstream low for the control forecast and the forecast without Gert in the initial conditions every 1 h, and (b) time series of hourly minimum MSLP at the center of transitioning Gert and the upstream low for the CNTRL and NOTC.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 13.
Fig. 13.

Horizontal sections of humidity mixing ratio (g kg−1) at 850 hPa [gray shading: white is below 7.5, light gray is 7.5–10.0, gray is 10.0–15.0, and dark gray is larger than 15.0; also shown are PV contours equal to 2 PVU at 850 hPa (thick, solid line), and geopotential height contours at 1000 hPa (thin, solid line; 40-m interval)] from (a) CNTRL and (b) NOTC. Symbols are as in Fig. 1.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 14.
Fig. 14.

As in Fig. 13, but for 0000 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 15.
Fig. 15.

As in Fig. 13, but for 1200 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 16.
Fig. 16.

Horizontal cross sections of isentropic PV on a 330-K surface with values from −1 to 0.2 PVU (black), from 2 to 4 PVU (dark gray), and from 4 to 14 PVU (light gray) [also shown are potential temperature contours (thin, solid line; 2-K interval) at 950 hPa] for (a) CNTRL and (b) NOTC. Symbols are as in Fig. 1.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 17.
Fig. 17.

As in Fig. 16, but for 0000 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 18.
Fig. 18.

As in Fig. 16, but for 1200 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 19.
Fig. 19.

Horizontal sections of humidity mixing ratio (g kg−1) at 500 hPa [gray shading: white is below 1.0, light gray is 1.0–2.5, gray is 2.5–5.0, and dark gray is larger than 5.0; also shown are PV contours equal to 2 PVU at 500 hPa (thick, solid line) and potential temperature contours at 500 hPa (thin, solid line; 2-K interval)] from (a) CNTRL and (b) NOTC. Symbols are as in Fig. 1.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 20.
Fig. 20.

As in Fig. 19, but for 0000 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 21.
Fig. 21.

As in Fig. 19, but for 1200 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 22.
Fig. 22.

Horizontal cross sections of isentropic PV on a 345-K surface with values from −1 to 0.3 (black), from 2 to 4 (dark gray), and from 4 to 14 (light gray) PVU and potential temperature contours (thin, solid line; 2-K interval) at 950 hPa from (a) CNTRL and (b) NOTC. Symbols are as in Fig. 1.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 23.
Fig. 23.

As in Fig. 22, but for 0000 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 24.
Fig. 24.

Quasigeostrophic vertical velocity (ωQG) at 700 hPa forced from lower levels (1050–750 hPa) with negative values as thick dashed contours and positive values as thick solid contours (0.008 m s−1 interval) and geopotential height at 1000 hPa (thin contours, 40-m interval) from (a) CNTRL and (b) NOTC. Note that ωQG is positive for upward motion and negative for downward motion.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 25.
Fig. 25.

As in Fig. 24, but for 0000 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 26.
Fig. 26.

As in Fig. 24, but for 1200 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 27.
Fig. 27.

Quasigeostrophic vertical velocity (ωQG) at 700 hPa forced from upper levels (650–50 hPa) with negative values as thick dashed contours and positive values as thick solid contours (0.008 m s−1 interval) and geopotential height at 300 hPa (thin contours, 80-m interval) from (a) CNTRL and (b) NOTC. Note that ωQG is positive for upward motion and negative for downward motion.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 28.
Fig. 28.

As in Fig. 27, but for 0000 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 29.
Fig. 29.

As in Fig. 27, but for 1200 UTC 24 Sep 1999.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 30.
Fig. 30.

(a) Maximum positive quasigeostrophic vertical velocity (ωQG, m s−1) at 700 hPa forced from lower (PL) and upper (PU) levels for the CNTRL and NOTC forecasts, and (b) ratio of the (ωQG) dipole strength forced from lower and upper levels (U/L ratio). The dipole strength at low levels (DL) is calculated by summing the magnitudes of the maximum ascent [see (a)] and maximum descent (not shown) at 700 hPa forced from lower levels. The strength of the upper-level forced dipole (DU) at 700 hPa is calculated in a similar fashion. The U/L ratio is DU/DL. Note that positive vertical velocity is upward and negative vertical velocity is downward. Low-level forcing is from 1000 to 750 hPa, and upper-level forcing is from 650 to 100 hPa. See section 3c for more details.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

Fig. 31.
Fig. 31.

This schematic shows the interactions involved during an extratropical transition classified as a “compound system” (Matano and Sekioka 1971) based on concepts from the QG theory. (a) Interaction between a vortex associated with the remnants of a TC PV tower (depicted by arrows with a hurricane symbol at the center) and a zonal baroclinic zone (dotted line) during the transformation stage of the TC. The result is cold/warm advection to the west/east of the vortex and an associated QG vertical velocity (ωQG) dipole (solid/dashed closed contours with positive/negative signs denoting ascent/descent, respectively). The dipole is forced from both upper and lower levels. (b) Upper- and lower-level forcing in the development of an extratropical cyclone associated with an upper-level positive PV anomaly collocated with an upper-level trough T (see dark gray shading) and surface thermal anomalies depicted by the cold/warm fronts. An ascending/descending vertical motion dipole in gray shading and hatch shading, respectively, is centered on the trough axis T. The ascending region is the area of favorable development for the surface low pressure center L. At low levels the ωQG dipole is centered around L (thick solid/dashed closed contours with positive/negative signs denoting ascent/descent, respectively). All of the ωQG dipoles are shown at the steering level of the extratropical cyclone. The transformed TC can interact with (c) the warm front or (d) the cold front of a preexisting baroclinic cyclone. See text for a more detailed explanation.

Citation: Monthly Weather Review 136, 6; 10.1175/2007MWR1637.1

1

Note that the impact of Gert during its tropical phase is not considered in this paper.

2

The Met Office data required to initialize the model were not available because they were only archived for a limited period of time.

Save